From b139ee7a94922285addbde8aeeb2ebc15f2c7748 Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Tue, 6 Dec 2022 17:07:09 +0100
Subject: [PATCH 01/15] First draft

---
 .../forecasting/neural_prophet_model.py       | 71 +++++++++++++++++++
 1 file changed, 71 insertions(+)
 create mode 100644 darts/models/forecasting/neural_prophet_model.py

diff --git a/darts/models/forecasting/neural_prophet_model.py b/darts/models/forecasting/neural_prophet_model.py
new file mode 100644
index 0000000000..bb1e2bf2b1
--- /dev/null
+++ b/darts/models/forecasting/neural_prophet_model.py
@@ -0,0 +1,71 @@
+"""
+Neural Prophet
+------------
+"""
+
+from typing import Optional, Sequence, Tuple, Union
+
+import neuralprophet
+import pandas as pd
+
+from darts.logging import raise_if_not
+from darts.models.forecasting.forecasting_model import GlobalForecastingModel
+from darts.timeseries import TimeSeries
+
+
+class NeuralProphet(GlobalForecastingModel):
+    def __init__(self, n_lags: int = 0, n_forecasts: int = 1, *kwargs):
+        super().__init__()
+
+        raise_if_not(n_lags >= 0, "Argument n_lags should be a non-negative integer")
+
+        self.model = neuralprophet.NeuralProphet(
+            n_lags=n_lags, n_forecasts=n_forecasts, *kwargs
+        )
+
+    def fit(
+        self,
+        series: Union[TimeSeries, Sequence[TimeSeries]],
+        past_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
+        future_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
+    ) -> "GlobalForecastingModel":
+        super().fit(series, past_covariates, future_covariates)
+
+        # TODO change to accept multivariate
+        fit_df = pd.DataFrame(
+            data={"ds": series.time_index, "y": series.univariate_values()}
+        )
+        self.model.fit(fit_df, freq=series.freq_str)
+        return self
+
+    def predict(
+        self,
+        n: int,
+        series: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
+        past_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
+        future_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
+        num_samples: int = 1,
+        verbose: bool = False,
+    ) -> Union[TimeSeries, Sequence[TimeSeries]]:
+        super().predict(
+            n, series, past_covariates, future_covariates, num_samples, verbose
+        )
+        print(series.columns)
+        predict_df = pd.DataFrame(
+            data={"ds": series.time_index, "y": series.univariate_values()}
+        )
+        predict_df = self.model.make_future_dataframe(df=predict_df, periods=n)
+        future_df = self.model.predict(predict_df)
+
+        future_df = future_df[["ds", "yhat1"]].rename(
+            columns={"yhat1": series.columns[0]}
+        )
+        future_ts = TimeSeries.from_dataframe(df=future_df[-n:], time_col="ds")
+
+        return future_ts
+
+    def _model_encoder_settings(self) -> Tuple[int, int, bool, bool]:
+        raise NotImplementedError()
+
+    def __str__(self):
+        return "Neural Prophet"

From 2422e47a9f6126359044799c581498769b351264 Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Thu, 8 Dec 2022 15:47:43 +0100
Subject: [PATCH 02/15] Allow multivariate time series

---
 .../forecasting/neural_prophet_model.py       | 65 ++++++++++++++-----
 1 file changed, 48 insertions(+), 17 deletions(-)

diff --git a/darts/models/forecasting/neural_prophet_model.py b/darts/models/forecasting/neural_prophet_model.py
index bb1e2bf2b1..5509e1d019 100644
--- a/darts/models/forecasting/neural_prophet_model.py
+++ b/darts/models/forecasting/neural_prophet_model.py
@@ -7,10 +7,11 @@
 
 import neuralprophet
 import pandas as pd
+from neuralprophet.utils import fcst_df_to_latest_forecast
 
 from darts.logging import raise_if_not
 from darts.models.forecasting.forecasting_model import GlobalForecastingModel
-from darts.timeseries import TimeSeries
+from darts.timeseries import TimeSeries, concatenate
 
 
 class NeuralProphet(GlobalForecastingModel):
@@ -28,14 +29,13 @@ def fit(
         series: Union[TimeSeries, Sequence[TimeSeries]],
         past_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
         future_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
-    ) -> "GlobalForecastingModel":
+    ) -> "NeuralProphet":
         super().fit(series, past_covariates, future_covariates)
+        self.training_series = series
 
-        # TODO change to accept multivariate
-        fit_df = pd.DataFrame(
-            data={"ds": series.time_index, "y": series.univariate_values()}
-        )
+        fit_df = self._convert_ts_to_df(series)
         self.model.fit(fit_df, freq=series.freq_str)
+
         return self
 
     def predict(
@@ -50,19 +50,50 @@ def predict(
         super().predict(
             n, series, past_covariates, future_covariates, num_samples, verbose
         )
-        print(series.columns)
-        predict_df = pd.DataFrame(
-            data={"ds": series.time_index, "y": series.univariate_values()}
-        )
-        predict_df = self.model.make_future_dataframe(df=predict_df, periods=n)
-        future_df = self.model.predict(predict_df)
 
-        future_df = future_df[["ds", "yhat1"]].rename(
-            columns={"yhat1": series.columns[0]}
-        )
-        future_ts = TimeSeries.from_dataframe(df=future_df[-n:], time_col="ds")
+        if series is None:
+            series = self.training_series
+        df = self._convert_ts_to_df(series)
+
+        future_df = self.model.make_future_dataframe(df=df, periods=n)
+        forecast_df = self.model.predict(future_df)
+
+        return self._convert_df_to_ts(forecast_df, series.end_time(), series.components)
 
-        return future_ts
+    def _convert_ts_to_df(self, series: TimeSeries):
+        dfs = []
+
+        for component in series.components:
+            new_df = (
+                series[component].pd_dataframe(copy=False).reset_index(names=["ds"])
+            )
+            component_df = (
+                new_df[["ds", component]]
+                .copy(deep=True)
+                .rename(columns={component: "y"})
+            )
+            component_df["ID"] = component
+            dfs.append(component_df)
+
+        return pd.concat(dfs)
+
+    def _convert_df_to_ts(self, forecast, last_train_date, components):
+        groups = []
+        for component in components:
+            simple_df = fcst_df_to_latest_forecast(
+                forecast[forecast["ID"] == component].copy(deep=True),
+                quantiles=[0.5],
+                n_last=1,
+            )
+            simple_df = simple_df[["ds", "origin-0"]].rename(
+                columns={"origin-0": component}
+            )
+            groups.append(simple_df[simple_df["ds"] > last_train_date])
+
+        return concatenate(
+            [TimeSeries.from_dataframe(group, time_col="ds") for group in groups],
+            axis=1,
+        )
 
     def _model_encoder_settings(self) -> Tuple[int, int, bool, bool]:
         raise NotImplementedError()

From 906de4cd5b1443f48b883606ef855a0468ca7241 Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Mon, 12 Dec 2022 14:55:22 +0100
Subject: [PATCH 03/15] Add examples and improve conversion

---
 .../forecasting/neural_prophet_model.py       | 113 ++++--
 neural_examples/examples.ipynb                | 337 ++++++++++++++++++
 2 files changed, 419 insertions(+), 31 deletions(-)
 create mode 100644 neural_examples/examples.ipynb

diff --git a/darts/models/forecasting/neural_prophet_model.py b/darts/models/forecasting/neural_prophet_model.py
index 5509e1d019..8d7e1ff883 100644
--- a/darts/models/forecasting/neural_prophet_model.py
+++ b/darts/models/forecasting/neural_prophet_model.py
@@ -20,6 +20,8 @@ def __init__(self, n_lags: int = 0, n_forecasts: int = 1, *kwargs):
 
         raise_if_not(n_lags >= 0, "Argument n_lags should be a non-negative integer")
 
+        self.n_lags = n_lags
+        self.n_forecasts = n_forecasts
         self.model = neuralprophet.NeuralProphet(
             n_lags=n_lags, n_forecasts=n_forecasts, *kwargs
         )
@@ -31,9 +33,10 @@ def fit(
         future_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
     ) -> "NeuralProphet":
         super().fit(series, past_covariates, future_covariates)
-        self.training_series = series
 
-        fit_df = self._convert_ts_to_df(series)
+        self.training_series = self._as_sequence(series)
+        fit_df = self._convert_ts_to_df(self.training_series)
+
         self.model.fit(fit_df, freq=series.freq_str)
 
         return self
@@ -52,49 +55,97 @@ def predict(
         )
 
         if series is None:
-            series = self.training_series
-        df = self._convert_ts_to_df(series)
-
-        future_df = self.model.make_future_dataframe(df=df, periods=n)
-        forecast_df = self.model.predict(future_df)
-
-        return self._convert_df_to_ts(forecast_df, series.end_time(), series.components)
+            series_list = self.training_series
+        else:
+            series_list = self._as_sequence(series)
+
+        raise_if_not(
+            self.n_lags == 0 or n <= self.n_forecasts,
+            "Auto-regression has been configured. `n` must be smaller than or equal to"
+            "`n_forecasts` parameter in the constructor.",
+        )
+        # TODO consider time series indexed by ints
+        # TODO check if series was used during training
+
+        predictions = []
+        for series in series_list:
+            df = self._convert_ts_to_df([series])
+            future_df = self.model.make_future_dataframe(df=df, periods=n)
+            forecast_df = self.model.predict(future_df)
+
+            predictions.append(
+                self._convert_df_to_ts(
+                    forecast_df, series.end_time(), series.components
+                )
+            )
+        return self._from_sequence(predictions)
 
-    def _convert_ts_to_df(self, series: TimeSeries):
+    def _convert_ts_to_df(self, series_list: Sequence[TimeSeries]):
         dfs = []
 
-        for component in series.components:
-            new_df = (
-                series[component].pd_dataframe(copy=False).reset_index(names=["ds"])
-            )
-            component_df = (
-                new_df[["ds", component]]
-                .copy(deep=True)
-                .rename(columns={component: "y"})
-            )
-            component_df["ID"] = component
-            dfs.append(component_df)
+        for series in series_list:
+            for component in series.components:
+                new_df = (
+                    series[component].pd_dataframe(copy=False).reset_index(names=["ds"])
+                )
+                component_df = (
+                    new_df[["ds", component]]
+                    .copy(deep=True)
+                    .rename(columns={component: "y"})
+                )
+                component_df["ID"] = component
+                dfs.append(component_df)
 
         return pd.concat(dfs)
 
-    def _convert_df_to_ts(self, forecast, last_train_date, components):
+    def _convert_df_to_ts(self, forecast: pd.DataFrame, last_train_date, components):
         groups = []
         for component in components:
-            simple_df = fcst_df_to_latest_forecast(
-                forecast[forecast["ID"] == component].copy(deep=True),
-                quantiles=[0.5],
-                n_last=1,
-            )
-            simple_df = simple_df[["ds", "origin-0"]].rename(
-                columns={"origin-0": component}
-            )
-            groups.append(simple_df[simple_df["ds"] > last_train_date])
+            if self.n_lags == 0:
+                # output format is different when AR is not used
+                groups.append(
+                    forecast[
+                        (forecast["ID"] == component)
+                        & (forecast["ds"] > last_train_date)
+                    ]
+                    .filter(items=["ds", "yhat1"])
+                    .rename(columns={"yhat1": component})
+                )
+            else:
+                df = fcst_df_to_latest_forecast(
+                    forecast[(forecast["ID"] == component)],
+                    quantiles=[0.5],
+                    n_last=1,
+                )
+                groups.append(
+                    df[df["ds"] > last_train_date]
+                    .filter(items=["ds", "origin-0"])
+                    .rename(columns={"origin-0": component})
+                )
 
         return concatenate(
             [TimeSeries.from_dataframe(group, time_col="ds") for group in groups],
             axis=1,
         )
 
+    def _as_sequence(
+        self, series: Union[TimeSeries, Sequence[TimeSeries]]
+    ) -> Sequence[TimeSeries]:
+        if isinstance(series, TimeSeries):
+            return [series]
+
+        if isinstance(series, Sequence[TimeSeries]):
+            return series
+
+        raise ValueError("Invalid type. Expected TimeSeries or Sequence[TimeSeries]")
+
+    def _from_sequence(
+        self, series_list: Sequence[TimeSeries]
+    ) -> Union[TimeSeries, Sequence[TimeSeries]]:
+        if len(series_list) == 1:
+            return series_list[0]
+        return series_list
+
     def _model_encoder_settings(self) -> Tuple[int, int, bool, bool]:
         raise NotImplementedError()
 
diff --git a/neural_examples/examples.ipynb b/neural_examples/examples.ipynb
new file mode 100644
index 0000000000..da92f4e30a
--- /dev/null
+++ b/neural_examples/examples.ipynb
@@ -0,0 +1,337 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import torch\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "from darts import TimeSeries\n",
+    "from darts.utils.timeseries_generation import (\n",
+    "    sine_timeseries,\n",
+    ")\n",
+    "from darts.metrics import mape, smape\n",
+    "from darts.dataprocessing.transformers import Scaler\n",
+    "from darts.utils.timeseries_generation import datetime_attribute_timeseries\n",
+    "from darts.datasets import AirPassengersDataset, MonthlyMilkDataset\n",
+    "from darts.models.forecasting.neural_prophet_model import (\n",
+    "    NeuralProphet as NeuralProphetDarts,\n",
+    ")\n",
+    "from neuralprophet import NeuralProphet\n",
+    "\n",
+    "\n",
+    "# for reproducibility\n",
+    "torch.manual_seed(1)\n",
+    "np.random.seed(1)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Univariate example"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "series_air = AirPassengersDataset().load()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<matplotlib.legend.Legend at 0x7fa41b10c7c0>"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "series_air.plot(label=\"air\")\n",
+    "plt.legend()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train_air, test_air = series_air[:-36], series_air[-36:]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Only trend and seasonality - equivalent to using Prophet"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 514\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4fcfd515771f44ec813f43821b22511d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/107 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 3.45E-01, min: 4.02E-02\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "df98919dae0d4576858e0762547063b3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/107 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 2.33E-01, min: 2.84E-01\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.00E-01\n",
+      "Epoch[514/514]: 100%|██████████| 514/514 [00:03<00:00, 140.22it/s, SmoothL1Loss=0.00124, MAE=10.9, RMSE=14.2, Loss=0.000911, RegLoss=0]\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 88.889% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 88.889% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n"
+     ]
+    }
+   ],
+   "source": [
+    "model = NeuralProphetDarts()\n",
+    "model.fit(train_air)\n",
+    "preds_simple = model.predict(36)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "With auto-regression"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 1000\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "255fc4100cbd481a84d083205c313f0d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/103 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-02, min: 2.41E-02\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c234eb4ee5d842eebbc626d1c78ea033",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/103 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-02, min: 2.41E-02\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.25E-02\n",
+      "Epoch[1000/1000]: 100%|██████████| 1000/1000 [00:04<00:00, 224.43it/s, SmoothL1Loss=0.000167, MAE=4.09, RMSE=5.28, Loss=0.000138, RegLoss=0]\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.278% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.278% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "model = NeuralProphetDarts(n_lags=36, n_forecasts=36)\n",
+    "model.fit(train_air)\n",
+    "preds_ar = model.predict(36)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# train_air.plot(label=\"train\")\n",
+    "test_air.plot(label=\"test\")\n",
+    "preds_simple.plot(label=\"trend & season\")\n",
+    "preds_ar.plot(label=\"auto-regression\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3.9.15 ('prophet')",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.9.15"
+  },
+  "orig_nbformat": 4,
+  "vscode": {
+   "interpreter": {
+    "hash": "2f14e79e1646dc5b749c3dc6e0dfef5e568c2efea6b930caf0398818dd8806ea"
+   }
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

From f9937d0233a4ae38d7b9ec9d57561132c48d989e Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Tue, 13 Dec 2022 15:44:18 +0100
Subject: [PATCH 04/15] Attempt at global model with past covariates

---
 .../forecasting/neural_prophet_model.py       | 62 ++++++++++++++----
 neural_examples/examples.ipynb                | 65 +++++++++++++------
 2 files changed, 95 insertions(+), 32 deletions(-)

diff --git a/darts/models/forecasting/neural_prophet_model.py b/darts/models/forecasting/neural_prophet_model.py
index 8d7e1ff883..1194138f00 100644
--- a/darts/models/forecasting/neural_prophet_model.py
+++ b/darts/models/forecasting/neural_prophet_model.py
@@ -15,7 +15,7 @@
 
 
 class NeuralProphet(GlobalForecastingModel):
-    def __init__(self, n_lags: int = 0, n_forecasts: int = 1, *kwargs):
+    def __init__(self, n_lags: int = 0, n_forecasts: int = 1, **kwargs):
         super().__init__()
 
         raise_if_not(n_lags >= 0, "Argument n_lags should be a non-negative integer")
@@ -23,7 +23,7 @@ def __init__(self, n_lags: int = 0, n_forecasts: int = 1, *kwargs):
         self.n_lags = n_lags
         self.n_forecasts = n_forecasts
         self.model = neuralprophet.NeuralProphet(
-            n_lags=n_lags, n_forecasts=n_forecasts, *kwargs
+            n_lags=n_lags, n_forecasts=n_forecasts, **kwargs
         )
 
     def fit(
@@ -34,10 +34,25 @@ def fit(
     ) -> "NeuralProphet":
         super().fit(series, past_covariates, future_covariates)
 
+        # TODO accept list of univariate series or one multivariate ???
+        # TODO series have to have the same frequency
+
         self.training_series = self._as_sequence(series)
-        fit_df = self._convert_ts_to_df(self.training_series)
+        self.train_past_cov = (
+            self._as_sequence(past_covariates) if past_covariates is not None else None
+        )
+        self.train_future_cov = (
+            self._as_sequence(future_covariates)
+            if future_covariates is not None
+            else None
+        )
+
+        fit_df = self._convert_ts_to_df(
+            self.model, self.training_series, self.train_past_cov, self.train_future_cov
+        )
 
-        self.model.fit(fit_df, freq=series.freq_str)
+        # TODO check if all time series has common frequency string
+        self.model.fit(fit_df, freq=series[0].freq_str)
 
         return self
 
@@ -80,10 +95,21 @@ def predict(
             )
         return self._from_sequence(predictions)
 
-    def _convert_ts_to_df(self, series_list: Sequence[TimeSeries]):
+    def _convert_ts_to_df(
+        self,
+        model: neuralprophet.NeuralProphet,
+        series_list: Sequence[TimeSeries],
+        past_cov: Optional[Sequence[TimeSeries]],
+        future_cov: Optional[Sequence[TimeSeries]],
+    ):
+        raise_if_not(
+            len(past_cov) == 0 or len(past_cov) == len(series_list),
+            "Number of past covariates has to be zero or equal to number of fit time series.",
+        )
+
         dfs = []
 
-        for series in series_list:
+        for i, series in enumerate(series_list):
             for component in series.components:
                 new_df = (
                     series[component].pd_dataframe(copy=False).reset_index(names=["ds"])
@@ -94,7 +120,19 @@ def _convert_ts_to_df(self, series_list: Sequence[TimeSeries]):
                     .rename(columns={component: "y"})
                 )
                 component_df["ID"] = component
-                dfs.append(component_df)
+
+            if past_cov is not None:
+                for component in past_cov[i].components:
+                    covaraite_df = (
+                        past_cov[i].pd_dataframe(copy=False).reset_index(names=["ds"])
+                    )
+                    covaraite_df = covaraite_df[["ds", component]].copy(deep=True)
+
+                    # TODO add checks if past covariate has full coverage
+                    component_df = component_df.merge(covaraite_df, how="left", on="ds")
+                    model.add_lagged_regressor(names=component)
+
+            dfs.append(component_df)
 
         return pd.concat(dfs)
 
@@ -129,15 +167,15 @@ def _convert_df_to_ts(self, forecast: pd.DataFrame, last_train_date, components)
         )
 
     def _as_sequence(
-        self, series: Union[TimeSeries, Sequence[TimeSeries]]
+        self, series: Optional[Union[TimeSeries, Sequence[TimeSeries]]]
     ) -> Sequence[TimeSeries]:
+        if series is None:
+            return []
+
         if isinstance(series, TimeSeries):
             return [series]
 
-        if isinstance(series, Sequence[TimeSeries]):
-            return series
-
-        raise ValueError("Invalid type. Expected TimeSeries or Sequence[TimeSeries]")
+        return series
 
     def _from_sequence(
         self, series_list: Sequence[TimeSeries]
diff --git a/neural_examples/examples.ipynb b/neural_examples/examples.ipynb
index da92f4e30a..0f8eb6e105 100644
--- a/neural_examples/examples.ipynb
+++ b/neural_examples/examples.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 1,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -39,7 +39,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -48,16 +48,16 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 18,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [
     {
      "data": {
       "text/plain": [
-       "<matplotlib.legend.Legend at 0x7fa41b10c7c0>"
+       "<matplotlib.legend.Legend at 0x7fbf0159e850>"
       ]
      },
-     "execution_count": 18,
+     "execution_count": 3,
      "metadata": {},
      "output_type": "execute_result"
     },
@@ -79,7 +79,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -95,7 +95,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
@@ -114,7 +114,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "4fcfd515771f44ec813f43821b22511d",
+       "model_id": "d0bb081224694a05bd9813637f34f081",
        "version_major": 2,
        "version_minor": 0
       },
@@ -135,7 +135,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "df98919dae0d4576858e0762547063b3",
+       "model_id": "5cf9537be2ca45319a0eaf26e20a927c",
        "version_major": 2,
        "version_minor": 0
       },
@@ -152,7 +152,7 @@
      "text": [
       "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 2.33E-01, min: 2.84E-01\n",
       "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.00E-01\n",
-      "Epoch[514/514]: 100%|██████████| 514/514 [00:03<00:00, 140.22it/s, SmoothL1Loss=0.00124, MAE=10.9, RMSE=14.2, Loss=0.000911, RegLoss=0]\n",
+      "Epoch[514/514]: 100%|██████████| 514/514 [00:03<00:00, 137.41it/s, SmoothL1Loss=0.00124, MAE=10.9, RMSE=14.2, Loss=0.000911, RegLoss=0]\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
       "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 88.889% of the data.\n",
@@ -177,7 +177,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 6,
    "metadata": {},
    "outputs": [
     {
@@ -196,7 +196,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "255fc4100cbd481a84d083205c313f0d",
+       "model_id": "47a9bdcf22c9446f94fb1121f02b96b2",
        "version_major": 2,
        "version_minor": 0
       },
@@ -211,13 +211,13 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-02, min: 2.41E-02\n"
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-02, min: 2.76E-01\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "c234eb4ee5d842eebbc626d1c78ea033",
+       "model_id": "f0e614f3c33e475091129626e81c8008",
        "version_major": 2,
        "version_minor": 0
       },
@@ -232,9 +232,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-02, min: 2.41E-02\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.25E-02\n",
-      "Epoch[1000/1000]: 100%|██████████| 1000/1000 [00:04<00:00, 224.43it/s, SmoothL1Loss=0.000167, MAE=4.09, RMSE=5.28, Loss=0.000138, RegLoss=0]\n",
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 8.73E-03, min: 2.25E-01\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.11E-02\n",
+      "Epoch[1000/1000]: 100%|██████████| 1000/1000 [00:04<00:00, 210.17it/s, SmoothL1Loss=0.000164, MAE=4, RMSE=5.23, Loss=0.000136, RegLoss=0] \n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
       "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.278% of the data.\n",
@@ -278,19 +278,19 @@
     }
    ],
    "source": [
-    "model = NeuralProphetDarts(n_lags=36, n_forecasts=36)\n",
+    "model = NeuralProphetDarts(n_lags=36, n_forecasts=36, n_changepoints=20)\n",
     "model.fit(train_air)\n",
     "preds_ar = model.predict(36)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 7,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -305,6 +305,31 @@
     "preds_simple.plot(label=\"trend & season\")\n",
     "preds_ar.plot(label=\"auto-regression\")"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "air_year = datetime_attribute_timeseries(series_air, attribute=\"year\")\n",
+    "air_month = datetime_attribute_timeseries(series_air, attribute=\"month\")\n",
+    "air_covariates = air_year.stack(air_month)\n",
+    "scaler = Scaler()\n",
+    "air_covariates = scaler.fit_transform(air_covariates)\n",
+    "air_covariates.plot()"
+   ]
   }
  ],
  "metadata": {

From b4f2f18aa39fa1d0411377628a2a59e9e8efc90d Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Thu, 15 Dec 2022 20:21:11 +0100
Subject: [PATCH 05/15] Add past covariates

---
 .../forecasting/neural_prophet_model.py       |  169 +-
 neural_examples/examples.ipynb                | 1638 ++++++++++++++++-
 2 files changed, 1681 insertions(+), 126 deletions(-)

diff --git a/darts/models/forecasting/neural_prophet_model.py b/darts/models/forecasting/neural_prophet_model.py
index 1194138f00..f40cb0583a 100644
--- a/darts/models/forecasting/neural_prophet_model.py
+++ b/darts/models/forecasting/neural_prophet_model.py
@@ -3,18 +3,18 @@
 ------------
 """
 
-from typing import Optional, Sequence, Tuple, Union
+from typing import Optional, Sequence, Union
 
 import neuralprophet
 import pandas as pd
 from neuralprophet.utils import fcst_df_to_latest_forecast
 
 from darts.logging import raise_if_not
-from darts.models.forecasting.forecasting_model import GlobalForecastingModel
+from darts.models.forecasting.forecasting_model import ForecastingModel
 from darts.timeseries import TimeSeries, concatenate
 
 
-class NeuralProphet(GlobalForecastingModel):
+class NeuralProphet(ForecastingModel):
     def __init__(self, n_lags: int = 0, n_forecasts: int = 1, **kwargs):
         super().__init__()
 
@@ -28,113 +28,99 @@ def __init__(self, n_lags: int = 0, n_forecasts: int = 1, **kwargs):
 
     def fit(
         self,
-        series: Union[TimeSeries, Sequence[TimeSeries]],
-        past_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
-        future_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
+        series: TimeSeries,
+        past_covariates: Optional[TimeSeries] = None,
+        future_covariates: Optional[TimeSeries] = None,
     ) -> "NeuralProphet":
-        super().fit(series, past_covariates, future_covariates)
+        super().fit(series)
 
-        # TODO accept list of univariate series or one multivariate ???
-        # TODO series have to have the same frequency
-
-        self.training_series = self._as_sequence(series)
-        self.train_past_cov = (
-            self._as_sequence(past_covariates) if past_covariates is not None else None
-        )
-        self.train_future_cov = (
-            self._as_sequence(future_covariates)
-            if future_covariates is not None
-            else None
+        raise_if_not(
+            series.has_datetime_index,
+            "NeuralProphet model is limited to TimeSeries index with DatetimeIndex",
         )
 
-        fit_df = self._convert_ts_to_df(
-            self.model, self.training_series, self.train_past_cov, self.train_future_cov
+        raise_if_not(
+            past_covariates is None or self.n_lags > 0,
+            "Past covariates are only supported when auto-regression is enabled (n_lags > 1)",
         )
 
-        # TODO check if all time series has common frequency string
-        self.model.fit(fit_df, freq=series[0].freq_str)
+        self.training_series = series
+        fit_df = self._convert_ts_to_df(series)
 
+        if past_covariates is not None:
+            fit_df = self._add_past_covariate(self.model, fit_df, past_covariates)
+
+        # TODO add future covariates to df
+
+        self.model.fit(fit_df, freq=series.freq_str, minimal=True)
+
+        self.fit_df = fit_df
         return self
 
     def predict(
         self,
         n: int,
-        series: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
-        past_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
-        future_covariates: Optional[Union[TimeSeries, Sequence[TimeSeries]]] = None,
+        future_covariates: Optional[TimeSeries] = None,
         num_samples: int = 1,
         verbose: bool = False,
     ) -> Union[TimeSeries, Sequence[TimeSeries]]:
-        super().predict(
-            n, series, past_covariates, future_covariates, num_samples, verbose
-        )
-
-        if series is None:
-            series_list = self.training_series
-        else:
-            series_list = self._as_sequence(series)
+        super().predict(n, num_samples)
 
         raise_if_not(
             self.n_lags == 0 or n <= self.n_forecasts,
             "Auto-regression has been configured. `n` must be smaller than or equal to"
             "`n_forecasts` parameter in the constructor.",
         )
-        # TODO consider time series indexed by ints
-        # TODO check if series was used during training
-
-        predictions = []
-        for series in series_list:
-            df = self._convert_ts_to_df([series])
-            future_df = self.model.make_future_dataframe(df=df, periods=n)
-            forecast_df = self.model.predict(future_df)
-
-            predictions.append(
-                self._convert_df_to_ts(
-                    forecast_df, series.end_time(), series.components
-                )
-            )
-        return self._from_sequence(predictions)
 
-    def _convert_ts_to_df(
-        self,
-        model: neuralprophet.NeuralProphet,
-        series_list: Sequence[TimeSeries],
-        past_cov: Optional[Sequence[TimeSeries]],
-        future_cov: Optional[Sequence[TimeSeries]],
-    ):
-        raise_if_not(
-            len(past_cov) == 0 or len(past_cov) == len(series_list),
-            "Number of past covariates has to be zero or equal to number of fit time series.",
+        future_df = self.model.make_future_dataframe(df=self.fit_df, periods=n)
+        forecast_df = self.model.predict(future_df)
+
+        return self._convert_df_to_ts(
+            forecast_df,
+            self.training_series.end_time(),
+            self.training_series.components,
         )
 
-        dfs = []
+    def _convert_ts_to_df(self, series: TimeSeries) -> pd.DataFrame:
+        """Convert TimeSeries to pandas DataFrame format required by Neural Prophet"""
+        dfs = []  # ID y
+
+        for component in series.components:
+            component_df = (
+                series[component]
+                .pd_dataframe(copy=False)
+                .reset_index(names=["ds"])
+                .filter(items=["ds", component])
+                .rename(columns={component: "y"})
+            )
+            component_df["ID"] = component
+            dfs.append(component_df)
 
-        for i, series in enumerate(series_list):
-            for component in series.components:
-                new_df = (
-                    series[component].pd_dataframe(copy=False).reset_index(names=["ds"])
-                )
-                component_df = (
-                    new_df[["ds", component]]
-                    .copy(deep=True)
-                    .rename(columns={component: "y"})
-                )
-                component_df["ID"] = component
+        return pd.concat(dfs).copy(deep=True)
 
-            if past_cov is not None:
-                for component in past_cov[i].components:
-                    covaraite_df = (
-                        past_cov[i].pd_dataframe(copy=False).reset_index(names=["ds"])
-                    )
-                    covaraite_df = covaraite_df[["ds", component]].copy(deep=True)
+    def _add_past_covariate(
+        self,
+        model: neuralprophet.NeuralProphet,
+        df: pd.DataFrame,
+        past_covariates: TimeSeries,
+    ) -> pd.DataFrame:
+        """Convert past covariates from TimeSeries and add them to DataFrame"""
+
+        # TODO add checks if past covariate Time series has enough coverage and the same frequency
+
+        for component in past_covariates.components:
+            covariate_df = (
+                past_covariates[component]
+                .pd_dataframe(copy=False)
+                .reset_index(names=["ds"])
+                .filter(items=["ds", component])
+            )
 
-                    # TODO add checks if past covariate has full coverage
-                    component_df = component_df.merge(covaraite_df, how="left", on="ds")
-                    model.add_lagged_regressor(names=component)
+            df = df.merge(covariate_df, how="left", on="ds")
 
-            dfs.append(component_df)
+            model.add_lagged_regressor(names=component)
 
-        return pd.concat(dfs)
+        return df
 
     def _convert_df_to_ts(self, forecast: pd.DataFrame, last_train_date, components):
         groups = []
@@ -166,26 +152,5 @@ def _convert_df_to_ts(self, forecast: pd.DataFrame, last_train_date, components)
             axis=1,
         )
 
-    def _as_sequence(
-        self, series: Optional[Union[TimeSeries, Sequence[TimeSeries]]]
-    ) -> Sequence[TimeSeries]:
-        if series is None:
-            return []
-
-        if isinstance(series, TimeSeries):
-            return [series]
-
-        return series
-
-    def _from_sequence(
-        self, series_list: Sequence[TimeSeries]
-    ) -> Union[TimeSeries, Sequence[TimeSeries]]:
-        if len(series_list) == 1:
-            return series_list[0]
-        return series_list
-
-    def _model_encoder_settings(self) -> Tuple[int, int, bool, bool]:
-        raise NotImplementedError()
-
     def __str__(self):
         return "Neural Prophet"
diff --git a/neural_examples/examples.ipynb b/neural_examples/examples.ipynb
index 0f8eb6e105..091a5d0eae 100644
--- a/neural_examples/examples.ipynb
+++ b/neural_examples/examples.ipynb
@@ -18,12 +18,12 @@
     "from darts.metrics import mape, smape\n",
     "from darts.dataprocessing.transformers import Scaler\n",
     "from darts.utils.timeseries_generation import datetime_attribute_timeseries\n",
-    "from darts.datasets import AirPassengersDataset, MonthlyMilkDataset\n",
+    "from darts.datasets import *\n",
     "from darts.models.forecasting.neural_prophet_model import (\n",
     "    NeuralProphet as NeuralProphetDarts,\n",
     ")\n",
     "from neuralprophet import NeuralProphet\n",
-    "\n",
+    "import neuralprophet\n",
     "\n",
     "# for reproducibility\n",
     "torch.manual_seed(1)\n",
@@ -54,7 +54,7 @@
     {
      "data": {
       "text/plain": [
-       "<matplotlib.legend.Legend at 0x7fbf0159e850>"
+       "<matplotlib.legend.Legend at 0x7fbcda1f37f0>"
       ]
      },
      "execution_count": 3,
@@ -114,7 +114,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "d0bb081224694a05bd9813637f34f081",
+       "model_id": "9dec31c451db4ddea6feb12df9a7dd36",
        "version_major": 2,
        "version_minor": 0
       },
@@ -129,13 +129,13 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 3.45E-01, min: 4.02E-02\n"
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 3.30E-02, min: 1.07E-01\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "5cf9537be2ca45319a0eaf26e20a927c",
+       "model_id": "cf05800d08644e4b83b9865a345b29e2",
        "version_major": 2,
        "version_minor": 0
       },
@@ -150,9 +150,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 2.33E-01, min: 2.84E-01\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.00E-01\n",
-      "Epoch[514/514]: 100%|██████████| 514/514 [00:03<00:00, 137.41it/s, SmoothL1Loss=0.00124, MAE=10.9, RMSE=14.2, Loss=0.000911, RegLoss=0]\n",
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 3.30E-02, min: 8.78E-02\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 4.78E-02\n",
+      "Epoch[514/514]: 100%|██████████| 514/514 [00:03<00:00, 161.81it/s]\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
       "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 88.889% of the data.\n",
@@ -196,7 +196,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "47a9bdcf22c9446f94fb1121f02b96b2",
+       "model_id": "3bc3d5351b6240d29c4d2b5b41249e08",
        "version_major": 2,
        "version_minor": 0
       },
@@ -211,13 +211,13 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-02, min: 2.76E-01\n"
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.97E-02, min: 4.44E-02\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "f0e614f3c33e475091129626e81c8008",
+       "model_id": "640c57be3d854c4da0ece842b0b53b78",
        "version_major": 2,
        "version_minor": 0
       },
@@ -232,9 +232,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 8.73E-03, min: 2.25E-01\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.11E-02\n",
-      "Epoch[1000/1000]: 100%|██████████| 1000/1000 [00:04<00:00, 210.17it/s, SmoothL1Loss=0.000164, MAE=4, RMSE=5.23, Loss=0.000136, RegLoss=0] \n",
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.97E-02, min: 4.44E-02\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 3.38E-02\n",
+      "Epoch[1000/1000]: 100%|██████████| 1000/1000 [00:03<00:00, 266.20it/s]\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
       "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.278% of the data.\n",
@@ -290,7 +290,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiMAAAHGCAYAAACmS4sdAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAADjtUlEQVR4nOydd3wUZf7H37Pp2fQCBAgpQGih9ybSiwVUQDw9FT3RO8upZ71Tz3Z6J2f92fEUK4pYAJVelGboHUKAFFKAFNLLtvn98ezO7oaEJKRsyvN+vXgxfZ7Z3cx85lsVVVVVJBKJRCKRSFyEztUDkEgkEolE0raRYkQikUgkEolLkWJEIpFIJBKJS5FiRCKRSCQSiUuRYkQikUgkEolLkWJEIpFIJBKJS5FiRCKRSCQSiUuRYkQikUgkEolLadNixGKxkJycjMVicfVQ6k1ruhaQ19OcaU3XAq3relrTtYC8nuZMQ19LmxYjEolEIpFIXI8UIxKJRCKRSFyKFCMSiUQikUhcihQjEolEIpFIXIoUIxKJRCKRSFyKFCMSiUQikUhcihQjEolEIpFIXIoUIxKJRCKRSFyKFCMSiUQikUhcihQjEolEIpFIXIoUIxKJRCKRSFyKFCMSiUQikUhcihQjEolEIpFIXIoUIy7kyiuv5MEHH2yw4z366KNcd911DXY8iUQikUiaAilGJBKJRNJmUS0qZWfKUFXV1UNp00gx4iJuv/12fv31V958800URUFRFFJSUjh8+DDTp0/Hz8+P9u3b88c//pGcnBxtv2XLltG3b198fHwIDQ1l0qRJlJSU8Nxzz/Hdd9+xYsUK7XibN2923QVKJBJJC+Dok8fZNOA39s0/IAWJC5FixEW8+eabjBw5krvuuousrCyysrLw9/dnwoQJDBw4kN27d7N69WrOnTvH3LlzAcjKyuKmm27ijjvu4NixY2zevJnrr78eVVX529/+xlVXXcXUqVO1440aNcrFVymRSCTNF9Wskv5VOgBnV54j4+tMF4+o7eLu6gE0BkOGDOHs2bO12tZsNuPm5tYg5+3QoQO7d++u1baBgYF4enri6+tLhw4dAHjxxRcZOHAgL730krbdxx9/TGRkJCdOnKC4uBiTycT1119PVFQUAH379gXAYrHg7e2N0WjUjieRSCSS6ilOKsFcatHmjz2TSLsp4XiGerpwVG2TVilGzp49S0ZGhquHUWcOHDjApk2b8PPzu2jdqVOnmDJlChMnTqRv375MnTqVKVOmMHv2bIKDg10wWolEImnZFB4odJo35hk59s9E+r/d10Ujaru0SjFSF8tAQ1tG6kNxcTHXXHMN//nPfy5aFxERgZubG+vWrWP79u2sXbuW//u//+Mf//gHCQkJmqVEIpFIJLXj/M7zFy3LWJJJ53mdCB0T4oIRtV1apRipravEYrGQmppKVFQUOl3Th894enpiNpu1+UGDBvHdd98RHR2Nu3vVX42iKIwePZrRo0fzzDPPEBUVxQ8//MCDDz6Ih4cH5eXlTTV8iUQiadHk7M7Vpj8rXcytvrcDsPf+/UzYMQ4374Z5UZXUjAxgdSHR0dEkJCSQkpJCTk4O9957L3l5edx0003s2rWLU6dOsWbNGubPn4/ZbCYhIYGXXnqJ3bt3k5aWxvfff092dja9evUCoFOnThw6dIjExERycnIwGo0uvkKJRCJpnqgWFUOSuEeeN59jtecqjhmPAmBMM5L43xOuHF6bQ4oRF/LII4/g5uZG7969CQ8Px2AwsG3bNsxmM1OmTKFv3748+OCDBAUFodPpCAgI4LfffmPGjBnExcXx1FNP8eqrrzJ9+nQA5s2bR1xcHEOGDCE8PJxt27a5+AolEomkeVJysgSlQgHgpPkkr7/xOj+1X4FJNQFw6s1kik8Uu3KIbYpW6aZpKcTFxbFjx46Lln///fdVbt+rVy9Wr15d7fFCQ0NZs2aNS1xOEolE0pIocAhePWlK4vr+s3j7x7d5fdSbXO9+A24WN9bfvJGZO69BURQXjrRtIJ9aEolEImlzOIkRcxJdu3alT58+jH9jHFlmUW/E87QXO/7zu6uG2KaQYkQikUgkbQ7HtN5c/1wCAwMBuHn+zaSOT9HWZfw3i5zTOZV3lzQwUoxIJBKJpE2hWlQKDgoxct58nnbd2jmt//uSJzmoPwCAXtXz5dVLZKn4RkaKEYlEIpG0KUpOlWIuFmUVbC4aR7y9vbnh++soUUUAa/dzcXz41w+bfJxtCSlGJBKJRNKmKDhQoE2fMp26SIwAdB/SHZ87vLV53WcebN20tUnG1xaRYkQikUgkbYrCg0XadFWWERtXvTKDgg5CuHRy68RX85Zw/vzFVVsl9UeKEYlEIpG0KRwtIydN1YsRRacwfekUzIpw6Uw1Tue+6+9zqpwtaRikGJFIJBJJm0G1qBQeEJaRHEs2+Wp+tWIEIKhPEJ3uigDAQ/FgyMFhPPfsc00y1raEFCMSiUQiaTOUJpdiKhJVVk+aTuLt7V1jk9P+z/RD6SAKn8V79CXhlZ2sWrWq0cfalpBipA3z7LPPMmDAAFcPQyKRSJqMypVXY2Nja6xa7ebjxtB3Bmvzd/j+ib/c8hfS0tIabZxtDSlGXMSVV17Jgw8+6Oph1IjFYuHxxx+nY8eO+Pj40K9fP5YvX+7qYUkkEsll4Vx59eQlXTSOhF0ZSsc5wl3jr/Pnhoo53HrrrY0yxraIFCPNGFVVMZlMLh3DF198weuvv85rr73GsWPHeO2119Dr9S4dk0QikVwuhfudLSO1FSMAvV7ogVugGwATvCZSsK2QnBxZnbUhkGLEBdx+++38+uuvvPnmmyiKgqIopKSksHnzZhRFYdWqVQwePBgvLy+2bt2KxWLh5ZdfJiYmBh8fH/r378+yZcu0423evBk3Nze2bdvGsGHD8PX1ZdSoUSQmJjqd99///jft27fH39+fO++8k/Ly8hrHqtPpCA8PZ968eURHRzNp0iQmTZpU434Gg4H77ruPiIgIvL29iYqK4uWXX9bW5+fn86c//Ynw8HACAgKYMGECBw4c0NafOnWKBQsWEBERgZ+fH0OHDmX9+vVO53j33Xfp3r073t7etG/fntmzZ2vrKioqeOCBB2jXrh3e3t6MGTOGXbt2OX1miqKwYcMGhgwZUu1nJpFIWg+qaq+8mmPJ4YJ6oU5ixCvci97P99Tm/6y/l5TklIYeZptEihEX8OabbzJy5EjuuususrKyyMrKIjIyUlv/xBNP8O9//5tjx47Rr18/Xn75ZT777DPef/99jhw5wkMPPcQtt9zCr7/+6nTcV199lYULF7J7927c3d254447tHVLly7l2Wef5aWXXmL37t1ERETw7rvv1jjWiRMnUlBQwNNPP12na3zrrbdYsWIFS5cuJTExkS+//JLo6Ght/Zw5czh//jyrVq1iz549DBo0iIkTJ5KXlwdAcXExV155JevWrWPfvn1MmzaNa665RvPR7t69mwceeIDnn3+exMREVq9ezRVXXKEd/7HHHuO7777j008/Ze/evXTr1o2pU6dqx7fxj3/8g1dffbXKz0wikbQuSpNLMRUKa/Mp00mAOokRgM5/6ERJhKjM2tmtM2kJMm6kIXC/nJ0+/fRTli5dSlFREZ07d2bRokXo9XoWL17MF198gcViYebMmTzwwANa6+UjR47wwgsvcObMGfr06cNzzz1HREREg16MjSF3WTibV/N2qGA2d8LNDVAs9T5vhxDYvahmfRcYGIinpye+vr5VRnE///zzTJ48GRBv+C+99BLr169n5MiRAMTGxrJ161Y++OADxo0bp+33t7/9jXHjxqHT6XjiiSe46qqrKC8vx9vbmzfeeIM777yTO++8E4AXX3yR9evXX9I6UlpayuTJk/nDH/7AunXrKCsrY+HChdp3GhAQwMcff+xkkbCRlpZG9+7dGTNmDIqiEBUVpa3bunUrO3fu5Pz583h5eQHw3//+lx9//JFly5axYMEC+vfvT1BQEFFRUeh0Ol544QV++OEHVqxYwX333UdaWhp6vZ6rr74af39/oqKiGDhwIAAlJSW89957LF68mOnTpwOwaNEi1q1bx//+9z8effRRbSz/+te/tM+w8mcmkUhaF5XjRaDuYkTRKeiGKrBCzOftvNBg42vL1FmMLF26lB07dvC///2P9u3bc/LkSTw8PNi6dSvffvstixcvxtvbm3vvvZeoqChmzZqFwWDgscce46677mL69Ol89NFHPP3003z00UeNcU2czYOM7NpufVl6rFEZMmSINn3y5ElNFDhiMBi0h6+Nnj3t5kOb0Dt//jxdunTh2LFj3HPPPU7bjxw5kk2bNlU7jsWLF5Ofn88777yjWSrmz5/PRx99RHp6OsXFxYwePbrKfW+//XYmT55Mjx49mDZtGldffTVTpkwB4MCBAxQXFxMaGuq0T1lZGadOnQKEZeSll15i69atZGVlYTKZKCsr0ywjkydPJioqitjYWKZNm8a0adO47rrr8PX15dSpUxiNRqexeXh4MGzYMI4dO+Z0zn79+lX7mUkkktZFYaVMGp1O52SxrS3ho8IoXlEGgPl4/V9kJXV8EpvNZj7++GM++ugj7Y2+e/fuAPzyyy9cd911dO7cGYBbbrmFlStXMmvWLPbs2YOHhwezZs0C4M4772TixIlkZGTQqVOnBrwcQYeQWm6ogtlsws3NHZQmPG8NOAaIFhcLc+DPP/980WdlsyrY8PDw0KZt1guL5fL/UA4ePEifPn3w8PAgODiYdevWMXbsWK677jq6d+/OtGnTqrVuDRo0iOTkZFatWsX69euZO3cukyZNYtmyZRQXFxMREcHmzZsv2i8oKAiARx99lLVr1/Laa68RFxeHj48Ps2fPxmAwAODv78/evXvZvHkza9eu5ZlnnuHZZ591igupDQ39mUkkkuZL5bTeyC6ReHp61vk4sRNj2asewF1xR5/p15BDbLPUSYycP3+e8vJy1q9fz1dffYWfnx9//OMfue6660hOTmbq1Knatt26ddPeck+fPq2JFhAdETt37szp06erFCMGg0F76GgDdXev9Y9m5we1ux6LxcKZMxlERkbWmGdeW2r7IPPw8MBkMjltb5u2WCzadM+ePfHy8iIlJYWxY8dWeb7qjuG4vlevXvz+++/ccsst2ra///77JcfcsWNHfvjhBwoKCvD39ycsLIy1a9cybtw4fvrpJ3bt2nXJ6/Xz82POnDnMmTOH66+/nhkzZpCTk8OAAQM4e/ZstW8lFouFbdu2ccMNNzBz5kx0Oh3FxcWkpKSgqqp2Tp1Ox4QJE5gwYQJPP/00ISEhrF+/nqlTp+Lp6cmWLVu0WByj0ciuXbv461//6vSZVZ6u6jNtCCqfoyXTmq4FWtf1tKZrgYa9HlVVKbQGr+ZZ8shT8xjQdcBlHbtDlw6ctvxAnFscIWUhlOdW4BnsUeN+ren7qcu11Ob5WmcxUlxcTFpaGitWrODMmTP8+c9/Jjo6mtLSUqc3er1eT1mZMGOVlZVdlA6q1+spLS2t8jyffPIJixYtclo2Z84c5s6dW5fh1pozZ840ynEvRVhYGFu2bGHbtm34+voSFBTEuXPntPEUFNh7J/zpT3/ioYceIjs7myFDhlBUVMSePXvw8/Pjhhtu0PZzvJasrCwAMjIyUBSFm266iUcffZTo6GgGDx7M8uXLOXz4MJGRkaSmplY5xilTpvDGG28wbdo0/va3vxEcHMyuXbsoLCzEx8eHN998k+eff77KfT/66CPatWtHnz59UBSFxYsXEx4eTkFBAd27d2fgwIFcffXVPPHEE8TExHDu3Dk2bdrElClT6NevH506dWLNmjVMnDgRRVF47bXXMJvNFBYWkpqayoYNGzhz5gzDhg0jMDCQTZs2YbFY8Pf3Jycnhz/84Q888sgjmEwmOnbsyIcffkhxcTGTJ08mNTW1ys+68mfWGLjit9ZYtKZrgdZ1Pa3pWqBhrseQbsCYb6u8mgRAu3btqr3/1US6zxniDHEAnPjlBP5jam8haU3fT22uJSYmpsZt6iRGbG6Bu+66C29vb7p3786UKVO0B2pJSYm2bUlJCT4+PgD4+Pg4rbOt9/X1rfI88+fP5+abb3YeaB0sI7VFWEbONKhlpLb885//ZP78+UydOlWLlWjfvj0AkZGRmrsCRPZNt27deP/99/n73/9OUFAQAwcO5MknnyQqKork5GRtW9u1XLgggqo6depEVFQU9957L4WFhSxcuJDy8nKuv/56/vznP7N27Vqn4FJHoqKi2LlzJ0888QT33XcfhYWFDBkyhM8++wxfX1+mTp3KgAEDeOihhy7at0uXLrz//vskJSXh5ubG0KFDWbVqlfajXL9+PU899RRPPvkk2dnZdOjQgbFjxzJgwAAiIyN55513uPXWW5kzZw5hYWE89thjmEwmAgICiIqKIi4uji+++IL/+7//o7y8nO7du/Pll19qsTXvvPMOjz/+OI8++ihFRUUMGTKENWvWaDEits/M8bOu/Jk1JK78rTU0relaoHVdT2u6FmjY6zm73/7SdtIsxEj//v0v+2+9rHMpnBbT5mNmom6u+Tit6ftp6GtRVFVVa7txWVkZEyZM4IcfftBiRhYuXIiPjw+ZmZl07dpVy9b46aefWLFiBR9++CE7duzglVde4YcffgCgvLyciRMnsnTp0kaJGaktFouF1NRULWOjJdOargXk9TRnWtO1QOu6ntZ0LdCw13P8uROcfku8hDxf9CwJxt9ZunQpc+bMuazj/fWPDzL1F5Gt59HfnckbJ9a4T2v6fhr6Wup0BB8fHyZOnMj//vc/DAYDycnJrFu3jtGjRzNjxgy+//570tPTyc3N5csvv2TGjBkADB48mIqKCpYvX47BYODjjz+mV69eLhUiEolEImk7VA5ehbqn9TrSoXd7zpuFtcVwzIjF1PLjQFxJneXM448/Tn5+PpMmTeKvf/0r99xzDwMHDmTMmDHMnj2b2267jdmzZzNixAhmzpwJgKenJwsXLmTJkiWMHz+effv28cILLzT4xUgkEolEUhlVVSk8IGLDit2LyVVzgfqJkejoaI6ZRKkAxaBQdKS4/gNtw9S5yIa/vz8LFy6sct38+fOZP39+lev69OnD119/XdfTSSQSiURSL8rSyrTg1WSLCPQIDQ0lMDDwso8ZHR3NEtPXjPO6EoALCRcI7B9Q77G2VVq200oikUgkkhpwdNEcKTkM1M8qAkKMHDUd1eYv7Mqv1/HaOlKMSCQSiaRV41h5NakB4kUA2rdvT6Z7BuWqaKlxYWd+vY7X1pFiRCKRSCStGueeNA0jRnQ6HZFRkSSajgNQnl5OWUbNndAlVSPFiEQikUhaLaqqamLE5Gsix5ID1F+MgHMQK0D+7vx6H7OtIsWIRCKRSFot5enlGPOMAFwIsrdzbzgx4hA3Il01l40UIxKJRCJptRTst7toUhV76feGEiPHHSwjUoxcPlKMSFodKSkpKIrC/v37XT0UiUTiYhzjRQ4VHwREAc/qOo7XhejoaIrVYtLMQuQUHizEXGau93HbIlKMtGAWL17s1MNGIoiMjCQrK4v4+HhXD0UikbiYggP2pqM7srYDEBsb2yDNMG1dx48ZhXVENalOlhhJ7ZFiRFIlqqpiMpka5dhms7lRW2i7ubnRoUMH3N3rXNNPIpG0IkTlVSEO3ILdyCjPAIQYaQg0MeIUN3KhQY7d1pBixIWsXr2aMWPGEBQURGhoKFdffTWnTp0CYPPmzSiKQn5+vrb9/v37URSFlJQUNm/ezPz58ykoKEBRFNzc3HjjjTcA0X321ltvJTg4GF9fX6ZPn05SUtIlx2I736pVqxg8eDBeXl5s3boVi8XCyy+/TExMDD4+PvTv359ly5Y57btixQq6d++Ot7c348eP59NPP3Uau82Cs2LFCnr37o2XlxdpaWlUVFTwyCOP0KlTJ/R6PcOHD2fz5s3acTMyMrj22msJDg5Gr9fTp08ffvnlF+0ab775ZsLDw/Hx8aF79+588sknQNVuml9//ZVhw4bh5eVFREQETzzxhJPYuvLKK3nggQd47LHHCAkJoUOHDjz77LN1+DYlEklzozyjHEOuCF5Vo+wvQA0RLwKi1oiXl5cMYm0ApBhxISUlJTz88MPs3r2bDRs2oNPpuO6662plNRg1ahRvvPEGAQEBZGVlkZGRwV133QXA7bffzu7du1mxYgU7duxAVVVmzJiB0Wis8bhPPPEE//73vzl27Bj9+vXj5Zdf5rPPPuP999/nyJEjPPTQQ9xyyy38+uuvACQnJzN79mxmzZrFgQMHuPvuu/nHP/5x0XFLS0v5z3/+w0cffcSRI0do164d9913Hzt27ODrr7/m4MGDzJkzh2nTpmnC6ZlnnqGiooLffvuNQ4cO8Z///Ac/Pz8Ann76aY4ePcqqVas4duwY7733HmFhYVVeU0ZGBjNmzGDo0KEcOHCA9957j//973+8+OKLTtt9+umn6PV6EhISeOWVV3j++edZt25djZ+ZRCJpnji6TAqD7e6ahhIjOp2OqKgoMiwZFKlFAOTvykdV1QY5fluiVdqxt07YgeF8RY3bqQiXwWm3VOrvPQTPdl6M2Tiy1tvfcMMNTvMff/wx4eHhHD16tJo9HM7l6UlgYCCKotChQwcsFgsVFRUkJSWxYsUKtm3bxqhRowD48ssviYyM5Mcff6yxXfbzzz/P5MmTAaioqOCll15i/fr1jBwpris2NpatW7fywQcfMG7cOD744AN69Oih9Svq0aMHhw8f5l//+pfTcY1GI++++y79+/cHIC0tjU8++YS0tDQ6duwIwCOPPMLq1av55JNPePHFF8nMzGTevHn07dtXO7eNtLQ0Bg4cyJAhQwC7ubQq3n33XSIjI3n77bdRFIWePXuSmZnJ448/zjPPPKO1v+7Xrx///Oc/AejevTtvv/02GzZs0D4PiUTSsig8aBcjZ9zPaNMNJUZA3HtOnDjBMeNRhnkOx5BrpPR0Kfqu+gY7R1ugVYoRw/kKyrNqFiM2TDRObERNJCUl8cwzz5CQkEBOTo5mEUlLS8PX1/eyjnns2DHc3d0ZPny4tiw0NJQePXpw7JgIspo+fTpbtmwBICoqiiNHjmjb2h7uACdPnqS0tPSih7HBYGDgwIEAJCYmMnToUKf1w4YNu2hcnp6e9OvXT5s/dOgQZrOZuLg4p+0qKioIDQ0FhIXn6aefZt26dUyaNIkbbrhBO8af//xnbrjhBvbu3cuUKVOYNWuWJr6q+kxGjhzpFLA2evRoiouLSU9Pp0uXLgBO4wOIiIjg/PnzVR5TIpE0f5x60pTa73MNLUZAxI0M8xT33QsJ+VKM1JFWKUY823nVajubZcTNza3BLCN14ZprriEqKopFixbRsWNHLBYL8fHxGAwGzR3haO6rjZulNnz00UeUlZUB4OHh4bROr7f/ARUXi5bYP//8M506dXLazsurbtfq4+PjJAaKi4txc3Njz549uLm5OW1ru/Ybb7yRm266iVWrVrF27VpefvllXn31Ve6//36mT59Oamoqv/zyC+vWrWPixInce++9/Pe//63TuByp/FkoitKogbYSiaTxcKy86hHswcGMA4BwrVzKklpX7GLEod7Irnw6/6FTNXtIqqJVipHaukosFgupqalERUVppvqmIjc3l8TERBYtWsTYsWMB2Lp1q7Y+PDwcgKysLIKDgwEuqpvh6emJ2eyc096rVy9MJhMJCQmapcB2rt69ewNcJCyqwzHYdNy4cVVu06NHDy2o1MauXbtqPPbAgQMxm82cP39eu35HbCIgMjKSe+65h3vuuYcnn3ySRYsWcf/99wPiM7rtttu47bbbGDt2LI8++miVYqRXr1589913qKqqCaJt27bh7+9P586daxyrRCJpeZRnVmDINgAQ2D+A05tOA+Ke4unp2WDnsYmRE6ZEVEVFURUZxHoZyABWFxEcHExoaCgffvghJ0+eZOPGjTz88MPa+m7duhEZGcmzzz5LUlISP//8M6+++qrTMaKjoykuLmbDhg3k5ORQVlZG9+7dmTlzJnfddRdbt27lwIED3HLLLXTq1ImZM2fWaYz+/v488sgjPPTQQ3z66aecOnWKvXv38n//9398+umnANx9990cP36cxx9/nBMnTrB06VIWL14McMk8/ri4OG6++WZuvfVWvv/+e5KTk9m5cycvv/wyP//8MyDiV9asWUNycjJ79+5l06ZN9OrVCxDBrcuXL+fkyZMcOXKEn376SVtXmb/85S+cOXOG+++/n+PHj7N8+XL++c9/8vDDDze5CJVIJE2DY6dezx6eXLggUm4b0kUDdjFSQQUF1iDZ4uPFGAsaxpLdVpB3Yheh0+n4+uuv2bNnD/Hx8Tz00ENaECgIl8GSJUs4fvw4/fr14z//+c9F2R+jRo3innvu4cYbb6R9+/Z88MEHAHzyyScMHjyYq6++mpEjR6KqKr/88stFboja8MILL/D000/z8ssv06tXL6ZNm8bPP/9MTEwMADExMSxbtozvv/+efv368d5772nZNDW5cj755BNuvfVW/va3v9GjRw9mzZrFrl27tBgOi8XC/fffr503Li6Od999FxBWoSeffJJ+/fpxxRVX4Obmxtdff13leTp16sQvv/zCzp076d+/P/fccw933nknTz31VJ0/D4lE0jJwLHZWHF6kTTeWGAFI80zRpmXTvLqhqG04B8mVbpqGpjldy7/+9S/ef/99zpw5U/PG1dCcrqchaE3X05quBVrX9bSma4H6Xc+ueXvIXic69BY/X8iN980F4N///jePP/54g47R19eXiooK/hB7Mzfn/xGAbo/EEvdk94u2bS3fT0NfS8v+NCTNgnfffZddu3Zx+vRpPv/8cxYuXMhtt93m6mFJJJI2ilPwapA7J/PtRR8b2jJiqzUCsPXcFm25jBupG60ygFXStCQlJfHiiy+Sl5dHly5d+Nvf/saTTz7p6mFJJJI2SkVWBYbzIng1oH8gp06f0tY1tBgBe62RtJI0PKM8MJw1kr+nAIvJgs5dvvPXBilGJPXm9ddf5/XXX3f1MCQSiQRwri8S2M+fU5vtYqSh+tI44hg3osQpcBbMJWaKjxUT0Degwc/XGpGSTSKRSCStigKHyquBAwK1nl+hoaEEBgY2+PkcxUhRB/u5paum9kgxIpFIJJJWhWNar1dPLzIyRLfexnDRgLMYSfdJ16alGKk9UoxIJBKJpFVha5DnHujOWUuWVsm6KcRIYvlxdD7i0XphV36jnK81IsWIRCKRSFoN5WcrqDgnepMF9gvg9OnT2rqmECPJackEDRSuoLLUMsrP1r5PWltGihGJRCKRtBoci50FDgjQ4kWg8cRI+/bttSKPKSkpBA0N0tZd2HmhUc7Z2pBiRCKRSCStBsd4kYD+TSNGHGuNCDFiD5LNl66aWiHFiEQiaXZk/pDF5qFbOPTwESqypZlbUnuc0nr7N42bBuyumpKSEtSu9sLmMoi1dkgxIpFImh0nXjpJ6elSznyazq9Dt5L8XgoWo8XVw5K0ALTg1QB3fGN8NcuIj48PERERjXZex7iRjMJ09N30YjwHCjGXm6vZS2JDihGJRNKsMBaaKD1dqs2bikwceyqRLWO3k70px4UjkzR3Ks5VUGENGA3o54+qqiQnJwOi2NmlOonXF0cxkpKSQvCwIABUo+pkrZFUjRQjEomkWVF0xN5h1TvCC6zPj5KkEnbN3sPum/dSklxazd6StoyziyaQjIwMKiqEOGlMFw1UL0YA8qWrpkakGJFIJM2KwkP2B0q3R7syat0Ip+yE86uz2TJqK4kvnMBUbHLBCCXNFafKq5WCVxujDLwjlcWIc0ZNfqOeuzUgxYhEImlWFB62W0YC4v0JGhjIyFXD6P9eX7zai/RJi0Hl1BvJ/Dp8KxnfZmpFrSRtm+LEYm06IN6/STJpbFQWI35xetwDRfu3C7vy5W+0BqQYkUgkzQpNjOjAv5c/AIqi0GluR8YljCH2rzHoPIXvpuJsBQfuOcTvM3ZSsL+gukNK2gglJ0vEhA6n4FVofDHSoUMHvL29ASFGFJ1CsNU6Ysg2UJpS1qjnb+lIMSKRSJoNFqOF4mNCjPh10+Pm6+a03t3fnZ7PxDF222jaTQvXll/Ymc+2Sb9z6MHDVOQYmnTMkuaBqqqUnBKxRL5RPug8dU0qRhRFcao1oqqqJkZAxo3UhBQjEomk2VCcVILFIMzZ/n39q91OH6tnyJeDGLp0kJZCiQpnPs9gx7TfsZhkGnBbo+JsBeYSkUJr+03YxIhOp3NyozQWjrVGcnNzCXIIYpVxI5dGihGJRNJsKHQIQAyID6hx+/CJ4YzdMoqeL/QAX7GsNLmMIoe4E0nbQHPRAPquzmIkMjIST0/PRh+DU4+a5GSCBgVqT9kLu2RZ+EshxYhEImk2OAWvXsIy4ojOU8fedrtZdOEDbVnGr1kNPjZJ88bmogHQd/UlLy+P/Px8oPFdNDYqB7G6+7kTEC9+x0VHizEVyeyv6pBiRCKRNBuKDjln0tSGt99+m7lz53K4/LC2LOs3KUbaGiWnHCwj3fRNGi9io7IYAexxIyrk75ZB1tUhxYhEImkWqKpK4WHhpvHq4IVXuFeN2//jH//g/vvvF5U2zacpV8sBKDtU3ujjlTQvik86Wkb0TdaTxpGqxIhT3IhsmlctUoxIJJJmQXlGOcZ8YcauyUVjNBq54447eOmll7Rlw0YOI8l0AgBdro7ys7LBXlvCZhlx83XDO8Kr+VhGhgVry2QH3+qRYkQikTQLCp1cNNUHr5aUlDBz5kwWL14MiJTKt99+m9dee41jpmPadvLG33awGC2UWet46Lv6ougUl4iR9u3bO9UaAfCJ9NaK9eXvLkA1y+JnVSHFiEQiaRY4loGvzjKSnZ3N+PHjWbVqFQBeXl58++233HvvvcTExHDcQYxIk3jboTSlTHvIV86kgcYvBW+jqlojiqIQPDwIAHOxmYrTsg5OVUgxIpFImgWFNQSvnj59mtGjR7Nr1y4AAgMDWbt2LTfccAMA7dq1I8U9WdteWkbaDk7Bq11FjrdNjISGhhIYGNhkY7G5akpLS8nJEV2mHZvmlR2QlViros5iZMGCBYwaNYqxY8cyduxYHnjgAQBWrlzJ8OHDteVjx47l7Nmz2n5Hjhxh3rx5jB49mgULFpCVJaPdJRKJHVtar5veDd8YX6d1e/fuZdSoUSQlJQHQqVMntmzZwhVXXKFtoygKIdEhZJgzAMjfX4C5QhY/aws4pfV201NeXk5GhvgdNJWLxsYlM2qAsoNSjFTFZVlGnnrqKbZs2cKWLVt46623tOWDBw/Wlm/ZsoUOHToAYDAYeOyxx5g3bx4bN26kf//+PP300w1zBRKJpMVjLDBSliZu0gHx/ig6RVu3fv16xo0bx7lz5wDo1asX27dvp2/fvhcdJyYmhmOmowCoBtWpiJqk9eJc8MyX5ORkrTFdcxAjAf0C0HmJx23pQZnpVRXuTXGSPXv24OHhwaxZswC48847mThxIhkZGXTq1Omi7Q0GAwaDs1/N3d29wSvoWSwWp/9bMq3pWkBeT3OmMa6l4KC9/oJ/H3/t2F999RV33HEHRqMRgFGjRrF8+XJCQkKqPH90dDTHNhxjktdkAPJ2XiBw8KUrucrvpvlS2+spdhAjPrE+JG1J0uZjY2Mvub+qQmIamMzg6w2+XuJ/Hy/wuIwnZJcuXbTp5ORkcW53CBwQwIWEfIzpRsrOlePT3rvuB29G1OW3ptPVbPe4LDHy2muv8dprrxEXF8dDDz1E9+7dATh06BATJ04kJCSEG2+8kdmzZwPC12vbBsDb25vOnTtz+vTpKsXIJ598wqJFi5yWzZkzh7lz517OcGvkzJkzjXJcV9CargXk9TRnGvJa8rbYS2UbOhpITU3lo48+ckrdnTx5Mm+++SZFRUUUFVVd7j0oKMgpiDVzcya6GbUbg/xumi81XU/RCWEBcwtxIzM/k927d2vrAgMDSU1NrXbfJz8O4Ztfqw6YdndT8fZU8fG0WP+3znup6L0tzL2imCmDnd0utmwaEM9E7dyxQIKYTN6ajH6IsyuypVKb31pMTEyN29RZjDzwwAPExsai0+n45ptveOCBB1i2bBmDBg3im2++oUOHDhw9epRHHnmE4OBgJk6cSFlZGXq93uk4er2e0tLSKs8xf/58br75ZueBNpJl5MyZM0RGRtZKuTVnWtO1gLye5kxjXEthRrE2HTMuis9Xfu4kRO666y7efvtt3N0vfcsaMGAA/zb/m1K1BF9Fj+GokS5duqAoSrX7yO+m+VKb6zEVmTiWI+rLBHT3JyoqigsX7OJ22LBhWoZLZc6ch2+3VH9+k1mhuEyhuKzqc2874suZZRDqEB/r5WUv1pebm2s/d38dF5bkA+BX4keXqMjqT9wCaOjfWp3FSHx8vDZ92223sWLFCg4dOsSIESOctpk3bx6bNm1i4sSJ+Pj4UFJS4nSckpISfH2rVoaenp5N0tTIhk6naxV/uNC6rgXk9TRnGvJabMGriptCQK8A3pprj0V77rnnePrppy8pKGx07doVCxYSTYkM9BhExdkKDJkGfCJ9atxXfjfNl0tdT1my3TKh76ZHp9M5VV/t3r17tfsuXqVisYjYkjH9ILIdlJaLf2UG+3Rphf3/sgrh2gGoMMKKbQp3Xm3/bUZERODt7U15eTmpqanauf262V/Iy1LKW83301C/tXrHjFQ3CEVRtACi2NhYli1bpq0rLy8nPT29yXK/JRJJ88VisFCcKCwj+u56jIpRe5gMHjyYZ555ptbHspmDj5mOMtBjECDqjdRGjEhaJpUzacCe1uvj40NERESV+5nNKv/7WTyjdDr46mmFyPY1C15VVfn9CIz6i9j3m42qkxix1RpJTEx0qjWid8gQK02u2ivQlqmTnCkqKuL333/HYDBgNBr58ssvKSwsJD4+nu3bt2umsePHj/PNN99oaXeDBw+moqKC5cuXYzAY+Pjjj+nVq1eV8SISiaRtUZxYjGoUN/aAvv4kJSVpLzK9e/eu07GCgoIICAhwLn62M7/BxippfhRXyqSxWCwkJ4t6M7GxsdVa1NbuEm4agOnDqZUQASE2RvSBGKvG2bgPsvOdq6pWVWvEu7M3irs4hxQjF1MnMWIymXjnnXeYNGkSU6dOZcuWLbz55pv4+fmRkJDA3LlzGTNmDH//+9+59dZbmTp1KiDcLgsXLmTJkiWMHz+effv28cILLzTKBUkkkpaFU7GzvgEcP35cm+/Zs2edjqUoirUSq/0YsvhZ68a54JmejIwMLRvzUmm9i1baBcRdV9dOiNhQFIW548W02Qzf/+q8vqr0Xp27Dp8uwkJXmlyqCW6JoE5umuDgYD7//PMq1z300EM89NBD1e7bp08fvv7667qNTiKRtHpsnXpB1Bg5/tvlixEQrpoDBw6Qakolyj2KwkNFmEpMuOubpJKBpIkpsXXr1YFvjC97dtgzaaoTI1k5Kiu3i+mIULhqZN3Pe+MEhf98JQTF0k0qd8+0C5rKYmTo0KEA6GN9KT1dirnUQsU5A94dLt2Zui3ROiJoJBJJi8UWvApWMVIPywjYHwRa8TOzSsF+WfysNaKqqlbwzLeLD25eulo1yFu8WtQVAZg/A9zd62YZARjQHbpZIw0274dzeXZLR1WWEQDfGHvsknTVOCPFiEQicRmqqmpuGu+O3niGempiRKfTXVb1TMcgVhvSVdM6qThnwFwiVEVtG+RZLCof/WQXDndeVXchAlZXzQTbMeE7B1dN9WLEHsRacto5w7StI8WIRCJxGWVpZZgKTYAIXrVYLJoYiY2NdarZUFtsYkR28G39OMWLdHNukAdVW0Y27YPTmWJ68hCI7Xh5YgSEq8bGNxtrYxlxyKg5LS0jjkgxIpFIXEblTr0ZGRlaMcTLcdGA/UGQYcmgwqMCEJYRGTDY+nDuSeNsGdHpdE6iwIZT4Oo1ly9EAPrGQg9r9fctByEzRxy7ffv2WiVWJzES62AZkW4aJ6QYkUgkLsMpXqSemTQ2bJYRFZUzXqJUtSHXKN9EWyFONUYqiZHIyMiLimdm56t8/5uYDg+CmWPqd35FUbjR6qpRVVi22b7cVnnVVmsEwCfSG9zENjJmxBkpRiQSicsoPOSQSdO3/sGrAH5+foSFhQFwuOKQtly6alofTm6a7r7k5eWRn58PVO2i+Ww1GIVXkNumgadH/SwjAHPH24+xdNPFrhrHWiM6Dx0eER5i+WmZ3uuIFCMSicRl2Cwj7v7u+HTxaRAxAnbryM68BG2ZFCOtj5IkIUZ0Pjq8I7wvGS+iqiqLfmo4F42NPjEKfax94LYdgjPnxDmqixvxjBRixFRsxpDt3J2+LSPFiEQicQmGPAPl6eUA+Mf7o+iUBhMjtgdBovG4dpeTGTWtC4vRQmmq6Eujj/VF0SmXFCNbD0JimpgeNwDiIhtGjICzdWSZNaumWjHS2UOblnEjdqQYkUgkLqFyfRFAEyNhYWGEhoZe9rFtlpFyysEaYFh0tBijNXNH0vIpSytDNQkrROWeNHCxGPmwHhVXa8JWjRXsWTXVipEu9jgWGcdkR4oRiUTiEpyDV/0pLCwkM1PkXNbHKgJ2MQJQ3MF6HhXy9+TX67iS5kNxFZk0x47Z07m7deumTV8oUrXg0mB/uGFcw46lZ5RCP6v2STgKKVlqtWLEI9JuGZFBrHakGJFIJC6hyCmtN4DExERtvr5ixPFBkOGbrk3n786v13ElzQfHTBo/q2Vk7969AHh4eDg1WfxiLZRbwzNunQreXg1rGQHnmiPLNl/KTWO3jJRIy4iGFCMSicQl2HrSKO4Kfj39GixeBJwtI0cNR7Rp2cG39VBSqVtvSUmJJmj79u2rpfWqqtqgtUWqw8lVs0mtttaIZycP7ckrxYgdKUYkEkmTYy43U3xCPEz8euhx89I1qBix1XgAOHT2EJ7txIMpf3cBqkWmU7YGnKuv6jl48CAWiwWAgQMHaut2HoNDp8X0yD4i+6Ux6NZZYVCcmN59HJKzqLLWiOKh4BMpu/dWRooRiUTS5BQfL9aCDwPiAwAaVIx4e3sTEREBQEpqCsFDggAwFZo0ESRp2di69XqGeeIR5KG5aAAGDRqkTX+4ovGtIjacao5srLrWCNgb5pkKTRjzjI06ppaCFCMSiaTJqRy8CnYx4unpWWUZ77pic9WcO3cO/UC9tlym+LZ8TEUmKs6KUv/6rqLEelVipLBE5euNYlmA3tmV0hg4Hn/ppuqDWJ0b5klXDUgxIpFIXEDlnjQmk4mkpCQA4uLicHNzq/c5HONGSjvarSEybqTl4/gAt2XS2MSITqejX79+ACxZD6WilA03Twa9T+NaRmI6Kgy1GvX2JYFfuN1dJBvmXRopRiQSSZNjC14FUfAsOTkZo1GYq+vrorHh+FZ61jsLxVr6W1ZibflU7tZbUVHBkSMiULlXr174+oqHvVPF1QauLVIdjlk1Z8pGaNOOYkQvG+ZdhBQjEomkSVEtKkVWN413Z288gz0bNF7EhqNlJDkjmYC+IjalJKkEQ54sw92Sqdyt98iRI5qYtblo9iaq7LFmiw/uAQPjmkaMzHFw1exJsxdec7aM+GjTstaIQIoRiUTSpJSmlmEqNgMQ2Lfhg1dtOImR5GSChwVp8/m7CxrkHBLX4NStt5veKV7ElknjaBVZ0MiBq450aa8wso+YPnXWF3zE79lRjPh08QHrkBytPG0ZKUYkEkmT4tip179S8Co0ohgZGqTNS1dNy0azjCgi/qJy8GpJmcqX68S83gdumtS043PMqnGPuAlwFiNu3m54dxI1SKRlRCDFiEQiaVIqB6+Csxjp0aNHg5ync+fO6HTiFpeSkkKQgxjJl0GsLRZVVTUx4tPFBzcvnZMYGTBgAEs3QZH1GT9vAvj7Np1lBJxdNbr28wDnWiNgjxsx5pswXJBuQylGJBJJk+Kc1huAqqpaT5HOnTvj5+fXIOfx8PAgMjISEJYRn07e2tto/r4CLCZLg5xH0rQYzhs0N5++qy8mk4kDBw4Aoh9NYGCgc1O8JnTR2OgUrjCmn3W87t3At8/FtUZiZUaNI1KMSCSSJqXI6qZxD3DHJ9KbnJwcLly4ADSci8aGzVVz4cIFCgoKNFeNucRM0dHiBj2XpGlwjLHw66YnMTGR8nKRvzto0CAOn1b53doBoG8sDOvlilE6u2oInwNUyqiRtUackGJEIpE0GRU5BsqzRLGqgL7+KIrSKPEiNioXnQpyDGKVcSMtEsduvb5dLw5edexDs+AaBUVpessIwOxxoJ06bC5wiVojMm5EihGJRNJ0FFVy0UDjBK/auCiIdUigNi+Ln7VMnDJpujoHr8b3G8Lna8W0tyfcPKWpR2cnIkzhiv7WGd8eoO9Hamqqtt5WrA2kZQSkGJFIJE2IYyZNVcGrjS1GAvoGoPMWt7383fkNei5J0+BYY8Svu7Nl5EzZUC5Y9e6c8RDs7xqriA3HAmiEzXG2jETJWiOOSDEikUiajEv1pIHGd9PoPHUEDhAWmdKUMirOVTTo+SSNj61Bns5bh2cHT/bt2wdAZGQkP+/017b701WuFSIA118BOsXqNgqfy+nTydo6Nx83vDuKgGppGZFiRCKRNCE2y4jioeAXJ7JmEhNFmUw/Pz86duzYoOerbBkBZL2RFozFZKE0RTy49bG+JKckU1QkBG7fgaNYt1ts1ykcLZvFlbQPURg/yCqKfLqRmOHrtN43VlhHjHlGjPltu3uvFCMSiaRJMJeZKU4SJnb/nn7oPHWUl5drIqFHjx4NHmzYsWNHPDw8ALsYcao3Il01LYqytDJUk7A0VK68qu80G4P1eT5rDOh0rreMgHNWTXr5yEq1RhziRtq4q0aKEYlE0iQUHSsGa2kPf2u8yMmTJ7FYxMKGdtGA6OAaFRUF2ItOOVlGZBBri8LmogERAOooRrKM9qZ011/RPIQICFeNgqiLYgq+jpycXG2dzKixI8WIRCJpEhw79TZFJo0Nm6umuLiY3NxcvNp5aY3KCvYXYjHI4mctBaduvY6ZNIoX+1I7ABASgD2LpRkQFqTQSW/t2Ocdw+Y9Dtcga41oSDEikUiaBMcy8IFNELxqo6q4kaAhQQBYKixO45I0bxwtI77d7GIkMPoGSsrF42zmGHB3bz6WEYDhXe0pvWv2BGjTsgqrHSlGJBJJk+D40PdvgrReG1UGscriZy0Sx7TefO98cnOFy8Mv+o/a8uvGNi8hAjB9aAVYREDLrpRorJ5JfKNleq8NKUYkEkmjo5pVio4KMeIT5YNHgAgqtYkRnU5Ht27dGuXcldN7wTmjJn9XQaOcV9Lw2Nw0HiEeHDp90LrUjTxGA6JD7+QhLhrcJYjv2QnyRRvhIkMQ2w6L5e56d7w6eAEygFWKEYlE0uiUJJdiLhFBfLZiZ6qqamIkJiYGb2/vRjl3VZYRv15+uOndAJlR01IwFZu0VgJ+jpk0gWMpM4mslKtGgLdX87OMREdHw/kl2vzna+zrbN17DdkGjIWmJh5Z80GKEYlE0ugUVVHsLCMjg5IS8aZb2UWzcpvKwDstTH7YwhPvW1i2WSU5U3VKi6wtVYkRnbuOoEGiNHx5ZgXGs227xkNLwDHA09cxeDXsOm15c8qicaRdu3Z4lawGk/g7+HYzlFWI37LMqBG4u3oAEomk9eNcBv7SmTRms8qChSpn88T8+t0A4sYdEgBDeqgM7gFDeigM7gFd2nPJ+iTh4eH4+vpSWlrqVI47eFgQuVvEScoOlsPwel6kpFFxzqTRs/eHvYCCEnY9KuDpAdNHVLu7S1EUhZgu7Tie8x10uJ3CElixFW6caLeMgBAjgf0DLnGk1osUIxKJpNGpSxn4jXvRhEhl8gph7S7xzyZQwgJhcA+VIT1gSE+FyUNA72MXJ4qiEB0dzdGjR0lJScFisaDT6ZyKn5UeKqv/RUoaFccGeaZwI5mZmeA/DNVTVO2dPAQC9M3TMgLCVXN8x2fQ4XYAPlujcuNExSmjpi2n90oxIpFIGh2bZcQj2APvTiI2pDox8tV6uyvmvb8pdAyF3YkqexJhdyKcv+B87JwCWLNT/AOVXlFw4BPwcEjvjImJ4ejRo1RUVHD27Fk6duxIkEMH37IDUow0d0qS7JaR5DJrj5fQWdqy5uqisdG9e3dWr34bylPBO4o1u+Bsrlqp1kjJJY7QupExIxKJpFGpOFdBxTkDIIJXbS6VqsRIWYXKd7+KZQF6uH0aXDtG4fk7dfz8io6zPyqcWabww78UnroVpg0XlhFHjqXC6gTnZVVl1HgGe+IXJwIfyxMrMJeZG+iKJY2B5qZRYH+maI5H2PUA6HRw7WgXDayWjB07FlDh/BcAmM2wZAP4RsuYEZBiRNLI5OXlMX36dHr27ElSUpKrhyNxAYVHLnbRgF2MhIaGEhYWBsBP26HIej++YdzFmRGKotC5ncKssQov/EnHqoU6zq9QSP1W4ZU/27f9dLVzoGtVQazg0KfGDPl7ZYpvc0VVVc1N4xPpw55De8C3D/h0B0TF1bCg5m0ZGTdunJiwihEQv1N3f3e82nsCUoxIJI1CeXk5M2fOZPXq1SQmJvL++++7ekgSF1B48OLg1aKiIjIyMoDqXTQ3T67dw0VRFLq0V3hoDrQPEctWbIPcAvuxqhMjISODtem8rZX8P5JmgyHbgMma9qqVgW8BWTSOtGvXjj59+kDZCSgSprsDJ+HASVXLqKk4Z8BU3DbTe6UYkTQKFouF22+/na1bt2rLTp8+7cIRSVxF3g77Qz5wgBAjiYmJ2jKbGLlQpPLL72JZRChcOaBu53F3V7hlspg2muDrDfZ1VblpAELHhtjHubWaqFmJy3EMXnXv7CYEpUO8yKwxLhjUZXDllVeKiXOfacs+X6PK9F4uQ4wsWLCAUaNGMXbsWMaOHcsDDzygrVu8eDGTJk1iwoQJvPnmm041AY4cOcK8efMYPXo0CxYsICsrq2GuQNIseeKJJ/jmm2+clqWmplaztaS1YjFaNDHi1d4TvTVGo6p4kWWb0VrAz5sIbm51f9u9bZp9n8Wra7aM+HT20Upy5+8pwFwq40aaI6UOYiTHIwe8Y8BvIADDekFk++ZvGQEHMZL9DTpFWEC+XAc+UQ5BrFKM1J6nnnqKLVu2sGXLFt566y0Atm7dyrfffsvixYtZunQp27dvZ/ny5QAYDAYee+wx5s2bx8aNG+nfvz9PP/10w12FpFnxzjvvsHDhQkCU+fb1FX9oUoy0PQr2F2qVV0PGhFwyePVyXDSV6dtVYVCcmN59HI6miGMGBwcTGCgiXR3FiG1cAKpR5cLO/Ms6r6Rxcawxcrr0tJNVpDn2oqmOK664QkyYLhBg/A0QaeynkA3zGsxN88svv3DdddfRuXNnwsLCuOWWW/jll18A2LNnDx4eHsyaNQsvLy/uvPNOjh07pvmMJa2HFStWOFnL3nnnHYYMEc0i8vLyKCqSHVLbEraiYgBhY0O16cpiJP28yq/7xXxcJJqguBwcrSOfrrrYOpKWlobJZPfLh4yxx43kSldNs8TRTbM3cw+EOsSLjHPFiC6PsLAwTXwXJL2tLf8pxd4wr61aRi6rzshrr73Ga6+9RlxcHA899BDdu3cnOTmZqVOnatt069aNU6dOASJWoHv37to6b29vOnfuzOnTp+nUqdNFxzcYDBgMBueBurvj6el5OcOtFou1daLt/5ZMc7iWXbt2MW/ePG0Mjz32GAsWLGDbtm3aNsnJycTHx9d4rOZwPQ1Ja7qeulxL7pZcbTp4VJC2j02MeHp60qVLF95YpmLz6v5hksieuJzS7wDzJsAj74q4kc/Xwgt3WnB3h6ioKPbv34/ZbObMmTNERUVp49LGuzUXi6XrZZ23OdCafmdgvw6bZUTnpWNz4jEIHQlA7ygL3TrpsFgu77fS1FgsFkaMGMHx48dRc3/B37uConIvvj7qw1XWbUpOlbSI768uvzWdrma7R53FyAMPPEBsbCw6nY5vvvmGBx54gGXLllFaWoper9e20+v1lJWJQkJlZWVO62zrS0urVoCffPIJixYtclo2Z84c5s6dW9fh1oozZ840ynFdgauuJS0tjRtuuEH7zq+++moWLFhAamoqQUFB2na7du3C39+/mqNcTGv6bqB1XU9N12IxWMhLEPEi7h3cOa+cR0lVMJlMWpp3dHQ0GRkZLP4lAhAvG1f0yiA1tX4ZBeP7hbN2ry9ZufDVqnOM61dOaKjdMvP77787be8Z5YEh1Uj+3gKSjyWj823Zsf2t6XemmlStMqlbRx1JmX0gTHw/E/oXkpraslKyR4wYweLFi0E10tlrG8fKJ5Bn9sDk5457sYmik0UtyqVdm9+aY8xWddRZjDi+1d52222sWLGCQ4cO4evrqzW9AigpKcHHR5iefHx8nNbZ1ttiCSozf/58br75ZueBNpJl5MyZM0RGRtZKuTVnXHktubm53H333eTmirfgK664gqVLl+LlJVpj9+3bV9u2vLxceyO9FK3pu4HWdT21vZa8bXmo1mZg7caFaxktJ0+e1Cyf8fHxFFuiOJYm9hnWC8YNu9haWlfuvg7WWvuord7Xnluvgf79+2vrS0tLtd+hxWIha8g5DKkFYAa/TH/CJ4bVewyuoDX9zkBcz6kdp8CqTdX2KlTYXTS3XxNEVFSQawZ3GVgsFoYNG4aiKKiqijnzY9BPAOCst57OxQWYss10CuuMu7WrdHOloX9r9S4HbxtETEwMJ0+e1Aq7nDp1iq5dhbkzNjaWZcuWafuUl5eTnp5ObGxslcf09PRscOFxKXQ6Xav4w4Wmv5by8nKuu+46LVWzV69e/Pjjj5oQBZy+57S0tDqNrzV9N9C6rqema8nblq9Nh10Rqm174sQJbXmvXr34eqN9n5snK+h09Q9IvHqUSligSk4B/LgVCksUp7ezyr9D/RBf8r8Tb9gXtl2g/eR29R6DK2lNvzNDqr2jcpYuDwJnAhCqL2RQXOAlmyQ2R4KCgujfvz/79+/nxL6viZv3KSfS3Thu9KUz4jdYfqacgN61tyC7kob6rdXpCEVFRfz+++8YDAaMRiNffvklhYWFxMfHM2PGDL7//nvS09PJzc3lyy+/ZMaMGQAMHjyYiooKli9fjsFg4OOPP6ZXr15VxotIWg4Wi4Vbb71Viwlp3749v/zyC8HBwU7bOVpCWpL5UVI/HINBHet5OAav9ujRk6/WiWmdDm6c0DDn9vRQ+MMkMV1hgKWbqk/vBfAdbBfPMoi1eWFItccP7swzgs4DgCmDS1ucELGhpfiqKsOixYtclqdDRs2pthfEWicxYjKZeOedd5g0aRJTp05ly5YtvPnmm/j5+TFmzBhmz57NbbfdxuzZsxkxYgQzZwoF6+npycKFC1myZAnjx49n3759vPDCC41yQZKm4/HHH+fbb78FwNfXl59//tmpuJSNyMhIbVqKkbaBqcRE/u58AHxjffHpZH/YO4oRg/dgUs6K6UmDoX1Iwz1cbp/uUHNkler026wsRtxD3PHrKeLaCg4UYiw0ImkeVKTZxUhCid1idefMkKo2bxFoYgTwuPAtigJZXm07o6ZObprg4GA+//zzatfPnz+f+fPnV7muT58+fP3113UbnaTZ8vbbb/Pf//4XEGa6pUuXMnjw4Cq39fb2pkOHDpw9e1aKkTbChZ35qEYRL+JoFQFnMfL7absL73Jri1THgO7QNxYOnYYdRyDzgp6wsDBycnKcqrDaCBkdQvHxErCIqrHtp7ZsV01rwZBmF4YnvIcC4G7J4cpB4a4aUr0ZO3YsOp0Oi8XCnu0/MmHCP0nd2rarsLYOp6KkSVm+fDl//etftfn33nuPq6666hJ72F01WVlZVFRUNOr4JK7Hsb5I6JiqxUjHTlH8uFWY3L094borGnYMiqI41Rz5bLWquWoyMjIu+h061huRpeGbDzY3jclXR5GnaCcQG3josir0NheCgoIYOFBUkD148CDXjSrmrGfbLnwmxYikTiQkJHDTTTdpueVPPvkkCxYsqHE/x7iR1pR2KKma6sRITk6OlnUV1v0mcqxZmdeOBn/fhn+43DwZ3KxJCZ+tgehoYYlRVZW0tDSnbUNGBYN1CDJupHlgKjFjOidSabI8PbTlE/q3rHTeqhg/frw2HWjaiOrnQYGbuMbiUyXV7dZqkWJEUmuys7O55pprtFoif/jDH3jxxRdrta8MYm07GAuNFOwXDwu/nn54tfPS1jm6aMr8ZmnTDe2isdEhVGHaMDF95jzoQuwRspVdNZ4hnvj3ERkMhYeKMFxwLrwoaXoc3RVJiHL+mPKZ08KzncA5biRh+wZuuMIexFqRVYG5rG31SZJiRFJrli5dSnZ2NgDjxo3j448/rnVKlxQjbYe8HRfAWpSxOhcNOj2pxQMACPaHacMbbzyOgazJpfba4ZWDWMFhvCrkbb9w0XpJ01Jy0m4hOGN10ZD7E0MG93PRiBqOsWPH4mY1223atIlbpylkOrpqUstcNTSXIMWIpNbs3LlTm37llVe0oma1QYqRtkNeNSm94CBGQq/FYBYm6TlXilTcxuKaUULwABzIiAU3P6AGMYKMG2kOOKa4Zlgf1BGevxMQEOCqITUYAQEBWtD/kSNH6N0xm5Jge0ZN5sG25aqRYkRSaxISEgCRqu1YzbI2SDHSdtDiRRQIGe1cc0YTI+3+oC1rLBeNDS9PhXkTxXSF0Q3CZgMXu2nAGjdivSvKuBHXU+IQyJnu5QvmUkb1KnbhiBoWR1fN1q2/0n2Y3TLy+7q2FcQqxYikVuTn52tVVgcMGFAnqwg4i5GqHgKS1oEhz0DhIdGZOaCvP57BzpWUjx8/Dh5hEDwFgMh2MKYJLO63O2TV0P5WoGrLiEegBwF9xVt30dFiKnJk3IgrcXTTZHn6woU1DBtSc6PNloJjEOumTZsYP8MuRpL3SsuIRHIRu3fv1qaHD6+7gz8gIEBrmCctI62XvG32OIvQsaFO68rLy4UACJsDiihxdNNEGqT8e00M7QU9u1hnAseBd0yVYgScXUt526V1xFWoqkqJ1U1z3sMbg84Ncn5g0KBBLh5ZwzF69GgtbmTz5s30GWNvKOuZXcbh0y2jG3FDIMWIpFY4xosMGzbsso5hs46kp6djNretSPG2Qu6WXG26cvDqyZMnRUp4+E3aspunNE2tiMo1R2j3R86fP39RA09wHrdjirKkaTHkGjEViLTedE9fsBgh7yetPkdrwN/fn6FDRSG3Y8eOkVeRh9lXCPUIQymfrZFiRCJxQosXwZMh/Ydc1jFs5bhNJhOZmZkNNTRJM8IWZ6G4KQSPcI4XSUxMBK9oCBwNQHwM9OvadIWr/jhV9L8BoN0fAaVKK13IiGAUa0EtGcTqOhxdNBleesjfSJdOgYSGhl5ir5aHo6tm8+bNBHQTrppwYzlfrzZjNrcNQSLFiKRGVFUlISGBMF0YnwV/QeqMM5x+OxmLyVKn48gg1tZNxbkKihPFAyRwQAAeAc7dJo4fPw7t7FaRPzRy4GplOoUrTLJ1LPCJhYAxVbpq3P3dCRwg4kaKT5RQcU5WDHZEVVV+2aGycptKaXnjPShLHAp/ZXj6Qu6PrcpFY8MxiHXz5s0EdBdiRAdwtowNe1wyrCZHihFJjaSnp3Pu3DnGe07EXwnAXGrh+D9PsGNqAoVHimp9HClGWje526pP6QU4duy4UxaNratuU+JYc4T2t1UbNxLi6KrZJq0jjixeBVc9rnLtkyph16jc8JSFL9eqFBQ3rDApOemY1usDuctbpRgZPXo0Hh4izX3Tpk3oY+1xIxGGsjbjqpFiRFIjtniRvu59nZYX7C9k24QdnPj3ScwVNVtJpBhp3TiVgB97sSl93wkz+PYGYFS8hagOTd9bZNZY0HuJOATCZpN0KqPK7RzFlEzxdebjX+wPx7IK+P43uOVFlfBrVWY8auGjn1TOX6jfA1RVVTI22j/3DMNRMJ5rlWJEr9drcXgnTpzAEGy3xHWsKOX736CotHEFycFTDS8m64oUI5IaSUhIQIeOXh7iQeIR4oFfD6HeVZPKyYWn2DZ+Oxd25V/yOFKMtG5swauKh0LwsCCndaqqklRgD3y+ZYpbUw5Nw8dL4eoR1sqW7v5sP9Ghyu2ChwWhWAuxSTFi5/wFlW2HxHRIALRzCAsymmBVAtz1ikrEdSpXPmDhrWUqZ85d+iFnMKoknVFZnaDy9ncqD/2fhftuyabisGgpkOql51z+CoBWKUbA2VVzJOeINt3RUEpZBSzb3HjnVlWVec+K7+yOf1swmlwjSqQYkdTIzp076ebWDV9F+DLDrgxl9KZRdHskFsVd3LCLE0vYMT2Bo/84jqnEVOVxpBhpvZSll1GaLB7ywUODcPN1Fhvp6ZkYg64HQMHEnCubeoR2/nyD3Qx+/ELVmWHueneCBoteKKWnSinLKG+SsTV3ftoOqvVZteAayPxe4de3FP46W9SMsWGxwK/74a9vqXSZozL8bgv/+VLlu80qr3ylcvdCC5MeshAz14LPZJW4m1WmP6py/5sqbyxVidt8UjvWl+FdIfdH2rdvT0RERNNecBPhGMT6W+Kv2nSEQbiqFq9qPIGw7RAcSxVWrpPp4OHumm7IUoxILonZbGb37t3Ee9grU4WMCsHNS0fck90ZvXEkAf2tpZlVSHk/lS1jt5Pza+5FxwoLC8PHR5Q7lmKkdeHoogkZc3G8yNK1Z8GrMwAxgScIC3Jd+/cr+rvhbkwBoNhtKGnnqt5Oloa/mOVb7Q/FmWMU3NwUrhig8MYDOlK/Vdj5gcITN0P3zs777TwGT3ygMvsZlcffV/lwJWzYAylnhXBxZERRNt3LRSxaup+e7dkvQ0VKq7WKAIwcORJPT1EgcO32tbhbg7+jLELg/3YAjqY0jiBZtNJ+3AXXuu7vUooRySU5duwYJSUlTvEioaPsttmAPv6MWjucns/FofMWP6ey1DJ2Xr+bg389jLHAqG2rKIpmHUlNTUVV20ZgVlvA0ZURVkXw6vdb7BV7J/St5unfRCiKQmfPjdYZHR+tqNrq4VRvRIoRSspU1u4S0x1CYFgv5/WKojC0l8LLd+tI/FLh8KcKz9+pMKB79ccM9INBcTBnPDx5C3z0CDzjdkpbH35bDmrW/wGt10UD4OvrqxWTPHnqJB6dhRgJKS/D3arW3v2h4e+XF4pUlm4S00F+cMO4S2/fmLjXvImkLWOLF+njLkowe4Z7oo/TO22jc9cRe18M7We049CDR7QqnOlfZJC9Loc+C3vR4ar2gHDVHD9+nLKyMnJycggPD2/aC5I0OKqqapYRnY+OwMFBTusrDCp7UqPFjLmYORO9m3aAVTC0y3FSrMa5z9fCHZMv3iZoaBA6Lx2WCosUI8C63VBurY5/7ehLV85VFIU+MdAnBp6+TeFUhspP26GoFGI7QrdO0LWTiDtRFPtxMn/IYn+K6D0TOCiQVeYt2rrWLEZAuGq2bBHXm++VjxfeKCpE68o4iZ7P1sDLd6v4+zac9eLLtfbv9NapIqbKVUjLiOSS7Ny5kxi3WPQ6IUBCRgY73Twc0cfqGf7jUOJf7Y27n4gZqDhXwd5b97PvzgOYy80ybqQVUppcSrk1piJ4WDBuXs63lVUJUGG29tzI+ZGB/eKaeogX0TcuEPKFdSQt25u9Jy/uteTm7UbQUBE3UpZWRmla22rpXpkftzi7aOpC104Kf52j8NRtCn+YrDCst0JooOJ0L7GYLCT9224Vift7N/bt36fNt3Yx4hjEerrktDZ9Uy8RN1JUCp+vabjzqarKhw4umruucZ0QASlGJDWwc+dOJxdNiIOLpioUnUKX2yMZu30M4VPsVo+sH89y+q1k2TCvFeJoNaiqvshX6+03PL/ynwgLC2uScV2K6OhoOPeZNv/9Vn2V28nS8AKTSeWnHWLazwcmNIIuyPw2S6u6GjwymLArQ9m3T4iR4OBgp3tHa2TkyJFaA9LdZ3Zpy6d2sYvgt79XG8y9vfMYHLJqnpF9ID5WihFJM6W0tJRDhw7R1zF4dfTFD5uq8OnkzZCvBtL/g77aryzz+7PSMtIKcaovUil4tahUVOoEwJhNfOez1VrWmpKYmBjI+R7MwiXw0049ZVUUWg0dY6+X0paDWLcdhlyRacu04eDdwOZ8i9FC0kIHq8iT3Th79ixZWVkADBw4sFn8bhoTb29vRo4cCcCBcwe05UGFZVpn62OpsHlfVXvXnUXNyCoCUoxILsG+ffuwmC1avIhHsAf+Pf1qvb+iKHSa3ZEQa4+SkqQSOusitfVSjLR8VFXVLCPufm4EDgxwWv/jFig3WG902d/Su+clohmbkJiYGLCUQM53ABSV6liVcPF2gYMC0fmI22Tu1rw2G3TtmEUzq44umtqQ/lUGZanCAhA6LpTQ0SGaVQRoVc3xLoXNVZNltvfuKjldwr3X2T/zdxogkLWwRGXJBjEdoIe54y+9fVMgxYikWhISEoh2i8Zf5w9Y40Uuo917+6vba9N+x+xiRoqRlk9xYgmG8yICLnhUCDp351vKEgcXDdlf07Nnz6YcXrVERESIVMrzS7Rl3266eDs3Lx3Bw4SYLs8spzS59OKNWjmqqvKjNY7UzQ1mjGzY45vLzZz8rz1GIu7JbgDs3btXW9ZWxIit3ki+mo/RTWQiliaXcv0V0N5qdPxxK6Sfr58gWbIeSq1JZDdPBr2PtIxImjE7d+4k3t3RRXPpeJHq6DDDXg2pbGs57u4iiUuKkZaPU7xIJRdNdr7K2t3WmfJUKNxOjx49mnB01aPT6YTLsGATmERNnJ92UGXjt7ae4nv4NCQLbwlXDoBg/4Z9cJ35LJ3yTPFkDJ8STvDQIPbt28dbb72lbdNWxMjw4cPx9hbZZlmq+NDL0spxR2XBNWIbsxmnwNPLwSlw9WrXCxGQYkRyCXbu3ElfD4f6IrWMF6mMT6SPVhit8EAR/Tr1B6QYaQ3YSsDDxcGryzaLGycA2d8AarPKiIiJiQHVBNnfA+JN8ZffL96urfep+XGrfbquWTQ1YS41c/J1B6vIE9347bffuPLKK8nOzgZg6NChzUbENjZeXl6MGjUKgNSyFABUs0rZmTLuvlbBzVrY+MOVooz+5bAnUWXvCTE9pCcMjJNiRNKMyc7OJjk5mXhrvIh7oDv+vf0xm1UKS+r+R9DBwVUz3s9qiszPp7CwsGEGLGlyVIuq1ZTxCHInIN7faf1XTi6aJURHR9O5c6XSnC4kOjpaTOR8qy1buuni33bggADc9OIpkNcG40acq6427LFTP07T3HwdrmnPljO/MXXqVO2+MHr0aBYtWtTqg1cdsblqMi2OcSOldApXmGX9/M/liQaFl8OiZmgVASlGJNWwc+dOurhFEagLAkS8SHoOdL5BJXC6Svg1Fkb/xcL8ly289LnKss0qB06qlJRVfaPucJXdVdO3wu76kdaRlkvRkWKMF4RfO2R0iFM8Udo5la0HrTOlR6HkIGPGNPCTrJ7ExMSIifzN+HmJ4MmftnPRb1jnodOCsCvOGShJKmnScbqSM+dU9iSK6YHdoUv7hnt4mYpMnHozWcwokBh/nFmzZlFeLlw206dPZ/Xq1QQEBFziKK0PmxhxDGItPS1ileobyFpcqvLVejGt94GbJtVjoA2MrMAqqZKL64uE8H/fq5y1WqlzCsS/7YcBnP8oOoerdO8McZEQF6nQvTMMitOj766nJKmEsLxwgpQg8tV8UlNT6du3L5KWx6XiRb7e4DBjDRIdO3ZsUwyr1mhiBDNdgw5y4Nxwyirg5x0wd4LztiFjQsjekAOI6/aLq31WWUtmxTb79KyxDfsWnfJBKsY8IWaL4gv5w6M3aetuuukmFi9erMWXtSWGDh2Kr68vmYYMbZktcPrKgdA7Go6mwNaDcPCUSr+utf9elm4SxdMAbppIg1ZzrS/SMiKpEhG8ahcJQSOC+GKtmHZzg06XqOKeng2b9sEHK+Bv76hc+6To3FkyUFhHFBSGe44ApGWkJeNYdyP0ilCndUs2OLpovgFodpYRzU0DhFo2atNVuWraahDrj43kojHmGzn9TgoAqqLy8JaHtHV/+ctf+OKLL7TGcW0NT09PRo8eTaY5S1tWYhUjiqLwl1kO1pHv62YdcQxcXdAMaos4IsWI5CJUVXUKXnXzc2OXwZ8sa6zi1SMh/TsdxWsU9n+ssPQ5hX/dpXDbNFHJLyzw4mNaLPBFid1VM8pjNCDFSEtFNankbRfxIp7hnvj1sFcwPZaisj9JTLuV7oHyU4SEhDSbtF4bdssIGLPX086aLPbzDmHOdiSgnz/u/uItPW/bhTYRN5JfpGoFtqI7QL+uDXfs5HdTMBWaAFhXvpZMi7ACPPXUU7z99tvodG370XTllVdyQc2jTBXuQ0fX4B+niiq4AF+sE99TbTh4SiXhqJju300ErzYn2vY3LqmSU6dO4ZPvS7BOvA2GjAjm8/V2FX3rVDGt91Ho301hzniFv/9RYfHfdWx/T0f2Sh25Pyn8/r7C508pdLHGrn6d4o9HR5G21t9jAL6KrxQjLZTy4+WYi0WqTOiYEKcAQ0eriDnrC0BYRZrbAyY8PBxfX9EzJz09leuvEMvLDWilz23o3HVaKwRDjoHi48VNOVSXsCoBTNZsqJljaLAgUkOugeT3xd+9UTWypOwrAF5//XVeeOGFNhWsWh22uJEUcwoApSllGPOFSytAr3DrVLFdaTl8urp2x1xUySrS3D7n5nV3kDQLKqf06ocGa5Hbwf5wVS2KHoUEKAzvrXDLFIX508UyFYWzccI64qF4MNRjmBQjLZSS3fZ+GY6pr6qqssQaIKegQvZSoPm5aEA8XG3WkYyMDOZcadHWfbPx4rfNkDbWp8axMV5DxoskvpaEuUSonLUVq8lVcli8eDEPPvhgg52jpTNkyBD0ej1JphPasoL99szDe6+3fx/v/qhisVzaOlJarmpudh8vUeisuSHFiOQiKseL7PUK1vp2zJsIXp51uzHdPNm+/XdGB1eN52gpRloopbvslUgd4yl2H4eT1ri79t5HwXgWaH7BqzZscSMGg4Gu4WfpYL2UVQlclMLeluJGKgyqVh4/JADGNFCMeXZSNqc/EBk0BtXAD5bv+f7777ntttsa5gStBA8PD8aMGeMkRvL3FWjTvaMVxlvrwJ04Axv2XPp4yzZDvtWYN3c8BPo1L6sISDEiqYKdCXbLiM5Xx+IT9voRNhdNXegeqTC8t5heeSEIXbAHAEM8hpJ3Nk9L5ZO0DCwGC6X7hWXEu6M3vrG+2jpHF40xU7hovL29m1WxM0e6drUHQiQeP8rsK8V0hQFWbnPeNiDeH48ga9zI9guoNbyNtmQ27bNnXVw9Etzd6//wOn/+PG9NfhsPVfz9r7es5atVX3HttdfW+9itkfHjxztbRhzECNQtzXfRT803cNWGFCMSJ4xGIxn7MwnTiXQZnwHBbDggfibdO6OJirryxyniD8CiKGTECuuIt+LNII9BpKWl1X/gkiYjf08BaoW4uYWOtceLmM2qltLr4aaSm/ghIEpcN9fMCEeRtHv3buaOt9+oK2fVKDqFkFHCOmK8YKToSFHTDNIFOBc6q//Dq6CggGtGXcOQwiEAVFDOrctu1RrDSS5m/PjxpFvSKVVF8GrBPucCkTPH2LMaV24XtX2q4miKveZP72gYGd9YI64fUoxInDh48CA9LfYw65PB9n40t069/KCnuRPA3VrK+FuD3VUzUrpqWhx526quL/LbAbSMq36RmWDKB5qviwaEULKxc+dORveFCGuW8uqdUFDc9lw1FovKcmsJeG9PmDqs/sdctGgRw7JG4KEIUdrulnYMmzS0/gduxQwaNAg/fz9Omk4ColFjxbkKbb27u8Ld11pf8izw/vKqxchHPzXvwFUbUoxInNi5cyfxDsGr3+UEadO3TLn844YHKUyz3vc3mEIweoo/kBEeI0g9LcVISyJ3ywVt2jF41bH8e4h5rTbdHINXbcTFxWkVPnft2oVOpzDnSrHOYHQu+gUQ0gb61Ow+bheVk4Y0TEfXbd9vY7KXuIEovgpD/zm43sds7bi7u9O/f3+STEnasvz9zq6au64GD2tduI9+ErE+jpRXqHy2Rkx7eYq04OaKFCMSJxwrr6oesK5QFA0ZNwCiI+p3U7rFGshq0uk4FSGO66fzJ39HwaV2kzQjzGVm8nfnA+AT7YNPpCh4UGFQ+e5XsY3eB7IOCxeNTqdj5MgG7jnfgOh0OoYOFW/omZmZpKenM3dC9a4a/55+eIaKmIcLCfmtst6IY6GzWQ3gosm/kM/wIyNxV8RTs+u9sXiGNE+3XXMjPj6eE+ZEbb6yq6ZDqMIN48R0dj58u9l5/x+2QK719jp7nMhybK5IMSJxIml7Eu3cRGGQ3I5BmKy1IS4ncLUy144Bf2us4y9uXbTlHgfkjamlcGFnPqr17cvRZbFmJ1ywhlDMGGbgyKFdAPTv37/Z9xaxiREQYnxkH7svfs1O56JSik4hcKAQ0sYLRsozW1/wtc1Foyhw9aj6H2/TM5vp5y46dZfpy4i9N7r+B20jxMfHXzKIFZwDWd+uVJHVqSleMw1ctSHFiESjsLAQ71R7ZsSvFvGw8fZEyzKoDz5einacbZ7hVCAqMEZkdmzVmQmtiZxNOdp0yFh7PJFjFk3v8EOaxaA5u2hsDBtmD4rYuXOnk6vGaLI/nG3497FnlxUeal1BrElnVI6miOlR8dA+pH4PsLIzZei+sfeX0T/go1WyldRMfHw85yznKLAIEVKwr+Aia9zovvbquAlHYU+iWJ90RmWTtYJuXCRc0b/Jhn1ZSDEi0di9ezfx7vZQ651u4mFz3RWi6l9DYHPVlLu5s99bRLT6mfw007+k+VKRXUHa4nQxo4PQ0UKsFpfaAx5DA6Esc7m2T0sUI8Als2oC+jqIkcOtS4w4Cq/6ZtGoqsrBBw/jYRZurQ3mdYy//8p6HbOtER8v7sc264gh10h5urM1TlGUKtN8HQNX77q6+Qau2pBiRKLhGC9i1Ckk+ghzdEO4aGyMG2A3gW8L6aUtz1xxtsHOIWkckv5zClORsGYFzQzEq70XIII8bUXxZo+DHdt/0/ZpCWIkIiKCiIgIQAhys9nM8N4QaU36WrsLLji4agLi7W6nwkPOPvyWjnO8SP2Olf5lBrmbRZBvtjmbtDFpeHl51e+gbYzQ0FAiIiJIMjsEsVbhqrl5MgRaG0kvWQ9nc1U+WSXmPdzhtulNMdr6IcWIROPwliNEuHUEINEnEKPOjQ4hMKkBA9/d3BT+MElMJ/h3wIwowZ25MqtVBgO2FoqOF5P26RkA3PRuhN9t79LrmEUze5xJsy7ExsbSsWPHph3oZdKvXz8AioqKOH78uJOrxmSGH7fYt9XH+qLzEbfO1lRr5PwFle2HxXSvKFGs8HIpyyjn2FP2wMv/K32TyddOqu8Q2yQXxY3svViM6H0Ubp8mpssNMPefKtn5Yv66sSKbsbkjxYhEo3SPvcT3IV/horl5csNUX3TkFmsBtGJ3Dw55CROuMb11F5Fq6Rx/NhGrbiT2rzG4hwm/f26ByhqhPegcDl7lu6ioEGaSlmAVsdG/v92hrrlqqsmqUdwUAnoLV01pchlGa/fZls7KbWB7H5hVj9Iwqqpy+OEjmhVtfcU69hh3M2PGjAYYZdsjPj6eE05BrFVb4/7i4KrZctC+vLkHrtq4bDFy8OBBhg4dykcffQTAypUrGT58OGPHjtX+nT1rN70fOXKEefPmMXr0aBYsWEBWVlb9Ry9pMDIyMuhc1Fmbt4mRW6c1/A+5X1dFC7jaHtJdW372p/MNfi5J/cnelEP2OhG46t3Jm+h77JlQyzbbO7vOmwjbt9uDDppzsbPKDBgwQJu2iZFhvdA6Tq/fLYSXDX+HuJGio61DRDdU1dWMrzPJXi9+L7mWXBaVfkD//v3p3LlzDXtKqiI+Pp4Lah45lmwACg4UVhnwHxepMHmI87LYjjCheXZiuIjLEiMWi4XXXnuN3r2da4MPHjyYLVu2aP86dOgAiCZUjz32GPPmzWPjxo3079+fp59+uv6jlzQYotiZMFUbFYXjvkH07yaEQ2Ngs47s8A/Xlp37+VyjnEty+ahmlWNP283tPZ7ujpuPmzbvmEVz00SFLVvs/oyWZBmJj4/XAvxsYkRRFOaKTu4XuWqc4kaaSRCrqqo89wmM+Gsn5v4TVieomM21c32WlKms2y2mI0JhaM9Lb18d5VnlHP3HcW3+7ZK3KFaLpVWkHtiDWEXciKnIRMmp0iq3dQxkBfjT1Qo6XcuwjFxWjtX3339PfHw8xcXFtdp+z549eHh4MGvWLADuvPNOJk6cSEZGBp06dbpoe4PBgMFgcB6ou3uD97ewWCxO/7dk6nst+zbtY4TbaACSvAOo0LnxxymN99ncOB4ef18lz8ObY15e9KqooOhoMUVJRei76lvVdwMt97d25ot0io+Jv/PAgQF0uK69dg1p5yz8Zu1b1CMS+saa2bZNlCwNCwuje/fuLeJ6LRYLfn5+9O7dmyNHjnDw4EFKSkrw8fFhzpXw36/Fdt9sUpk/Qzzc/Xrrtf0LDxU2i+t89hN44VMAd777Fb77VaVzONw2TeX26eItuTpW7xSxBgDXjAKouS19ZVRV5fAjRzEVCPfMqfYn2ZknWv9Onz79sj6jlvp3Ux2Xcz29eolA/yTTCUZ6isIv+Xvz8e3qc9G2M0YIa17aOXDTwa1T6/491pa6XItOV7Pdo85iJD8/nyVLlrB48WJeffVVp3WHDh1i4sSJhISEcOONNzJ79mwATp8+TffudnO8t7c3nTt35vTp01WKkU8++YRFixY5LZszZw5z586t63BrxZkzZxrluK7gcq8lc6PdbXZIH4xOURnbI53U1Ma7CQzpFsKuJH+2B3Wh1zmh+hO/PEHobfZiWq3pu4GWdT3mEgunXkzW5oPuDSTtjL2p4aIf8lFV8V1NG5zPhg07yc/PB0RfjZbWALFXr14cOXIEk8nEqlWrGDx4MGHeEBnekTPZHmzco7LvcDoh/hYsfhZQABVy9ua4vL/Sx2v8efGrkIuWp2fDvz4X/0b2LmPO2BKmDSnF29P5AbVkTSgg0jFGxp0jNbXuxdwKVhdyfrVwJbiFuPHSmRcBCAoKIiIiol6fUUv6u6kNdb2eLl26cCLTHjdy5rd0jMMMVW776l1evPZ9INePLsFQXEJqLWwGxVuL8ejsiWeUR51TgGtzLTExMTVuU2cx8u6773LTTTfh7+/vtHzQoEF88803dOjQgaNHj/LII48QHBzMxIkTKSsrQ6/XO22v1+spLa3a1DR//nxuvvlm54E2kmXkzJkzREZG1kq5NWfqcy1msxm/dH+wWt8P64OZPFRhSP/IRhipnbuvU9n1Cuzwb8edVjFSsd1I1DNRreq7gZb5Wzvx0knMeSIgpP3V7eg1U9jubdeyeq+96Nk9NwSx4adT2vzkyZOJiopq2gFfJrbrGT9+PMuWLQPEDfb6668H4KZJ8MoSMFsUdiVHsuAasd+Z2ExKT5ViOGUkslMkOnfXfK+LV8GLX9nn/35THoN6BbF4tY5Vv4PZ+j6x46gPO4768PyX4prmT4dBcWA2wyZrwKO/L9w4tT1edbzVVpyr4OSrduGquwPO/l3EDE6bNo3Y2NjLuraW+HdzKS73evr378+mM5u0efWUWu3fV1QUXD8JwAcIq/HYqlll43O/YrxgJKCvPyM3DK+VIGno76ZOYuT48eMcPXqUxx9//KJ1jhaO+Ph45s2bx6ZNm5g4cSI+Pj6UlJQ4bV9SUoKvr2/lwwDg6enZpC3HdTpdq/ihw+Vdy/Hjx+mJeNCYUTjmE8Qj0xrf1zhnvMpd/ykjy8uX057exBrKKdhTgOGsAc8O4vtvTd8NtJzrKUsvI+U98SareCj0fLaH07hPn3Vn7wnx+xjcA3pG6Xhh+3Zt/RVXXNEirtMRxw6+ommeGP+NE1ReWSIsCcs2wz0zxfLAvgGUnirFUmGh7HQ5/j39mnzMP/ymctdCu5Xjmdvg9glFREWFcMOVOjJzVD5bDR//opJkrVeXXwzv/Sj+9e8G4/pDnjVBY/pw8PGu2/emqipHnziO8YIRgIhZHVhy4Utt/dVXX13v30JL+bupLXW9nr59+7Jy5UoyzZl0dOsoKv+aQedR/8/kwp587bvzjfHFzc2thj2caajvpk5H2Lt3L6mpqcyYMYOpU6eybt06PvvsM5577rmLtlUURasbERsby8mTJ7V15eXlpKenX7ZaljQsuzbspoubUNknffxx93dnZhPEHgboFTp4iGDBHYF2h/bZX2RWjatJfCEJS7l4pY5eEIU+xvnFYeUOu6XzpolClGzdKjJpfHx8GDhwYBONtOGIj4/H29sbgISEBG35wDjoan3X2rRP1OOASmXhDzd98bMNe1TmPadic9k/cAM8c7vzNh3DFJ64RSHxS4Xf/k/htmng621ff+AkvPWdff5ysmjOLj/HOWsmnGeoB73/04uff/4ZEA+qadOm1fmYEmcqV2K1lFsoTqxdzGZNZK/P1qbDJ9VsSWks6iRGrr/+en744Qe+/PJLvvzyS6644grmzJnDww8/zPbt27lwQbQWP378ON988w1XXHEFILJsKioqWL58OQaDgY8//phevXpVGS8iaXrS1tp9uYd8g5lzJfh6N00E9oCORwDYHtBOW3buJ5lV40ry9+STuUzEEHmEeNDtb84vDaoKK34XYkRR4MYJkJaWpsWIjBgxAg8Pj6YddAPg4eHBoEEiD/L06dPk5Ij0VMesGosFvrcWmHUsC1/UxBk1CUdVZv5dxSBeaLl1Krx+v0J11nVFURjbX2Hx33Vk/aDw4aMKw52TIXF3EwGQdaEix8CRx45q831e6U1WcSZHj4plI0aMIDQ0tLrdJbWkb19RGTvJXHO9kbpyfp2931T4hBYiRry9vQkLC9P+eXl54ePjg7+/PwkJCcydO5cxY8bw97//nVtvvZWpU6cCwu2ycOFClixZwvjx49m3bx8vvPBCo1yQpO4YD5m16cP64EapLVIdo3uXgDGHFC8/sjxFdHje9gsY8qoOzpI0LqrqnMrb/dGueAQ6C4t9SZB8Viy7oj90bqdoVhFoWfVFKlPZVWPDqVfNRmEZCYh3TcO8w6dVpj+qUlIm5q8dDf97vPZu1QC9wl3XKPz+vo7Dnyo8PFe4a16/TyHIv25/+0efOIYhVyii9le3o8PM9vzyyy/aepnS2zDExcXh7u7uVPysqrLwdaXiXAWFB4SoCejnj3eEdw17NB71ap/47LPPatMPPfQQDz30ULXb9unTh6+//ro+p5M0AmVlZXQo6Aw6MAOFMUGM7dd054+NiYTsb6DjvWz3b8cNuamoZpXza3KgAdqXtxXOX1C57SUVH0+4/waFKwdyWY2xzq48x4WEfAD0XX3pMv/iIOYl6+3Tf5jk7KKBllVfpDKVm+ZNny6aevTvBt07Q1I6/HpA9P5o38ELzzBPDDkGCg8XoapqozcjO52pMuVvKhes2mf8QPjmWeWyqyT3iVF49b7L2/fsynNk/SCCVD2CPYhf2BtFUTQXDcBVV111WceWOOPp6UmPHj04deQkFtWCTtFR0ABiJHuDg1VkUvgltmx8Wk9EkOSy2PvbXqJ14oFz2tufOTM8mrRITlRUFJz/AqjkqpEF0OrE60tVVifAD1tgwoMqw+9W+W5z7YteAZgrLCQ+Z3/z6vl8j4sC5CwWlW82iml3N7hhnJi2FTvT6XSMGFFHW38zwlGMOMaNVOWqURRFixsx5BioONe41rysHJXJD6tk5Yr5oT1h+csK3l5NX9TKkGfg8KN290zvl3vi1c6L0tJSNm4UP5COHTs6ldmX1I/4+HjKKeeMWaTSFh0txlxurmGvS+MsRlznogEpRto8x360V0s8pA/mjw3Yobc2REVFQdFOKEsi0SeQXHeRRZO7OQ9zSesodNQUrNjmPL/rOMx+RqX3rSqLVqpUGGoWJakfpVGaImz/oWNDaDfV+U1JVVWWb4UM6/1r6jAIDVS4cOEChw+LDmsDBw68KO2/JRETE0NYmLgp79y506l5Y1W9ahzjRhoziDWvUFhETmeK+d7R8MsrCv6+rqmuefTJ4xiyhfhqNy2cjrNF1+NNmzZRXi5qlMyYMaPZt61vSWhBrNa4EdWk1itWyWKykL1R/DF7BLkTNDiw/oOsB1KMtHHObbffbItifImrR6fOyyEiIkIEO57/ElVR2OEvrCOWCgslO0pq2FsCwnR/NEVMd+skXAo2TpyBBQtVYm5UeeUrlYLiqkWJIdfAyf9a64Qo0OuFHiiKQlmFys87VP7ymoXouSrXP2Xff95E8f92h5TeluyiAWHtsFlHcnNzSU62187oGws9rG15fjsgLBVNETdSXKoy4zGVw9ahRHWAta8qhLmoE+u5Nee1AGf3QHfi/9tbEx2O8SLSRdOw2IJYGypuJH9XASZrk8ewCWEuq5NjQ4qRNk7QedEJzAKMu7ndpTduBHQ6HZGRkXBeVG3a4eCqKdrYMKlrrZ2f7FqAO65S2Pc/hdX/VRjvkF2blSvK73eZo/LE+xaycpxFSdIrp7QbU9Csjnx5yp+rH7cQcpXK1Y+rvPejKDFto12QiZmie0CL7UdTHZXjRmw4umpUFb77FQL62nvUNEbX6QqDyqx/qCRYPSLtQ2D9awqdwl0jRIyFJo48ckyb7/2vnlrQo6qqWryIp6cnkyZNcskYWyv29F57gHl9MmocU3rbuTheBKQYadNknDxHlEG4RVK89Myb5ZpI6qioKCg/BYU7OKQPpshNxFUXb62/T7QtsHK7ik5VuSY3jbGHT5GzOZcJPc1sfFPHzg8UbhiHlvJZWAL/+Qqib1RZsNBC0hmVgmPFpH4s/NAGNx3XHOnKX15T+XmHvV8JgKcHTBkKb9wPK5/PQm9tjdFagldtVBc3As5ZNUs2qOi7+aLzErfRhraMmEwqf3heZcMeMR/kJywi3TpXL0Qqsg0UrCnEVGxq0LHYOPGvJMozhRsmbHwonebZ6wMdPXpUK/k+btw4/PyavghcayYmJgYfHx9Om5MxIb7f+gSxnl9vjxcJm+D69GspRtow3711UvsBnAmxEBLgmrctrazx+S8wKzoSrJ18LaUqub/luWRMLYXCEpVf98OUCxncczaRwkWn2DV7D+u6bmDLuO34fnKMtwac49DCcu66RggKAIMRFq2EHreovHV1IlibaS0NiSbPwy5KI0LhT1fDD/9SyF2psOZVHfffAOGBIp6nvLxcS4Ht1q2b1qm7JTN06FBt2tEyAtAnRvwD2H4Y9p9S8LNWXi05VYKppOFEwH1vqFpNE19v+Pk/yiW7aKsWlT037iXzH2f5ffrOBhckF3blk/o/UUvGzdeN+Fd7O8WEyJTexkWn09GnTx9MGEkxCZ9dcVIJpqK6f8/lmeVavEngwAC8wr0adKyXgxQjbZiMLXZTfchQ11kgNDGS/S1uOosWNwJolR0lVbN2FxhNMLQ4x3mFRRTiSv3fGfYvOEjq9C3M+2wLmzoe4o2YNPpRiE610K8ol8H5Yt8cdy9+CI9meG94/k6FPYsUMr5XWPSYjlljFfyqCJbctWuX1mG7NVhFQHQc7tq1KyCqThuNRm2doijcf739c3j1G9UexKqKDIeGYN8JlQ9WiGkPd/j+RYVRfS/9spC9PkezzhQfL+Hg/YedAnDrg8Vg4dCDR8B6uO5PdMM3yrkqr0zpbXxsrhotbkSFgoN1d9U4ZdFMdr2LBqQYabMUlapEZNu//qv/3NllY9HEiCmXXu1T2esXSqlO9Ec499N56aq5BCu3qSiqSu/SfEAEFEbdGSkekJX+usszy8n/5Szdf0rk5SMJ/HhqM3/PPKSt97yjG6kr3Pn9fR1P36YwqIdSYzZEayl2Vhmbq6a8vFzLFLJx6zQIsyYefLMJzFEOlVgbKG7kv1/bRcQr9yhMHVaz1TL5vRSn+bMrznH6/1Kq3LaunHormeLjQmgF9A8g+u4uTuvz8/O130L37t2durRLGo7KZeHh8lw15x3jRSa6NqXXhhQjbZQVaw10LRM3lzR3I72Hdqthj8bDsftktO82DDo3tlutI6YiE+fXZle3a5vGbFb55XfoXFFCgFm8vYeMDKbPK70Zs3kUk09NZOiywXR7tCuhV4TgpndugOVWYcbP+tYf0D+AG//ViXbBdXPVtbZ4ERuXihvx8VK4z2odMZvhpyx7bERDxI2knlX5xtqgNSwQFlxb8z6FR4s0l6ZboP22nvjCCbI35VS3W60oPlHMqVdFppXiptD3jT4XZV6sXbsWs1m8NEgXTeOhZdTUoyy8xWAhd7MoVuMZ6kHgQNem9NqQYqSNsvWzs7hZba5ZwVkurQcQHR2tTXsUrSZAD5uCIrRlmd9muWBUzZ+Eo5BTAH2sVhEQYsSGR4A74ePDiHuiG8N/GMrk0xMYvWEEvV/qScSsDnh1EH5inY+OPv/phVLHYndms5lt20SBk3bt2rWqt+HqMmps/OU68LY2Fv+/vQ5ipAFqjby+VMX6XOe+65Va9YlKed/eXyrsrlC6PmLtJ2SB/QsOUppWdlljUS0qhx4+isVapybmL1EE9gu4aDuZ0ts02CwjaeZUjDrxIlHX9N4LCRcwFYsfWNiEMBS35lELRoqRNkh+kYqHwxucz4B81w0G6Ny5syaG0tNOMnscHNSHkGctgJa9PhtjvvFSh2iTrLTWiIkvvaAtCxkRXN3m6Nx1BA4IJPruKAb+rz8TDo9j/MErmHBgHMFDg+p8/iNHjlBQIG6EY8aMaVUFrgYOHIi7u8jqqkqMhAeJDrgA2UYPDGEitajoaDFqHareVia3QGXRT2Laxwvuva7mfSqyK+x1PwLcCbo2kG6PxhI+RcQCGPOM7L1tH+ayurs7z3yezoUd4vflG+1D98cutqBaLBZWrVoFgF6v1xqkShqeiIgIgoODsWAhlRQAylLL6tTLyzGLpp01XqSoqGkbPVaFFCNtkOVbVAYUCTOdEQudJrk2A8LT05OOHUWKYGpqKrdMUbAoCr8FiHFZDCpZK866cojNElt9EZtlROejI6CKt9bqUBQFn04+eIZ6Xtb5bVYRaF0uGgAfHx/69RNNmo4ePVrlzfqhufYuuYcRcSPmUjMlyaWXfd73foRSkTnLHTOoVWGztMVnsFSI7KbOf+yEzleHolMY8H5ffGNFkGnhwSIO/+1onQJay7PKOf5Puzsg/tU+uPm6XbTd7t27OX9eBJpPnjwZLy/XZ2a0VhRF0awjh0vtsUwF+2tvkcu2iRGdSM8GkUHWu3dvHnvssQYLeq4rUoy0QdYsL6GDUdzxjlhSiB8c7+IR2eNGzp8/z7Ae5XQMg01BdpFke/OTCFKyREXOcEMZ7azfZfDgIHSeTfcn7VjsrDUFr9qwuWpUVWX37t0Xre/RReFaa+G3o4rdVXO5JbrLKlT+73vxINDp4OEbaxYi5gqLViMGHUTdaW9s6BHoweDPBmixQhnfZJL6UVqtx3PkiWNa2minmzoSdmXVtShkSm/TUmUQ697auWrKzpRpgchBg4PwDPEkOTmZxMREjh07xu+//+4yC6cUI22MvEKVsu252vzeik307NnThSMSOAaxpqenMXc8nPQOIN1TvNnlbbtAWcbl+b1bIyutVpHeDvEiwSOrd9E0NKqqasGrer2eAQMGNNm5m4qa4kYAHpknbtynvR3Lwl9e3Mhnq+G81eM2exzEdqz5oZD1XRaG88JE3+Ga9vhE+jit9+/lT7+37C8bx55KJG/HBWri7M/ntLR6zzBPej3fo9ptHVN6pRhpfOpTFt7JRWNtjLd69Wpt2bRp0xpiiJeFFCNtjB9+gwGFdjGS3eE8Pj4+l9ijaXAUI6mpqaLviaKwOdDBOvKddNXY+MkaL+IUvHqJeJGGJiMjg4yMDABGjBihxVe0JoYPH65NVydGRveF4b0riZHLsIyYzSqvfmM3jz96U81CRFVVkj+wB67G/Dm6yu0iZnUg9n6xTjWp7L1jv1ZFtSqMhUaOPOZQ8v2lnniGVO3KO3v2rGY1GjBgAJ06dapx3JL6YbOMZFkyMXqIWLraumkcS8Db6otIMSJxCcvWmelbIlIAc9VCQvo13QPsUlQWI0N6QFQ7I5sDHbJqlmW6YmjNjqJSlc37xfQAg3jLVdwUgoY0XYqeo9uiNbpoAHr06KF1IK5OjCiKwt9uVMj28NbaGFyOGFm+FZLSxfT4gTCkZ81iJG9rnr2K5qDAS37/cU91J/SKEAAM5w3snb8fc0XVXbETn0+i4mwFINrKR1xffUyZLXAVpFWkqejTpw8AKioZnuKFoOJsBeVZ1QtMEC49W/q3ZztPAvr6U1FRwYYNGwBo3769Sy2cUoy0IbLzVc5tu4C3Km5C+wwJxPd1fbwIXCxGFAWuHl5Klpcvx33ETbboSDFFx1wf9e1q1u0S5dz9TQY6lYjOxgF9/XH3azrrhK0EPLS+4FUbbm5uDBkyBID09HQyM6sWw9eNhZiOCsleQrhUnK2gIrui1udRVZVXltitIo/9oXY+++T3Ha0iUZf09evcdQz8qD8+kaLUf/7uAo7+/dhF2+UlXCDtExGD4qZ3c+rIWxUypbfpCQkJ0QL+D5Uc1JbXVG8kb3se5lKRURU+MQxFp7Bt2zZKrPeQqVOnotO5ThJIMdKG+P5XZxfNXuMeTWW7mspiBOCaEeKPZJODqyZD1hzRUnpdFS8CdsuIm5ubkzujtVGbuBF3d4WH5iqX7arZehCtK2/fWJg67NLbg+iDc36NMLl7d/SmwzXta9zHM9STQZ8OROctbvtnFqdz5ot0bb25wsKhvx7R5uP+0f2iGBRHjEYja9euBSA0NLRV/w6aGzZXzcGiA9qymuJGsqtI6W0uLhqQYqRN8c1GlcHFQoxYVAv7jHu1H7WrqUqMxHU2Eh8DWwI7YEa8nWUuy0K1uCb1rDlgNotuugD9Dfna8qaMF8nNzSUpKQmAQYMGterurLURIwDzp8O5YLsYyUiofRCro1Xk0ZtqLsEPkPJhmtYnJupPkeg8ancrD+wfQPyrvbX5I48eJX9PPgCn3jhNSZJ4AQgcFEj0n7pUdQiNrVu3UlgornPatGm4uV2c9itpHGxBrEnmJG1ZTWXhbWJEcVO0zCibm01RFKZMmdIYQ601Uoy0Ec7mqhzaXU5MhUjrSjInUepWSlxcnItHJtDr9YSGij+QlJQUbfm8iVDg7sleP+HvLs8o58LvNWcDtFZ2HYfsfDE9gnxteXATipHt27dr063VRWOjNkGsAH6+CsOusouR/etqZxk5mqJq9WIi24nfe00YC4ykLxGxAm6+bnS5LbKGPZzpPK8TUVahYTGo7L39ALlb8zj1+mkAFHdR8r2mypwypdd12F4isy3nMfmK9OuC/YXV1ggpSS6l5KQQmkHDgvAI9CA9PV3ruzRs2DDt/usqpBhpI3z3KwwozNPm9xr3EBcXh6fn5RW8agxs1pGMjAytU+rc8WKdYyBrRhuuOWJz0XhZzLTLEW+l+m56vMKa7ntsrf1oqqJTp06af37Xrl1YLFUHfQLcfpcfRqtVoyKxiKLSmi14jg3xHpyj4OFes1XkzOfpmEuE77/TvI54BHnUuE9ler3Yg+ARQYBooJgwaxeqUYwl9r5oAvr4X2JvgS2lV6fTudzE39ZwtGhn+4sUbOMFI2WpVZc/cMyisaX0rlmzRlvWHL4/KUbaCN9sVBnk0GZ+r3FPs3HR2LCJEYvFoqWNdu0Ew3rB7/7hlCvi53p2+VkshuofCq2Zldaipz3KClCsZcdDRgY16Rhac+XVqrC5agoLC0lMTKx2u04d3SgJ0wPQsayUT364dPn1zByVL0TIBYF+cNc1NY/FYrKQssheuCx6QdQltq4enYeOQR8P0PoT2Vw+vl196fZI1xr3T05O5tgxEQA7cuRIQkJCLmscksujV69emjsv0XBcW15d3Ej2Bvu935bS65gJNX369MYYZp2QYqQNkJGtsv2gygBrSm+JpYRE0/FmE7xqo6q4EYCbJiqUu7mzI0B08jXmm5yCsdoKqWdVDglLOlN87a6qpnTRlJWVacGrcXFxtGvXrsnO7SpqGzcC0Gm4KMfvhsp3nxdhMlVvHXnzWxWjsLDzl1ng71uzVeTcT+cpTxcpnOFTwvHrrq9xn+rwau/FoMUDUDzs5+37Wh/cfGqO/ZAuGtei1+uJjRXNEHec26EtryqjxlxmJneLuPd7R3jh39sPo9HIunXrAJGdY8sacyVSjLQBlm2G2LJCAq1t5g+Y9mPG3OwsI47dex3FyNwJoCg4FUDLaIM1R36yh2ow2GR/AwppwkyanTt3ai600aNHN9l5XUlt40YAuoy0uze8M4r5/reqtyssUXl/hZj29ID7r7+MdN57Ls8q4kjw0CAGfNgPv15+9Hw2jtAxtbNwOFZdlSm9rsEWxHq45JC2rKog1txteVjKhSU5fFIYiqKQkJCgBR9PmTKlWQQfSzHSBvhmo8rgIntK7x6jeLNtbmLE0TKSlmY3RXcMUxg3APb5hVLgJvzj59dkYyw0NfUQXcpPO6x9S1QLgen5gHjT8enSdBV021K8iI3BgwdrJvGEhIRLbusYaxFbXsR/v1arDCr8cAUUinhCbp0KEWE1i5H83fnk78oHwL+3n1bErL5EXNuBK7aOJvb+mFptX1payqZNmwDRcdvWUFDStNju3/lqPmqI+I0VHCi8qGt09joHF82k5umiASlGWj1p51R2HIFBJY5iZA9eXl507Vqzb7gpqc5NAzBvgoJZ0fGb1TpiKbdw7qdzTTo+V1JcqrJxr5ge6V2Eai1eFDwiuMkaW5lMJj799FNtvrVWXq1MYGCg1r/pwIEDlJdXX+nSP94uRmLKi9h1HLYccN7GYFR5Y5n9gfG3WjTEA2erSPQ9ly5y1phs3LhR+wxmzJjhsnG0dRxfJgtC8gEwl5gptqZngyioZwteVTwUQq8QGTOO9UVcndJrQ4qRVs7SjeBrNtKzVJjv0s1nyLacp1evXs3CNOfIpcTI7CvB3a1SAbQ2lFWzbreougow03rjgaaNF/n666+1+iIjR45sdmK2MbHFjZhMJvbv31/tdp7Bnnh3FlVOY8uLUFTVKWMGYMl6yLAmN8wcAz2jan6Yl6WXcXaFEN+eYZ50vCGihj0ajyVLlmjTMl7EdTiKkdPKaW3a0VVTcqqU0hSRYRMyPBiPAHfOnTvH3r3izWbgwIF06FB9uf+mRIqRVs7STSJw1c0aLr/HuAeg2QWvAgQHB2sFtBzdNAChgQpThkKiTyCZnsItkbslt8Z+DK0FW2M8qNQcr4niRcxmMy+++KI2/8ADDzTJeZsLdYkbCbBaR3wtZtoby1i5HY6niu9PVVUWOoiTx2rREA8g9aM0zfzeZX4kbt5Vv0h8++23/OlPf9LcKA1NdnY2y5YtA0TV1alTpzbKeSQ1ExcXh4eHcFvvybP3inIUI86N8S5O6W0uLhqQYqRVczpTZddxGFTsXAIeml+8CIgqgDbrSFpa2kU1HeZNVEBR+NVmHbFA5g+tv5OvxaLy8+9i2tdLxSNRZNK4B7jj36tpqp8uXbpUS2sdN25cmyv9XZeMmoB457gRgNeXCiGx6nc4kizWjYqHUX1rFiOmYhNpn4my7TpPhag7qi5y9s477zBv3jw2btzITTfdRGlpaY3Hrisff/wxBoMBgDvuuANvb+8GP4ekdnh4eGjuw03JG7Xljh18q4oXaU4l4B2RYqQVs3QjoKqaGLHoLBwyisZKzdEyAnZXTUVFBbm5uU7rZo4Bb0/Y5NTJt/W7anYdh3PWenU3dCvFmCv8NcHDg1B0je+vN5vNvPDCC9r8U0891ejnbG707dsXLy9Rk6PGINb4AG26p1mIkU/XwPkLlRri1dIqkv51JqYCEawdcUMEXu28LtrmjTfe4L777tPms7Oz+fjjj2t1/NpisVj44IMPtPm77767QY8vqTu2l8oCUwHune1doy0GC6YSE3nbxY3DJ9Ibvx56zGaz1k8oICCAESNGuGbgVSDFSCvmm00qnQ2ltDcKV0ZmQCYViG6izdEyAs5xI7bCZzYC9ApXjYQMLz1J1qZkhQcKKT5R3KRjbGocXTTT9PnadFP1o/nuu++0Aldjxoxh/PjxTXLe5oSnpycDBw4E4OTJk+Tl5VW7bUBfu2XkigDx26wwwJ3/Ufl1v1jeowtcU4vMaNWikvLhpdN5Fy5cyEMPPXTR8v/+979aGnZDsGbNGpKThVln6tSpbSpmqLnieB8v6yACVy0VFoqOFZO7JQ+LQdw7wieGoygKu3fv1l7yJk2apLl5mgNSjLRSTpxR2Z/k7KLZXSHMy3q93umh35y4lBgBUQANnMvDt3bryEqH+iKxFxyKnTVBvIjFYuH555/X5p955pk2mz3h6JratWtXtdv5dPHB3V+8pXYsKMLdGt7hWCfmkXkKulpYtc6vy6b0lHC3hI4JcbK6ALz00ks89thj2vw///lPxo0bB4gg8K+//rrGc9SW999/X5u+5557Guy4ksvHUYyke9o7MBfsK3B20VjjRRxdNM0pXgSkGGm1LLW6EB1LwK87Kyru9e7dG52ueX71jmIkPT39ovUzRoK/L/wa2AFbREnmd1nVNohq6aSdUzlwUkwP6Qnl+/MB0HnpCBwQ2Ojn/+GHHzhyRLSVHzlyJJMmTWr0czZXahs3ougU/PuIWB5jVjl/HOVsnegQArdMrt05U95zSOf9s/1vQ1VVnnvuOf7xj39oy1588UWeeeYZ/vznP2vL/v3vf1+yn05tSUtL46effgJEbZGrr7663seU1B9HMXKw+KA2nb+vgOwNInhV56kQOlbUpHEUI80t+Lh5PpEk9eabjSoeFjN9S8SbtC5UR4pZmFibq4sGaraM+HgpzBoLFzy8OKAXf2ClKWXk7750++yWys/2Ss9c36dCS9MLGhyIm1fj/vlKq4gzjmKkLnEjC+Kd3YgPzFbw9qr5cyw8UqSV8faN9aXdFBGAqKoqTz/9NM8++6y27SuvvKIJk2HDhmnVcY8ePcrKlStrPFdNLFq0SBM1d911F+7u7vU+pqT+REdHo9eLlgBb0n7TnujnfjpH2Rnhng8ZHYK73p3c3Fztd9unTx8iI+vW7bmxkWKkFXI0ReVwskgB9VbFDaQ01n5DbK7Bq1CzGAG7q2ZTkIOr5tvWWR5+5Ta7xWecR742HTy88V00y5cv5+BB8bY1bNiwZvcm1dR07dpVawi3c+fOS1rjHONG2hcUMc3q4Qn2h3tm1u58KY5FzhZ0QdEpqKrK448/zr/+9S9t3euvv86jjz7qtK+j6+bll1+ul+XQaDTy0UcfAeDm5saf/vSnyz6WpGHR6XTa/fx48nH0cb6A6N9lI9zapXfdunXa76C5uWhAipFWydKN4gfnGC9yWm8vitOcLSMdOnTA09MTqF6MTBoCoYGw3b8dFdZOvlk/nsVibJ6dfE9nqkx/FG7+Tzte/gJ2H1exWGp+OJSUqWzcJ6Y7hUNAij1epLE79aqqKq0ilVAURbOOZGdnX1SYzxHH9N7CQ0UseUbhv39R2PyWQrB/zZ9jwcFCMqwC2z3Anc43dUJVVR5++GEWLlyobff222/z4IMPXrT/VVddpfUuSUhI4Ndff63VNVbFjz/+yNmzIoV+1qxZdOzY8bKPJWl4bPdzVVUxd7m4RUZzT+m1IcVIK0NV4RtrvMhgmxhRYHuhve17cxYjOp2OLl26AJCZmVnlG52Hu8LscVDm5k6Cv/hDM+Qaydmce9G2ruZYisrY+1TW7oIdR3146iMYukCl3UyVm56z8MkvKhnZVQuT9btFFgbA1SPhwu9WMaKDoKGNaxlZuXKlVml08ODBstKmldrGjfj19ENxE6Kj8HARQf4Kf5un0K9rzULEXGbmwN0HUY3idxF9dxQ6Xx33338/b7zxhrbdBx98wL333lvlMRRF4YknntDmX3755RrPWx3vvfeeNu0YjyJpHjjez8/7nXda5xvjg76rLxaLRRMjer2+WfaVkmKklXHoNBxPg1BjOdEVwjUTODCQPSdEsbPAwMBm/2Zjc9UUFxeTn59f5TY3TbJl1TiUh29mrpr9SSpX3K+SmXPxutwC+HoD3PFvlc43qMTfZuHhty2s2alSWi4eQisdUnqv6W+i6Kj4PgPi/fEIaDyfvbSKVE9t40bcvN3Qdxe+/OLEYiyG2lvtEp8/QfEJkaYZ0Nef2Aej+fOf/8w777wDCKHx8ccfs2DBgkseZ+7cucTEiOZ3a9eu1UqA14Xjx49r1Vzj4uKYMGFCnY8haVwcxchxw3GndbaU3gMHDnDunGgnMGHCBK1mTnNCipFWxrfWKtADHVw0AaP9tPLq8fHxzf7BcqkeNTbG9IWOYbDHL4xCWyffVdmYiptHJ9/fj6iM/6tKjjWudlAcrHgui3ceguvGQoDeefsjyfD6Upj2iErI1SqTH7bwwxaxzscL+hvzsVb0b/R4kV9++YU9e4R4HTBgANdcc02jnq8lUbcgVuGqUY1qrWvhZG/MIeVD8beq89bR990+LPjLAj788EOxTKfjs88+Y/78+TUey93d3SmW5N///netxuCIY5Gze+65p9nfO9oiNnccQELm7+g87d9RVSm9zdFFA1KMtCpUFZZaxchghy69+ZH2WIPmHLxqozZixM1N4cYJYNLp2BrQDgBzqZlzv5yvcvumZOMelUkPq+Rbnz+j+8L61yA+2sA9M+H7f+nIXamw9R2FZ26HEX3AMdO6wiBcNHnWqs6TBkPx7nxtfVX9aFJSUoiLi2PEiBEcPXr0sscurSKXJjw8XLM27N69+5Il1x2DWAsPFdV4bEOugYP3HdLm457pxn0v38cnn3wCiODRL7/8kltuuaXW450/fz7t27cHYNmyZZw4caLW+5aWlrJ48WIAvL29ue2222q9r6TpaN++PaGhohvvwaMHCRwkUv7d9G6EjhYB16tWrdK2l2JE0ugcSfXkZAboVJWhZSIl0D3AnURTorZNc44XseEoRio3zHOkORZA+3mHyozHVUpEBi6ThsCa/yoEVmoh4+6uMLqvwnN36Njxno6clQrfPq9w1zXQpb3ztn+YrNjjRai68uqiRYtISkoiISGBESNGXHY655o1a7RYiH79+jFzZi1TP9oQEydOBETLgi1btlS7nX8fBzFy+NJiRFVVDj18hIpzIkgofGIYL+96iS+++AIQVo6vv/6aefPm1Wms3t7eWnVWVVV55ZVXar3vN998o7lJb7zxRi2TSNK8UBRFu69nZmbS+e+d6DgngoEf9cfNx42CggK2bxcV9+Li4oiNjXXlcKtFipFWxE8JIq2rW1khPtZ+86FXhHD42GFtm9ZiGQFRBKxrJzjqG8R5D9GwK2dzLhXZFY0+xqr4dpPKrL+rWtDptaNh5csKep+aLQvB/gqzr1T48FEdKUsVEr9UeO9vCt88qzB7lFnrxOkb64tX+4v9vYcO2d+oi4qKmDlzJi+99FKdUjpthbRsPP300822OJ4rmTJlijZt6/NRFU4ZNYcLq90OIP2rTM79JKx6HiEehD4RzEf/E+m0Hh4eLFu2jNmzZ1/WeO+55x4CAkTdk88++6zaLLXKyMDVloPjS2ay+TQD3u+n1aXZsGEDZrMZaL5WEZBipNExm1XSz6uNXiFUVeGXnSIQYUipQxngCWFaBU1oGZaR6OhobdrWE6UqFEVh3gRQFYVN1kBW1axy9qemd9UsXqUy7zkVk/ibZ95EWPZC7YpbVUZRFOIiFe6ZqTB3gkLh/iKtx0R1/WgcxQgIYfGPf/yDefPmUVJSUqvzrl+/nt9/F+2B+/Tpw/XXX1/nsbcFJk6cqLmu1q1bV+12XuFeeHUQwrHocFG194CS5FKOPmn/nfd9vQ+b9m3S5h9//PF6WagCAwO1rBuj0chrr71W4z579uzRSt4PHDjQKVZG0vxwvK8fPnzYaV1LcNFAPcTIwYMHGTp0qFYMB2Dx4sVMmjSJCRMm8Oabbzr98R05coR58+YxevRoFixYQFZW6+4nAuKBMO1RlcjZKv3nq3yw/P/bu/OwKKv2gePfYd9UEHADBHdRXjfcEHPftVxCsyzNLDXfTN9cMlPb1EwrtVXNyi311X4uaVZvhimuue8rioIigojILszz+2Och0EBAQdmGO7PdXnJbM+cmxmeueec+5yjkJRSPEnJwXMQFaebYdFRm10v4tnJQ31zenp6UqlSpWJ5fmPy9fVVx7nDwsLy3exr8IOhmj3ls8c2bv0vtngb+JCvNygM/1hBv+r2iN6wapoGWxvj1FrE7zPYj6a16yO337t3j4iICEC3f4rhgljr1q2jbdu2+Q53gfSKFEbFihVp3rw5oEsC8zuX6etG7idkknY97ZHbtZlajr9+kqxkXRbrPcSLKn0q88cff6j36d279xO3efz48Tg46HoPFy9e/MiO2A97uFdE6obMW17JiKIoavGqvb29um+ROSrS2Uar1fL555/ToEED9brdu3ezfv16li1bxrp169i7dy+bN28GICMjg8mTJzN48GBCQ0Np3Lgx06dPN04EZuz0FV0hIuim3I7+TMHrWYU3F2o5d9W4SYm+cNU56z5Vb+u6hJ3rOJPilKIuWFQahmhA1zOgn0KYnJyc76yFgJoaAmpAuEM54mx030Jv77pNVkpWibT1k58U3liQ/Vq++SwsmaTB2tp4J+/H1YsYFqz+61//YurUqWzevBkXF12hyrFjx2jevHm+9Q07duxgzx7dWjT+/v5FHhIoKwyHavLrHSnfMP8i1vAFV0g4mACAk58jDWbVJysri+3btwPg6upKixYtnri9lSpVYsSIEYDub+qrr77K874JCQmsXr1a1/7y5XnhhRee+PlF8corGTlz5oy6x1eHDh1wcnIq8bYVVJGSkQ0bNhAQEKBWlYNuOmD//v3x9vbGw8ODF198kW3btgG6Lj9bW1v69euHvb09I0aM4OzZswUeuyytftnz6HWJyfDl/4H/Swpd/qNl4y6FzMyiJSaKonD+msJ3WxRWPzgfNkuNR/NgdU/PTu6lbohGT18kCLoxz/w830UDGg2HyummsWnTtMTtKt4F0BRFYdp3WqYszn7t3n0JFrxZsN1YC/w8WQp3/kkAwK6SHU41Hz2ZGA7R6Kf5PfPMM+zfv1/d5j02NpZOnTrlmKppyLBXZNq0aVhbWxsrBItU4GTkX9l71DxcN5JwOIFLc8N1F6yg8aJG2JSz4dChQ9x5sDtzly5djPZaTJw4UT3WF198kefw3cqVK0lN1VVgDx06VN37RJgvV1dXvL29AV0yoh+VKC1DNACFXjkpISGBNWvWsGzZMj777DP1+itXruTYu6J27dqEh+v+0C5fvkydOnXU2xwcHPD29uby5ct4eXk98hwZGRlkZGTkbKiNjbpMuLHoN34yxq6Wudm8O/vnjbNgyx5Y8xekPqiv/Osw/HVYwacSjHxaYURvqJxPwXpWlq6HJexE9r9bd3Lep4999oewewd3dpwMVS/7+/sXW6zG1rFjR/Xn7du359uTNrADvPsdHCjnSY87ugQ35o9beHbzKJa2KQq89RV88X/Z181+Dd4eoktScqsNKOp7LfFkIpn3dGunuLVyzfX4hslIgwYN1Ofw9/dn//79DB48mL/++ovMzExGjx7NsWPHWLBgAba2uvVZdu7cya5duwBdtf3AgQPzbWdx/92UtKLE07JlS1xcXEhKSuLPP/8kMzMz12Etl4bZH+SJJ++pz5GZnMWx0SdRsnSvZa23alIhsHyOlTIBunbtWqh25RdL9erVGTx4MD/99BPx8fEsWbKEcePG5biPoig5hmhGjhxp0tdZ3msF17BhQ6Kiorhz5w5RUVF4eXnleC9169bNqM9bmFgKMuRb6GTkm2++4fnnn6dcuXI5rk9JScmRQTs7O6vZdWpq6iPZtbOzc55z9H/88Ue+++67HNcNHDiQQYMGFba5BRIZGWn0Y8bcseafs7pMtb5PBo29o2n8HIztY8X6MGd+Ci3H1Rjdh0HkLZj+PXy4TKFnyxRe7HSPwDrp3M+C0xF2/HPegX/O23PoogP3UvJ+USs4ZdLgiq54U2OnIcnnHvtX7ldv9/DwyHd2ijnRaDT4+fkRERHB/v37OXPmTJ7f0GyAxjWrcPxSRdI1VtgrWm7+dhOXsU5GH+vO0sK7P1Zk3a7s9//7L8YzuO09CvKrLex7Lf43g2yznpLr63fo0CH15woVKjxyn2+//ZaPP/5YXa9i0aJFHDlyhK+//hp3d/cc29CPGjVK7dZ9nOL4uzGlwsbTsmVLQkNDiYmJ4X//+x/+/v6P3EfRKGgcNSipCvHH7qivTfTsGFIu685/Dg0dsA2xVm8znJbdoEGDIv3N5hXLiy++yE8//QTAvHnz6NWrV44veQcOHFCLxvUJlzmcM8r6e60g9NtoAISGhhIYGKgOzXp7e+Pg4FAsr2VBYjEcRclLoZKRc+fOcebMGd5+++1HbnNycsrR7ZecnIyjoyMAjo6Oj3QJJicn5zl+NXz4cIYMGZKzocXUMxIZGYmPj4/Ri/X+OJH9c0hHuxzTVRs1gA9egz8PwTebdNvEKwrcz9Lwyz5nftnnTK1qEB0PKY/WvKnKO+sW1GrXCIL/pcU15jKRz+uKPSsGuVGjXo0chYudO3fGza34d3s1Bq1WS5s2bYiIiCAzM5OIiIh890cZ1hPe+hqOO1ekZVIcmbFZuN2tSIXG5fN8TFHMWwPrdJ0IWFnBkkkwvGdFIP81GIr6Xrtz4a76c80eNang+2g8+h7ISpUqERgYmOtxli5dSnBwMGPGjCEjI4N//vmHkJAQJk2axL59+wBdb+Ybb7zx2O3hi/PvxhSKGk/fvn0JDdX1PJ46dSrPbvDohre4e+gu96/fp5pbNe7sSyBhg+51tXayosX3zXCupUu0ExIS1D2B6tevT5s2bYwai6+vL3369GHr1q1ER0cTFhbGK6+8ot5uuJ/Nm2++meO8ZQryXiu4oKAg9Ut8bGws4eHh6ghD7969c8xSNAZjx1KoZOTIkSNcvXpV/VBISkrC2tqa69evU6NGDS5duqRW64aHh6vj1TVr1uTnn39Wj5OWlkZUVFSei6/Y2dkZPfHIj5WVldHfGFv26Lqu3O6n09PqHmR5YGWb/RxWVtCzte5fRLTCos0KS3/V7VkCEJ7LNiuertCuMbRrrOGpRtCoFmqhpFYLR7dm9zTp9yTQFzNVq1ZNXaWvtAgODlYL6Xbs2EGfPn3yvO/QHgrvLlX4p5wHLZN0U5vj/ozDramr0dqTcV9hwXpdt7qVFayZoZt6WxiFea8pisKdA7qeERsXa1wbVUDzUD1KbGysuudEQEBAvsceMWIEDRo0YMCAAdy8eZOrV6/yxhtvqLdPmzatUH93xfF3Y0qFjcdwWHr79u1Mnjw51/tVCCjP3UO6P+z4XXc4PSl7Gq//zPqUq5Pdy/b333+ra0J07969yL/f/GKZOnUqW7duBeDTTz9l+PDhWFtbExMTw8aNGwHdzLuQkBCzeX3L+nutIBo3bqz+fPr0aa5cuaJe7tmzZ7H9/owVS6GOMGDAADZu3MhPP/3ETz/9RLt27Rg4cCBvvfUWvXr1YsOGDURFRXH79m1++uknNWkJDAwkPT2dzZs3k5GRwQ8//IC/v3+u9SKWIClF4a8jYKVomRd1mIRJRzk48HCeW9z7VdUwZ7QVUT9rWD5VQ8sHvb0+lWBIV1g8UcPZlRpiNmv4+SMr3gzR0LTuozM2kvZn9z55dnInJiaG+HjdSqylqXhVLygoSB1m0c8uyIt7BQ1Du8M/D3bxBbj1h3Gn+K7bATd1v076P0WhE5HCSrmSoq7I6drSVd0F1pBh5XxBXuOgoCAOHTr0yAyNmjVrPtIbKfJXt25dtWt8165d6rD0wwyXhT859hQZcbrXtFIPT3yGeue4r+GUXsMiWWMKCgpSvzSeP3+eTZs2AfD999+r0+hHjBhhlpupibz5+/ur58uTJ0+qxau2tralYoPDQiUjDg4OeHh4qP/s7e1xdHSkXLlytG3blpCQEIYNG0ZISAitW7dWF+qxs7Nj3rx5rFmzho4dO3L06FE++uijYgnIHPzvoG5/kQYpCVRN0SUIt8PiOTvtfL6Pc7DXMLSHhgOLrcgI1XDtZytWTbdi5DMa6vtq8q1/yErLIuWw7mToUNUel/ouOT6oSsu0XkOurq40a9YM0K1rc+tW/ouZjR+o4batA+EOupP/3WOJpEXnM85VCIqS3Suif67iFr8vQf05r8XOCpuMAHh5ebFz584ce5y89957jx2eETlpNBq6du0K5L80vOFKrJlJul4PO087/rWgYY6/aUVR1GTEzs6uWNeEeOedd9SfP/74Y7KystSZVhqNhlGjRhXbc4vi4ejoSO3atQE4evSo2jPStm3bR2o8zdETnX3ef//9HJeHDx+e526SDRs2ZO3atU/ydKXG5t26D62gxJzfzK8uvUb5gHL4vOSd28NyKOyCWXf2J6Ck657Xo6MHGo2m1E7rNdSpUyd1B9nQ0NB89+ao76uhZyuFf255UCtNt6ZD7Pa4Av2+H2fvKTj8IJdsVldXq1Pc9EM0AG4FWHnVcPfOx3F0dGTFihUMHDiQrKws+vXrV+R2lmXdunXj+++/B3RLw+fWm+Hi7wIa1F2XARp9EYC9Z86eh4sXL6oFhk899VSxTqnt1q0bTZs25ejRoxw+fJgJEyao9WU9e/Y0en2BKBkBAQFcvHgxxwwXc5/Sq2c5g3BmIjNTYes+QFFok/Tgm7xBXnFq0hniD9zJ9bFPIm5H9pRej0662pDS3jMChVtvBOA/gzT845I9VBPzu3GWhn+4V6QkVqSMf7DYmcZWg+uDnTgfZvgaGy5CWBAajYZnnnmG/v37ywqbRVSQpeFtnG3UAlWA6i/7qPuGGDLc56a4hmj0NBpNjmLVhQsXqj/LPjSlV25fOnv27GmClhSeJCNGtveUbuv3GulJVMrQDRF4tHfHb6RubFm5r3Bk2DFSr+c+vlxUcaEP9qPR6J4PnuyDylwEBwerRZUFSUa6NAe7BuW5Y6N7zK2/b5OV+mSrsV69qbDhwQyayhVhUMf8719YSpZCWnQaCUfucvPXGCKWXuPc++dJCdcVJLs2rYC146MLXymKor7Gvr6+6mZoouS4u7urS8OfOHEiz6XhvQZXA6B84/LU/7BurvcxrBcxLI4tLs8++6zara/n6+tbaj68xKMe7h2tVq1aqekVl0FiI9MP0bROzP5GXrlXJXyGeXPvbBK3w+LJiM3g8EvHCNraEmunJ19d8fq6GySd09WmVGhWAbuKdiiKog7T+Pn5lYoxw9w4OTkRHBzMjh07uHLlCpcvX853C2yNRsO4QVYc3ONBt4QbkKbl9u54KnV99JtoQX29MXvfmTH9NNjbFa4XQVEU7h67y919iVy+f4X06AzSotNIi04nPTqN9FsZ6uJXuclriCYyMpJ793TDUYUZohHG1bVrV3VTue3bt/PSSy89cp9a42tQtV8VHL0dcsyq08vIyGDHDt2eDpUrVy6R19Pa2prJkyczcuRI9bqRI0fK6rul2MOJR48ePUpNr6f0jBiRoihsfrAEfJt72clIpR6VsLKxoukPjXHy0629kng8kRPjTz3Rbr6KonDp03COv55dN1C1n27DOMMPqtI6RKNX2KGaIV3hbOXs5OPKpqIP1SSnKnz3YA0qO1sYXYjNU5MuJHFh9kV2Ng9jX9d/uDHjJhc+usTVpdeI+fUWd4/cJS06Pd9ExNbNFu8Xcp91ZlgvUlq+/VgiwyEVw6EWQxqNBucaTrkmIgB79+5V12Lq1q1biU1jHTp0KNWq6XptbGxscqw5Ikqf2rVr55ieX5p6uaRnxIjOXoXw61A5I5WaaUkAVGhaHkcv3W6ZdhXtCFzVlL3dD5CVnEX0/92kfEB5ar35+NXpHqa9r+X0pDNErsze38c1pALVX/UBsIjiVb0uXbowbdo0QPfN87XXXsv3/o72GoJerMj9GRpsFYXrv8WhKEqRviGs+AMSdC8lL3SBSm75HyMtOo0bG29y4+doEo8n5ntfNLpZFQ5VHXCoao9DFd3/9lUdcKhmj0NVB5z8HLF2yP2balFm0gjjCwoKwtnZmeTkZP78888ivddKeohGz97ennXr1vH+++8zbNgwqlSpUmLPLYzP1taWhg0bcvToUaytrXN8kTN3kowYkX4vmtYGvSKVe1fOcZ9y/uVo/M2/ODLsGADnP7xAOX+XQg0jZN7L5Mgrx4gLzS5arfdeHTR9wMpG943KEopX9QIDAylfvjyJiYmEhoai1Wof+81x1CBbls6tSOC92zjcTePWkXtUDixcTYVWq7Dw5+xei3EhuX/A3E/MJGZrDNd/juZ22G14eDkZK3BvVxHrJtZUC6iKo5cDDlUdsK9sj5Vd0b8BSzJiHuzs7OjYsSNbt24lJiaGkydP0qhRo0IdwzAZ0U8XLinBwcH5bvYnSpdZs2bxzjvv8PLLL5eaFbdBhmmMKrteJHtKb5VelR65X5U+lakzRbc6LQoce+0ESRdz30HzYWnRaex/+h81EbGy09BkaSNqvOGX49uYJX1Q2djYqBvnxcXFceLEicc8Aqp6aKBFdoL31zeFXwDtj3/g/IPV9Ds0hSZ1sn+/2gwtMdtuceSVY/zlv4MTY09xe2fORKR84/L4z6pHp5MdaLE+EPcX3ajStzJuLd1w9HF8okQEsodprK2tqV+//hMdSzwZwwQir6GavNy6dYujR48C0LRpUypVevScIURB9ezZk2PHjjF+/HhTN6VQJBkxkug4hQNnoHxmBgEpuimZzrWccK6b+1oBtSfUosrTul6TzHuZHB5yhPt37+f7HPfO3mNv9wMkntTVgti62tByQwuq9a/6yH31wzRWVlYW8UFV2LoRgB5vZCcjCTti0WoLV5+TW69IVmoWZ945y1/+Ozj80lFubo5Bm5adgTj5OVJ7Yk3a7W9L29Agaoz2w6GK8VeyzMzMVDc0q1u3rqyWaWKGdSOF7WUwvH9JDtEIYU4kGTGSrbq9xmh5L1b9pVbuVSnPsWONlYZGXwVQrqELAMnhKRwbeSLPYsa4nbfZ1/Mf0q7rpgs7+joS9FsrKgY92g2n1Wo5c+YMALVq1VI3LCzNunTpov78uKXh9Vq0d+KWq+73W/3uXX7/Pb3Az3cmQuGPf3Q/16gKTz/Yr+z0lLNELLnG/YRM9b52Hnb4vlqdoN9b0f7QU9R9pw4udYpvwSqAS5cuqZtglfaeL0tQr149fHx09Vr5LQ2fm5JYAl4IcyfJiJGoq67eyx4OeLhe5GE2LjYErmyKbUVbQLda6PmPLjxyv6i11zk46DCZ93QfgBWalqfN761wqeuS63GvXLmingwt5YOqfv36VK2q6wHatWuX+kH8OO5ddL0jVsAf3xZ8qOYLg16Rsc/q9gFKupTM9TW6HQyt7K2oFlKV5mub0elUexp+4o9bC9cSm0ZnOAwn03pNT6PRqIlEWloau3fvLtDjFEVRh3WcnZ0LvUuvEJZCkhEjSEpR2H4Y7LVZNEvW1XLYV7bDNTD3VTMNOfk60eyHxuomaJe/jOD6et0HnqIoXPw0nBP/PoWSqftwrNTDk1abW2BfKe9ueUsqXtXTaDRq70hKSgr79+8v0OPaDM8eqnE5GcfJ8McP1cQnKqx48GXVxRFe0e33yKV54WrPVe0JNWmyuBGVunrmOV2zOMm0XvNTlLqREydOqLsud+jQQYbbRJklyYgR/HlItzFes6Tb2D1YHatSz0qPbPeeF/en3PGfnV3XcXL8ae4cTODkuNNc/PiSer3vCB8CVzTFxjn/SVCWNK3XUFHqRiq2cCWrnK7nqWnybb5Y8/jVWL/bAqkPRnRe6QUVXDTcO5fEjf/Tra5p526L30jfQrbeuCypQNlSFGRp+IeZakqvEOZGkhEjyG3V1dxm0eTHd4QPPi/pFrfSpmnZ1+sAUT9lryFS/8N6NPjEP9dt5B9miT0jkDMZKWjdiMZaQ9Xuut4RJ20WJ365w607efeO3M9U+GqD7naNRjdEA3BxziV1o7Oab9bAppxpZ8XrX2NHR8d8V6QVJcfDw4PAwEAAjh8/zs2bNx/7GMMeFElGRFkmycgT0m+MZ61oaZ2kq0mwcbGmYlv3Qh1Ho9HQcG4D3Fq56q54MEHDyl63cmvNf/sVuB5B/0FlY2ND3bq574NRGnl7e1OvXj0ADhw4QGLiYxYVe8Crd/ZQTbM7sSzanPd9N+6CqAelJX2CoLa3hrsnErm5RdeVbl/ZDt9XqhctACNJTU3l0iVdj1mDBg1k+W4zYliA+riEOTk5mbCwMEC3J0ydOnWKtW1CmDNJRp7QvtNw+y40TE7AJUtXYOrZ1RNr+8L/aq3srGi2rAkO1XQrttq62dJyY3Oq9i34qoj379/n/HndXvd169bNsTSwJdDXjWRlZbFr164CPcajgwcaG10i1+JeLN9s1JKekXvvyMO788KDXpEHao2vaZT9hJ7E2bNn1S3CZYjGvBSmbsSwELt79+6lZg8RIYqDJCNPKHsWjeGqq0VftMi+kj3BO4L414KGPLWrDRVbFW4FPUuf8lmUuhHb8ja4t60IQJX7aThEJ7M2l4f+c0Zh34Nym4Aa0LEZJBxK4NYfuq4SBy8HfIb5PFkARiD1IuZLvzQ8oC4NnxepFxEimyQjT0BRFN0S8IqiTunV2Grw7Fz0HWIB7D3s8HnJW+0hKQxLLV7V69Chg7oUfEHrRgAqdct+TVrei2X+euWRDwrDRc7GD9Sg0Wi4YFBAXHtCzSL1eBmbTOs1X/b29nTo0AGAmzdv5nitHqZPRqytrenUqVNJNE8Is2X6M2spdu4qXLoOtdLu4XlftxiZRzt3bMubrrjRMBmxpOJVPTc3N7VI8NSpUwUqEgSo1D1nMnL8Evx9NPv2G3EK63Q7uONRAV7oCvF744n7WzdV29HXMc/dc0uaTOs1bwXZxffatWucO3cOgFatWuHq6loSTRPCbEky8gT0G+MFGcyiqVzIWTTGVha68A1XYw0NDS3QY5z8nHCpp+s+r596l/KZGcxfl90T8s1GhcwHs35HPQMOduToFakzqZZJ1hPJjf41dnV1Vbd/F+ajIHUjhtfLqqtCSDLyRDbveWjVVQ1U6mHaZES/DLy9vT21atUyaVuKS1HqRiB7qMYKaJ4Ux9Z9cDFSITVdYdEvuvvYWMOY/hpu74wnfm/2HkPVBj66/48pJCQkEBUVBeiSTSl6ND/169fH29sb0BWppqWlPXIfqRcRIidJRoro5m3dxnhV01PwS08CwLW5a7FsilZQ6enpXLx4EQB/f3+LnfIZHByMg4Ounmb79u35FgkaMkwUW96LQ1Hgi/9TWP2nbkYUwKCOUNUdLsy+qN63zpTaWNmYx5+K1IuYv8ctDZ+VlaXWO7m6utKiRYsSb6MQ5sY8zrCl0NZ9oCjQ+p75DNFcvnyZrCzdWIOlDtEAODg4EBwcDOjG3sPDwwv0ONfmFbB1063GGpgUh41Wy4+/wby1OXfnjf0zjoTDuuzExd+Fqv0KPrW6uJWFYThLkF/dyMGDB0lISAB0Q46W+qVBiMKQZKSI1FVXDTfGM3EycuFC9iZ7lli8aqgou/ha2Vjh2cUD0K3G2jDlDsmpcP6a7vY2AdDCHy58nN0rUndK7QIv618SJBkpHfJbGl6GaIR4lCQjRZCcqrD9ELhmptMgJQEAl7rOuNQu3m3jH8cwGbH0D6oi143kmFUTl+O2cSEaYrbeIvHEPQDKNy7/RGvGFAeZSVM6eHh40KxZMwCOHTumboYHUrwqRG4kGSmCPw9BWoZuiqj+F2jqXhFArRcBy+8ZadasmTodMjQ0VF2R9HE8O2WvxtrxfqxurA3wqQT9gxUuGKy2Wved2mZVIKooitozUq1aNSpWrGjiFon85LY0fEJCAgcOHAB0ha7Vq5t2awEhzIUkI0WgrrqaaDBE07uyqZqj0veMODs74+tr2l1li5u1tTUdO3YEID4+nmPHjhXocbYVbKnYWreqbYV7qfhkJAPw1iANsVtjSDqnL0auoA7pmIubN28SHx8PSK9IaZBb3UhoaKha1yVDNEJkM+3WoybW5T+QllaZJvWgoZ9CAz/w94XKFcnzG3FWlsLWveCYlUmTZN0Hg0NVeyo0KV+CLX9UcnIykZGRgG7zNP0qpZasS5cubNy4EdB989R3iz9OpR6e3N6te+0Wt4sjtnM5hnTWsjvYsFekjln1ioDMpCltgoKCcHJyIiUlRV0aXupFhMhdmU1G7mcq7D4J9zMd2HcW1P3hAbdy0MBPwd8XGvhp1CTFp5JuY7y4uxCcdBs7RTc0UKlnJZMXOZ49e1ad4lpWvjU/XDcyefLkAj2uUjdPzk7TbSbodi6OnvNqErX6JsnhKQBUDHbDvb35DYFIvUjpol8aftu2bURHR3Pq1Ck1GbGzs6Ndu3YmbqEQ5qPMJiPRt3VJx607j9525x7sOan7Z5ikuDhCOSfdz0FmNKUXyuYsi7p16+Lt7U1UVBRhYWGkp6djb//4dV6caznjXNuZ5EvJ3DmQQPqtdC5+mj092Bx7RaBsvsalXbdu3di2bRsAX3/9NVevXgXgqaeeUjfUE0KU4ZqR6pU1RG+Ew19HsvMLWDxRw7gQ6NYCvPPY5y4pVZfE2Gi1tHgwE8OmvA3uwab/Fq1feRUsv3hVT6PRqL0jqamp7Nu3r8CP1c+qUbIUjo08QerVVAA8OrpTMahwOyWXFH0yotFoaNCggYlbIwrCsG5k6dKluV4vhCjDPSN6bi5amvhCuyYaIPvbcGKywrmrcOYqnIlQOHsVzkTAlWgISLmDizYT0HX5W9mZPqcrq9+au3TpwvLlywFd3Yh+x9THqdTdkytfRwBwOyxevb7uO7WN3USj0Gq16iaItWrVwsnJycQtEgVRv359vLy8uH79ulq4ClIvIsTDTP8paqbKO2to2UDDyz01zH3dii1zrAhfa0XSHxoWBxvOojH9EA1k94xUqFChTG2eZrj1emHWG3Fr6YpNhZy5eKUenrgGuhqraUZ15coVUlJ0NS1lKdks7QyXhterXLkyjRo1MlGLhDBPkowUkqMd3AvV1YtY2Vvh2dH00z/v3LmjzqQpa5unVatWTR2y+Oeff7h7926BHmdla/XI1N26U8yzVwTKbs+XJXg4GenWrVuZ+hsVoiAkGSmku0fvkn4zHQD39u7YlMt/pGvBggU0b96cLVu2FEt7FEVh3Lhx6uWy+I1LXzei1WrZuXNngR9XuXt2r1aVZypT/l+mnZ6dH5nWW3p16dIlR/IhQzRCPEqSkULQZmqJWntDvVzlMbNo4uPjmTBhAocPH+bZZ59lx44dRm/TV199xcqVKwFwcnLijTfeMPpzmLui7FMDUKVvZbyeq4ZHB3cazKpfHE0zGpnWW3oZLg2v0Wjo2rWriVskhPkp8wWsj6MoConHE7m+7gY3NtwkIzZDd4NGV2OQn127dqnLlN+/f5/+/fuzZ88eo812CQsL46233lIvz507l/r1zftDtTi0b98eKysrtFptoepGrGysaPxN8fcyJCUlMWnSJBRFYcGCBTg4OBT6GPqeEVtbW+rUqWPsJopi9umnnzJp0iQGDx5MpUrmUWcmhDmRZCQPKddSubH+BtfXR5N8MfmR26s8Uxl7z/zXtHi4J+Tu3bv06tWLffv2PXGR6fXr1xk4cCCZmbpZPZMmTaJXr15PdMzSqkKFCrRs2ZL9+/dz5swZbty4YVZFvFOmTGHRokUAVKxYkdmzZxfq8RkZGZw/r1ukrX79+tja2hq9jaJ4dejQgYMHD5q6GUKYLRmmMXA/4T7Xlkeyr88//N10FxdmX8qRiFjZaajydGWarWhCkyWPr834+++/dY+zsqJx48YAXLt2jT59+nDv3r0itzM9PZ2QkBB1J9DOnTszc+bMIh/PEhiuxhoaGmrCluR0+fJlFi9erF7+5JNP1I3SCur8+fNq0in1IkIIS1TmkxFthpaYX29xeNhR/vLfwam3znBnX85lWSu2cSNgfgM6n+1Is2VNqNK7MlY2+f/q4uLiOHHiBABNmzbl999/VzevO3r0KIMGDVI/YApr/Pjx7N+/H4Dq1auzdu1abGzKdieX4RRffRJoDmbMmJHjddZqtQwdOlSdplsQMpNGCGHpymwykhqZyumJZ7jY4zJHXz5OzNZbaDOyl353ruNM3Xfr0PFYO1pvaUn1oT7Yuha8e3zXrl3qzx06dKBKlSr89ttv6rb3v//+O6+//rq6n0xB/fDDD2qXv729PRs2bMDDw/TTi02tdevW2NnZAeaTjBw/fpzVq1cD4O7uriYSFy5cYOrUqQU+jiQjQghLV2aTETQQufw62kStepVdJTv8XvclODSIdvuCqf1WTRx9HIt0eMMPRP1W9/7+/mzevFn90Fy6dCkff/xxgY956NAhxowZo15evHgxgYGBRWqfpXFycqJVq1YAhIeHq+uumNLUqVPVZHPq1KnMnz9fLV5duHBhgWdXybReIYSlK7PJiKO3I25t3NA4aKg2sCot1gfS6WR7GsysT4XG5Z94USL9B42VlRVt27ZVr2/Xrh3Lli1TL7/77rusWrXqsceLjY1lwIABpKfr1jgZM2YMw4YNe6I2WhrDpeALs95Icdi1a5e6QZqPjw+jR4+mVq1azJo1S73P8OHDC1Q7pJ/W6+LiQvXq1YunwUIIYUJlNhkB+NeCBtT9Xy0afROAZyePx9aBFFRsbKz6bTYwMJAKFSrkuP35559nzpw56uVXXnkl36LLzMxMBg8erH7bb9OmDfPnzzdKWy2JYTJiyqEaRVGYMmWKevnDDz9Ue0TefPNNdev4q1evMmHChHyPlZSUxJUrVwDdBohWVmX6T1YIYaHK9JnNqYYTVk7G/xUYfivPa+O2yZMnM3r0aEC3BsmAAQNydMcbeuedd9RkpUqVKqxfv14d6hHZgoKCzKJu5JdfflF3EG7QoAEvvfSSepuVlRXLli1Tt4//7rvv1B6U3Bjuxiz1IkIIS1XoT+JZs2bRvXt32rdvz3PPPacWam7ZsoVWrVrx1FNPqf9u3rypPu706dMMHjyY4OBgRo4cSXR0tPGiMDO51Ys8TKPR8OWXX9K7d28gew2SGzdu5LjfunXr+PTTTwGwsbHh559/Nqs1NMyJo6MjrVu3BkxXN5KVlZWjOHX27NlYW1vnuE+NGjX4/PPP1cuvvvoq8fHx5MZw5VWpFxFCWKpCJyNDhgxhy5Yt7Ny5kxkzZjB9+nQSEhIA3ZBEWFiY+q9KlSqAbtGmyZMnM3jwYEJDQ2ncuDHTp083aiDmRF8vYm1tnaNe5GE2NjasXbtWLUKNjIykd+/eah3BqVOneOWVV9T7L1iwgODg4GJseeln6rqRVatWqb0ZQUFBPPPMM7ne77XXXqNHjx4AREdHM3bs2FzvJzNphBBlQaEXp/Dz81N/1mg0ZGZmEhsbm+9jDh8+jK2tLf369QNgxIgRdO7cmevXr+Pl5fXI/TMyMsjIyMjZUBsbow9N6Jdq1/9vDLdu3VI/jJo3b46zs3O+x3dycuKXX34hODiYiIgIjh07RkhICCtXrqR///4kJ+sWXRs6dCijR4/O81jFEYspFTUefT0G6JLCF154wajtyk9aWhozZsxQL8+ePRtFUVAUJdd4lixZQqNGjUhISGD16tX07duXkJCQHMc07Blp0KCBWby+8l4zX5YUC0g85qwwsRSk1q1IK2XNmTOHLVu2kJ6eTnBwMLVr1+bcuXOcPHmSzp07U7FiRZ577jn1xHr58uUc+2k4ODjg7e3N5cuXc01GfvzxR7777rsc1w0cOJBBgwYVpbmPZczufMPx/yZNmnD16tUCPW7JkiUMHDiQu3fv8r///Y/69etz545u8bWAgACmTJnCtWvXHnscc5jSakyFjadatWrY2dmRkZHBX3/9VeDfvzH88MMP6mvUvn17fH19H3n+h+OZMWOGur/Q6NGj8fPzw9Mze8+j48ePA7p1StLS0ko0nscp6+81c2ZJsYDEY84KEkuNGjUee58iJSNTpkxh0qRJHD58mPDwcDQaDc2aNeO///0vVapU4cyZM0ycOBE3Nzc6d+5MamqqWrCn5+zsnOcqlMOHD2fIkCE5G1pMPSORkZH4+PgYbZaCYbf6M888o666+ji+vr5s2rSJ7t27k5GRoSYi7u7u/PLLL489TnHEYkpPEk/r1q3ZtWsXV69excrKCh8fn2JqZbbExER1MTqA+fPn53jN8ornzTffZPfu3WzYsIE7d+4wc+ZMNm7ciEajITY2lri4OAAaNWpU4PdScZP3mvmypFhA4jFnxo6lyGuIW1tb07JlS9asWYOPj0+O2oiAgAAGDx7Mjh076Ny5M46Ojupwg15ycjJOTk65HtvOzq5EZ4tYWVkZ7Y2hr1OwtramXbt2hTpuhw4dWL58Oc8//7zarrVr1xYoq9QzZizmoCjxdOjQQS2sDgsL48UXXyyOpuUwf/58NXF44YUXaNq0aa73yy2eRYsWERYWRmxsLFu2bGHVqlUMGzaMs2fPqvcJCAgwu9dV3mvmy5JiAYnHnBkrlic+QlZWFlFRUY9cr9Fo1NUna9asyaVLl9Tb0tLSiIqKombNmk/69Gbl5s2b6gdIixYtcHFxKfQxBg8ezPfff0/Tpk358ccf6dKli7GbafFKer2RmJgYPvvsM0DXg/fRRx8V6vGenp4sWbJEvfzmm28SGRkpK68KIcqMQiUjSUlJ/P7776SkpJCZmcn27ds5dOgQTZs2Ze/everQwrlz5/jvf/+rFhMGBgaSnp7O5s2bycjI4IcffsDf3z/XepHSzHD2Rl5TegvilVde4ciRIwwdOtQYzSpzSnqfmlmzZqk9f6NGjSpSkt2vXz/19U5MTOSVV15RN1oEmUkjhLBshe4Z2bhxI7169aJz584sW7aMmTNnUq9ePQ4cOMCgQYNo27YtU6dOZejQoXTv3h3QDbvMmzePNWvW0LFjR44ePVrob4+lgeFeI3ktdiaKX0muN3LlyhW1VsTJyYlp06YV+VgLFy7E29sbgO3bt7Ny5Ur1toYNGz5ZQ4UQwowVqmbExcWFxYsX53rbf/7zH/7zn//k+diGDRuydu3awrWulNF/C7exsZH1QEzMsG5k586dxVY3MmPGDO7fvw/AW2+9pa6tUxSurq58//33ahKflpYGQPXq1SlfvvyTN1YIIcyUZVTQmIHo6GjOnz8PQMuWLR+ZPSRKVknUjZw4cYKffvoJ0M16mjhx4hMfs1u3brz++us5rpN6ESGEpZNkxEgKsgS8KDklUTcydepUtUh76tSpj2yIWFRz587NUXci9SJCCEtXppOR6Ohopk6dysWLF5/4WIYfeFIvYnrFXTcSFhbGr7/+CoCPjw9jxowx2rFdXFxYsWIFzs7OWFlZPbIqqxBCWJoym4zs3LmTevXqsXbt2hzbvReVvnjV1taWNm3aPPHxxJMrrn1qFEXJ8Z754IMPcHBwMNrxAYKDgzl//jxXrlyhefPmRj22EEKYmzKbjAQGBlKuXDkANm3aRFhYWJGPdf36dbV3pVWrVnku5iZKlmEyYjjT6Ult3bqVvXv3AuDv789LL71ktGMb8vLyonr16sVybCGEMCdlNhlxcXHhgw8+UC9PmDChyJsXyRCNeSqOupGsrCzeeecd9fLs2bOxsSnyQsZCCCEow8kI6PbAqVevHgAHDx4s8tRjKV41T4Z1I5cvXy7QRoOPs3r1ak6fPg3okp2+ffs+8TGFEKKsK9PJiLW1NVOnTlUvv/POO+raDoWhT0bs7OwICgoyVvOEERizbkRRFObNm6de/vjjj9FoNE90TCGEEGU8GQF46qmn6NatGwDXrl1j4cKFhXp8VFSUuu9O69atcXR0NHobRdEZc72R0NBQTp48CUBQUJAMyQkhhJGU+WQEYN68eequg7NnzyY2NrbAj5V6EfNmzLqRBQsWqD+PHz/+iY4lhBAimyQj6BaVGjFiBKDbpMywsPVxDGdpSL2I+TFW3ciFCxfYunUroFtXZMCAAUZroxBClHWSjDzw4Ycfqku4L1q0iHPnzhXocfpv2/b29uqHnjAvxqgb+eKLL9Sfx44dKzNohBDCiCQZeaBKlSq8/fbbgG765uTJkx/7mGvXrnH58mVANxxg7IWvhHE8ad3InTt3+PHHHwFwdnbm1VdfNVLLhBBCgCQjOUyYMIFq1aoBsGXLlsculCVTekuHJ60bWbp0KSkpKQC8/PLLuLm5GbN5QghR5kkyYsDJyYlZs2aplydOnJjvQmhSvFo6ODo6qlOuC1s3kpmZyZdffqleHjdunNHbJ4QQZZ0kIw8ZOnQoTZo0AeDIkSOsWrUqz/vqe04cHBxo1apVSTRPFFFR60Y2bNigbrLXp08f6tSpY+ymCSFEmSfJyEOsrKz47LPP1Mvvvvuu2kVvKCIigoiICEC35oTUi5i3otaNzJ8/X/35P//5jxFbJIQQQk+SkVx06tSJ3r17A7pFzQw/kPSkXqR0ad26Nfb29kDBk5H9+/ezf/9+ABo1aiSvsxBCFBNJRvIwb948rK2tAZgzZw43b97McbvUi5QuDg4OhV5v5OFFzmTpdyGEKB6SjOTB39+fkSNHApCUlMT777+v3qYoilov4ujoSMuWLU3RRFFIhakbiYyM5OeffwagUqVKPP/888XZNCGEKNMkGcnH+++/T7ly5QD47rvv1N1aIyIi1G/Wbdq0Ubv/hXkrTN3IV199RVZWFgCvv/661AQJIUQxkmQkH5UqVeKdd94BQKvVMmnSJECGaEqrgtaNJCcns2TJEkC3E/Prr79eEs0TQogyS5KRxxg/fjw+Pj4A/Pbbb/z555+yH00pVdC6keXLl5OQkADACy+8QOXKlUuqiUIIUSZJMvIYjo6OzJ49W708ceJENRlxcnKiRYsWpmqaKILH1Y1otVoWLlyoXpbdeYUQovhJMlIAL7zwAs2bNwfgxIkTREVFARAcHKwuMy5Kh8fVjfz2229cuHAB0PV6NW7cuIRaJoQQZZckIwVgZWXFp59++sj1Ui9S+jyubuTh6bxCCCGKnyQjBdS+fXv69u2b4zqpFyl98qsbOXnyJNu3bwegVq1a6sJ3QgghipckI4Uwd+5cbGxsAN1W8vqhG1G65DVUY1grMm7cOHXROyGEEMVLkpFCqFu3Ll999RU+Pj588skn2NramrpJoghyS0Zu3bqlbopYvnx5Xn755ZJvmBBClFE2pm5AaTNq1ChGjRpl6maIJ6CvG0lPT1eTkcWLF5Oeng7Aa6+9pi52J4QQovhJz4gocwzrRq5cucLFixf5+uuvAV2x8htvvGHK5gkhRJkjyYgokwyHasaMGUNMTAwAAwYMwM/PzzSNEkKIMkqSEVEmGSYj+hk0INN5hRDCFCQZEWWS4Xojei1atKBNmzYmapEQQpRdkoyIMsmwbkRv/PjxaDQaE7VICCHKLklGRJllOFRTrVo1Bg4caLrGCCFEGSbJiCizDFdYHT9+vKwbI4QQJiLrjIgyq0WLFqxfv55bt27J2jFCCGFCkoyIMi0kJMTUTRBCiDJPhmmEEEIIYVKSjAghhBDCpCQZEUIIIYRJSTIihBBCCJOSZEQIIYQQJlXoZGTWrFl0796d9u3b89xzz7Fr1y71tmXLltGlSxc6derEwoULURRFve306dMMHjyY4OBgRo4cSXR0tHEiEEIIIUSpVuhkZMiQIWzZsoWdO3cyY8YMpk+fTkJCArt372b9+vUsW7aMdevWsXfvXjZv3gxARkYGkydPZvDgwYSGhtK4cWOmT59u9GCEEEIIUfoUep0Rw+3VNRoNmZmZxMbGsm3bNvr374+3tzcAL774Ilu2bKFfv34cPnwYW1tb+vXrB8CIESPo3Lkz169fx8vL65HnyMjIICMjI2dDbWyws7MrbHPzpdVqc/xfmllSLCDxmDNLigUsKx5LigUkHnNWmFisrB7f71GkRc/mzJnDli1bSE9PJzg4mNq1a3PlyhW6d++u3qd27dqEh4cDcPnyZerUqaPe5uDggLe3N5cvX841Gfnxxx/57rvvclw3cOBABg0aVJTmPlZkZGSxHNcULCkWkHjMmSXFApYVjyXFAhKPOStILDVq1HjsfYqUjEyZMoVJkyZx+PBhwsPD0Wg0pKSk4OzsrN7H2dmZ1NRUAFJTU3Pcpr89JSUl1+MPHz6cIUOG5GxoMfWMREZG4uPjU6DMzZxZUiwg8ZgzS4oFLCseS4oFJB5zZuxYirwcvLW1NS1btmTNmjX4+Pjg5OREcnKyentycjKOjo4AODo65rhNf7uTk1Oux7azszN64pEfKyurUv/G0LOkWEDiMWeWFAtYVjyWFAtIPObMWLE88RGysrKIioqiRo0aXLp0Sb0+PDycWrVqAVCzZs0ct6WlpREVFUXNmjWf9OmFEEIIUcoVKhlJSkri999/JyUlhczMTLZv386hQ4do2rQpvXr1YsOGDURFRXH79m1++uknevXqBUBgYCDp6els3ryZjIwMfvjhB/z9/XOtFxFCCCFE2VLoYZqNGzcyZ84cFEXBx8eHmTNnUq9ePerVq0dISAjDhg1Dq9XSr18/+vbtC+iGXebNm8dHH33E3LlzadCgAR999JHRgxFCCCFE6aNRDFcmE0IIIYQoYZZRQSOEEEKIUkuSESGEEEKYlCQjQgghhDApSUaEEEIIYVKSjAghhBDCpCQZEUIIIYRJSTIihBBCCJOSZEQIIYQQJiXJiBBCCCFMSpIRIYQQQpiUJCOliKzcL4QQwhJJMlKKaDQaUzdBlAG3b982dRNEGXH//n1TN0GYCYtNRg4dOsTNmzcB0Gq1Jm7Nk9m2bRutWrXi0KFDpm6KUfzxxx8sX76cQ4cOqSej0trr89dffxEeHg5AZmamiVvzZLZt28ZTTz3Fzp07Td0Uo9i2bRvvvfcev/76q/oaleZzwe7du4mMjAQgKyvLxK15Mtu2baNNmzYW81777bffmD9/Prt27SIuLg4o3e81U5zXLG7X3lu3bjFz5kwOHjxIjx49eO+990zdpCKLiIjg3XffJSMjg1dffZXu3bubuklFpigKycnJzJo1izNnztCpUyfCwsJo164dw4YNo0KFCiiKUmp6f+7cucOcOXMIDQ3Fy8uLTZs2mbpJRXb9+nUmTJiAVqtlxIgRpf59lpmZyRdffMHu3bvp168fp06d4saNG3z22WdUqVLF1E0stLi4OGbPnk1YWBiBgYEsWrTI1E0qssjISN5++20yMzMt4r2Wnp7O3LlzOXLkCD179uTgwYO4u7szefJk3N3dS9U5DUx7XrOonpGsrCxOnDiBm5sbkydPJiYmhn379qm3lSZJSUksX76c1NRU1q9fT/fu3UlPTy91cehpNBpu3brF7du3WbduHePGjePtt9/m2rVr/PDDD6ZuXqHdvn0bOzs7VqxYwf379/n555+B0tc7otVqOXz4MOHh4axZs4bu3buTkpJCQkKCqZtWJBqNhuTkZE6ePMlXX33FsGHD+PDDD/H19S21X0wiIiJwcXFh/vz5JCYm8scffwCl75yWnp7Otm3biImJYd26dXTv3p20tDTS0tJM3bQi0Wg0JCQkEBERwbJlyxg1ahTTpk1Dq9Xy6aefmrp5RWLK85pNsT9DCbK2tqZ+/fo0aNAAR0dHYmJiWL9+PUFBQVhbW5eqLNXFxYXg4GA0Gg1//vknMTEx7N27Fzc3N3r27ElgYCCOjo6lKqZjx45hbW2NRqNBq9XSqFEj1q1bx9atW+nVqxf16tUz63gSExMpX748AH5+fowYMQI/Pz/GjBnD3LlzCQkJwcbGxqxj0NPHYmVlRXBwMB07dmThwoVUq1aNX375BTc3Nxo1akTv3r3x9vZGq9ViZWW+310MX5u4uDjs7e2JiYnBy8sLR0dH6tevz5dffsnOnTtp37692ceTkZGBnZ0dALVr1+a1116jatWqXLt2jR9//JHu3buXmnOaPhZ7e3uCg4OJiopi1apVaDQa/vrrL1xdXWnXrh3t2rWjYsWKZh+T4WsTHh6Ora0tSUlJuLq64u3tjaenJ+vXr2fw4ME0btzY7N9r5nJeM9/fUAHs2LGDF154gVmzZrFy5UoAvL29qVatGm5ubrRp04bMzEw1uzPnMTzDWFasWAFAmzZtcHNzY/bs2YSGhjJ06FBcXFxYv349GzZsAMy3qDW3eFq1asWZM2e4dOkSVlZW2Nvb4+rqSpMmTdi4cSNgnvGEhYXx9NNPM23aNGbNmkVcXBw2Njb4+fkB0Lt3b3x9fZkzZw5g3t9YDWOZOXMm8fHxuLu7079/f9avX88ff/zBjBkz6Nq1K1FRUXz++ecAZnsyfTiexMREqlatir29PRcuXFCLcTMzM2ndujXffPMNYL7x/P333wwYMID33nuPhQsXoigKrq6u+Pj4YGNjQ7du3ShfvjxLliwBzPu99nAsAP7+/jRu3JjFixezfft2xo0bh7+/P3v37lV7SM3xHAC5x1O/fn3u3LnDxYsXSUtLw8bGBkdHR9q0aaPGY67vNbM7ryml1MGDB5VevXopmzZtUrZu3ap07txZWb16tXLnzh31PikpKcratWuVkSNHKrGxsTker9VqS7jFecstlpUrVyrp6enKpUuXlF9//TXH/desWaN89NFHSmJioolanL/c4lmxYoWiKIqyaNEiZfDgwcrMmTOV119/XXnttdeUjRs3KnPnzlUyMzPN6nVRFEWJiIhQBg4cqPz222/KhQsXlFGjRinvv/++cvbsWUVRFOX+/fuKoijKyZMnlRYtWiiRkZE5Hm9O8eQVy4ULFxRFUZSwsDAlPT1dvf/ly5eVV199VTl16pSpmpyv3OKZMWOGEh0drezdu1cZM2aM8uqrryojR45UBg0apFy4cEEZO3bsI6+RuThz5ozyzDPPKL/++quyf/9+5dlnn1W+/PJLtb3691JoaKjy9NNPK9HR0Tkeb07vtdxiWbhwoRITE6PcuXNH2b59e477h4WFKRMnTixVr82CBQuU1NRUZePGjcrw4cOVCRMmKKNHj1ZefPFFZc+ePcq0adOUu3fvmrrpuTLH85p5pmwFcOrUKdq2bUvfvn3p3bs3U6ZMYd++fezevVu9j6OjI4GBgXh6erJ161YAtm7dilarNavsO7dYDhw4wPbt26lVqxadOnUCICUlBQBnZ2cOHz6Mra2tKZudp9zi2b9/P3/++SejRo1iwoQJVK5cmZYtW7JkyRLKlSvH2bNn1SEcc3Lr1i1cXV3p0KEDderUYcKECdjZ2bF27VoAbGxs0Gq1BAQE0K1bN3Ws+MsvvyQzM9Os4sktFltbW1avXg1AUFAQdnZ2pKenA5Camsrdu3dxcXExZbPzlNdrs3TpUoKCgvjggw949tln6dGjB//973+xs7MjMTGRihUrmrrpubp69Sr16tWjV69etGrViilTpnDjxg22bdsGZPcYBAYG0qhRI5YtWwbAqlWryMrKMqv3Wm6xREdH88svv6jDMkCOepHw8HCcnZ1N1eR85RXPmjVr6NevH1OnTqVFixa0a9eOlStX4ujoyNWrV9XhD3Njjue1UpeMKA8m/6SmpqpTjwC6dOlCrVq1OHbsGBEREer1tWvXpkuXLixfvpwWLVpw/PhxsxmuyS+W2rVrc/jwYSIiInBwcADAyckJ0BW0derUSb3eXOQXT926ddm3bx8RERE0b96cV199lZdffhnQ1ZL07NnTFE1+rISEBFJSUtTEr06dOrRr1474+HhCQ0OB7LinTJnCnj17aNmyJWfPniUzM9OspiznFkv79u25ffs2oaGhWFtbA2Bvbw/AgQMH8PX1xdPT02Rtzk9u8XTo0IGYmBi2b9+Oh4cHPXr0oH///gD8/vvvBAQEqPGZm4SEBKKiotTLzZs3p2nTply5coWjR4+q15cvX56hQ4fyf//3f7Ro0YJTp06RlZVldu+1h2Np1qwZ4eHhHD16VH3N9Oew06dP07RpU8qVK2eS9j5ObvEEBgZy7tw5jhw5Qu3atXnuued4/vnnAdizZw9PPfWUqZr7WOZ4Xit1yYg+I+vSpQt3797l+PHj6m3du3fn6tWrREdHq9edPHmSKVOm4Ofnx7Jly3j33XexsTGPut3HxXLt2jU1lgsXLvDzzz8zYMAADh48yDPPPGOSNufncfFERkaq8ejf9L169eL48eO0adPGJG02lNsfWNeuXblx4wbbt29Xr6tfvz41a9bk5MmTZGZmYm1tzalTp+jatSsNGzZkxYoVfPPNNzg4OJjs22pRYsnKyuLatWts2rSJvn37qj1Z+iTYlAoTT61atTh9+jRZWVmkpaWxa9cuevbsqY7565MuU8otnj59+hATE8P+/fvV61q3bo1GoyE8PFx9zKlTpxg6dCgBAQGsWLGCOXPmYGdnZ1bvtcfFArqekA0bNtCvXz927tzJkCFDzOLcXJh4rK2tuXz5MgD37t3j77//plevXuzdu5du3bqVWJsLyxzPa2abjMTHx3PmzJlHimb0b5SKFSvSsmVL1q9fr95Wv359XFxc1DdMZmYmt27dYuLEifz44480bNiw5AIw8CSx6KcmOzg4cPDgQYYNG8aKFSvw9fUtuQAeYozXxs7Ojrt37zJmzBhWrFiBl5dXyQVgICEhQU2QHv4D0y/INnToUObPn69e7+7ujouLCzdv3lRPnlWrVmXChAksW7aM+vXrl1Drc3rSWKytrXF2diY+Pp5Ro0axevVqateuXXIBPMQY8djZ2VGxYkXGjx/PmjVrqFmzZskF8JCEhASuX78OPBpPVlYWLi4uDBgwQC1OBfDx8cHFxYWzZ8+qj9FoNOo5zd/fv+QCMPAksZw5cwbQDTeHh4fz6quvsnr16lL92gBqQjh27FhWr15t0nO0fqg1t6JTcz2vmWUy8tlnnzF48GDmzJnD+++/z4kTJwDdL1H/RnF3d6ddu3ZER0erM0tANyyjn3ZlY2ND586dCQkJKfkgHnjSWOzt7VEUherVq/PJJ5/Qt29fk8ShZ6zXxsXFhf79+9OnT5+SD+KBzz//nOeee47p06fz5Zdfcv78eUA3dQ9QuzBfeOEF3Nzc+Pzzz9W6HW9vb+7fv68O+bm7u5v0fWaMWDIzM3F3d+eVV16hV69epgnkAWPEk5WVhZWVFQEBASZfXOuzzz5j4MCBTJ06lS+++IKLFy8C2fHoe2v69OlDSkoK33//vfrYGjVqoCiKmuw3bNjQ5Oe0J4kFdDMbq1SpwqRJk0x6DgDjvTb29va0b9/e5EPOX3/9NUOGDCEpKQlra2v1HKX/31zPa2aXjKxfv57Tp0/zyy+/MG3aNMqXL68u8qP/Ja5evZo5c+ZQu3ZtBg4cyMKFC1m2bBmrVq1i8+bNBAYGmjIElbFiMZfCNEt6bTZt2sS5c+fYsGEDY8aMIT09ndmzZwOoCdPatWsZPXo0t27d4sMPP+Sff/5h2rRpLF68mE8++YS2bduaxbQ9Y8ViDl3kYLx4zGE4BmDv3r2cPn2a9evX8+9//5uUlJRH4lmzZg1TpkzBycmJiRMnsmbNGj755BN++OEHli5dqq45ZGrGisUc/m7Asl6b1NRU5s2bx969e3FxceHrr78Gsnt69L9zcz2vmcU7wrCi+ubNm3h5eeHg4EDdunVxcXFRK5KTkpJ48803+e9//0uHDh1wd3enR48evP322yQmJrJr1y4+/PBDWrdubapQLCoWsKx49LFotVoiIiKoV68ezs7ONGvWjJCQEKKioli6dCkACxcuZPXq1bz88st4e3tTp04dPvjgA9q2bUtUVBSzZ8+mX79+EouRWFo8+m+aoFsC3draGldXV1q2bMmrr75KVlaW2kU+Y8YM1q5dS79+/fD09KR58+bMnDmTypUrc/LkSWbNmkXnzp1NFYpFxQKWF09ycjKgSzqCgoKYOHEi48aNY8+ePZw/fx6NRkNmZiYZGRl89dVX5vu3Y/TJwoUQHx+vvP3228r8+fPV67777jvl/fffV3bv3q1kZGQow4cPV95//31l3759SlZWljoPWlEUJTMz0wStzp0lxaIolhVPbrF8+OGHyqJFi5S0tDRFURQlOjpa6d27t9KlSxfl7t27OdalycrKKukm58mSYlEUy4xn4sSJypQpU9S1GtavX6/MnDkzR7v379+vtGvXTrl7965y7do19Xpz+7uxlFgUxbLj0bctNTVVURTdOiFz585Vxo4dm+MxV69eVX82t78dk/WMLF68mH79+uHo6Mi///1v9fp+/foRHBzM8uXLadeuHV5eXnh5ebFixQo+//xztYhGX+1rDiwpFrCsePKKpWvXruzevZtly5YRGxvLhg0b6Nq1K61bt+aXX37Bw8MDQK070FNMOH3SkmIBy4vnzJkzDBs2jHLlyjFixAh13L127dqcPn06x9TQJk2a0Lp1a1auXImPjw+gi8fw78aU8VhSLGD58ejbo58qbW1tTUhICNHR0fz555+ALobq1asDunO0Of3tgIn2pvn+++9ZvXo1s2fPJjg4GEBd997Dw4MuXbpw9epVAgMDGTVqFKDbonnjxo0kJCTg6upqNuPblhQLWFY8+cXSunVrbt68yYEDB5g4cSIODg5Mnz6dn376icqVK6vHeDipMtXYsCXFApYXD8Dx48cJDg7m7bffBnRDl3Z2djRp0gRfX182bNiAt7c3Hh4eWFtb4+XlRVZWlppUmVM8lhQLWH48ycnJOc67Go0GX19f+vbtyw8//EDXrl2xtrbm3r17lCtX7pGaEFPHAyWYjOhPNADt2rXj0KFDODk5cfLkSdauXYuXlxeVKlUiJCSE1NRUDhw4QPfu3dXHRUVF4evri6ura0k1OU+WFAtYVjwFicXDw4NBgwbRr18/+vXrR3R0NFWrVgV0dTHmsrKtJcUClhuPoijcv3+fa9euERwcTGRkJDNmzMDd3R03NzfGjh3LW2+9xfjx4/nf//5H9+7dcXd35969e1SrVs0sehEtKRYom/F4enoyZMgQvL291cf169eP/fv388EHH6gLTv773/82mwJiQ8WejKSkpPDNN99ga2tLmzZtCAgIoE6dOgQEBDB9+nTu37/P008/jb29PYsWLSIzM5PBgwfTsmVLtm3bxs2bN4mLi2PPnj288847xd3cMhOLpcVTmFiWLFmCRqOhV69eODs7U7lyZaKjo/niiy+Ii4sjICBAYpF4ChRPcHAwDRs2xNHRkbi4OH799Vc8PT1p27Ytbdq0YeHChSxYsIA333yTUaNGsXHjRrZv346npydHjhxh3rx5EovEY7R4li9fzvPPP0/NmjWxsrIiPT2d+Ph4Tpw4wZAhQ9TebHOkUYpxsOjixYtMmTKFBg0a4OHhwdmzZ6latSrvvfceiYmJLF++nGeffZZq1aoBsH37dlauXMm3336LnZ0dO3bs4NixYwC8/vrrJt0jw5JiAcuKpyixrF69ms8//xxXV1cuXbrEnDlz8PDwYPr06SbdH8OSYilr8Zw+fZoxY8bQsGFDdXfgixcvsmLFCrp168ZTTz3F7du32bNnD7du3WLw4MFm+3dT2mKReHT3X7lyJR07dqRjx46kpaUxbNgwKleuzJw5c8xiJeV8FWd17M8//6xMnjxZvXzlyhUlKChI+euvvxRFUZSEhARFURS1av7evXtKhw4dlEOHDqmPMZcKZkuKRVEsK56ixnLkyBH1esPdnk3JkmJRlLITz44dOxRFUZS3335bCQkJURQl++/j+eefVzZu3FjSTX0sS4pFUSQeRXk0HnPdNTg3Rh04iouL49q1a4CucvfevXs4Ozur87rd3d3x8PDgu+++A6BChQpA9sZcx48fp0mTJtSrV089pqnG7CwpFrCseIwVS506ddTrTVXvYkmxQNmMx93dXV0mfMyYMdy4cYNVq1aRlJREbGws5cqVM+my+nqWFAtIPAWJx1x3Dc6NUZIRRVH49ttveeaZZ1i3bh2JiYlYW1tTrlw5UlNTOXLkCKCbTtSqVStiY2PZtGkTAImJifzzzz/MmjWLGTNm0KZNG5N2j1lSLGBZ8Ugs5hkLlO14WrduTUxMDJs2baJ69erMnDmTw4cPM3HiRAYNGkTTpk1NWutiSbFIPOYfT1EZpWZk//79bN26Vf3lBQcH07VrVzIyMvjmm2/Yu3cv9evXJywsjFGjRqHRaLh58ybjxo0jPT2duXPnkpyczKRJk3B3dzdGXBKLBcYjsZhnLBJPznhA98Fx4cIFvLy81N4ficU4JB7zjqeojJKMpKWlcenSJfz8/Pj2228BeP755/H29ubevXtcvXqVM2fOULduXZo0acLMmTPx8/PjxRdfVB+vX6zF1CwpFrCseCQW84wFJB7DeLRarVlNnbSkWEDiMfd4isooUTg4OBAQEICLiwvt27cnPj6eAwcOAFCuXDkCAgIYNGgQTZo0ISYmhhs3btCgQYMcjzcXlhQLWFY8Eot5xgISj2E85vbhYEmxgMRj7vEUldEjadmyJX5+fpw+fZrTp0+r18fHxzN79mwGDBhA3bp1adasmbGf2ugsKRawrHgkFvMl8ZgvS4oFJB5LYtRkRD/i07lzZzIyMjh16hQAZ8+e5f79+zRq1Ih169Yxfvx4Yz5tsbCkWMCy4pFYzJfEY74sKRaQeCyN0Rc9Ux4sW/vbb7+xYcMGzpw5Q7169dRFjEoTS4oFLCseicV8STzmy5JiAYnHkhh9OXiNRkNaWhrr16/n8uXLvPnmmzz33HPGfpoSYUmxgGXFI7GYL4nHfFlSLCDxWJJi2Ztmz5491KtXj2+//VZdzKi0sqRYwLLikVjMl8RjviwpFpB4LEWx7E2j72qyBJYUC1hWPBKL+ZJ4zJclxQISj6Uo1o3yhBBCCCEex3ImKQshhBCiVJJkRAghhBAmJcmIEEIIIUxKkhEhhBBCmJQkI0IIIYQwKUlGhBBCCGFSkowIIYQQwqQkGRFCmMTIkSNp3rw5I0eONHVThBAmJsmIEEIIIUxKkhEhhBBCmFSxbJQnhBCGEhMTmT17NmFhYbi6ujJ8+PBH7rN161ZWr15NVFQUAJUqVcLf35+PPvqopJsrhChhsjeNEKLYTZo0iR07dgDg6+vLrVu3AEhNTaVZs2ZMnDiRIUOGoCgKPj4+2NnZER0dTUpKCocOHTJl04UQJUB6RoQQxSoqKkpNRIYNG8bYsWOJiIjgueeeU+8TGRmJoihUr16dn3/+GSsrK7Kysjh+/Lipmi2EKEGSjAghilV4eLj6c6dOnQDw8/OjTp06nDt3DoDGjRtTvnx5rl27RufOnalevTp169alZ8+eJmmzEKJkSQGrEMLkPDw8WLduHWPHjqV169akpKSwceNGRo8ezalTp0zdPCFEMZOeESFEsapZs6b6899//03Dhg25evUqFy9eVK+PjY0lISGBYcOGqdeFhIQQERHBsWPHCAgIKNE2CyFKliQjQohi5ePjQ4cOHfj777/58ccf2bFjBzExMVhbW5OVlQXA5cuX+fe//42bmxuenp4kJydz/fp1AGrXrm3K5gshSoAM0wghit306dPp1KkT9vb2JCUlMWrUqBy9HV5eXnTr1g1nZ2euXr3KnTt3qFu3Lu+++y6tW7c2YcuFECVBpvYKIYQQwqSkZ0QIIYQQJiXJiBBCCCFMSpIRIYQQQpiUJCNCCCGEMClJRoQQQghhUpKMCCGEEMKkJBkRQgghhElJMiKEEEIIk5JkRAghhBAmJcmIEEIIIUxKkhEhhBBCmNT/A5z7yOYMQGwSAAAAAElFTkSuQmCC",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -300,20 +300,211 @@
     }
    ],
    "source": [
-    "# train_air.plot(label=\"train\")\n",
+    "train_air.plot(label=\"train\")\n",
     "test_air.plot(label=\"test\")\n",
     "preds_simple.plot(label=\"trend & season\")\n",
     "preds_ar.plot(label=\"auto-regression\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Multivariate"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "complete_ts = AustralianTourismDataset().load()\n",
+    "ts = complete_ts[[\"NSW\", \"VIC\", \"QLD\", \"SA\", \"WA\", \"TAS\", \"NT\"]]\n",
+    "ts.plot()"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 9,
    "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 391\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f63c65898e8a46359badf46e5e392a79",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/110 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-01, min: 1.26E+00\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b4c803be617149dfa126e172663d2886",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/110 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-01, min: 1.04E+00\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 1.04E-01\n",
+      "Epoch[391/391]: 100%|██████████| 391/391 [00:03<00:00, 98.15it/s]\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# model supports only datetime indexing\n",
+    "ts = TimeSeries.from_times_and_values(\n",
+    "    pd.date_range(start=\"2000-01-01\", periods=len(ts), freq=\"D\"), ts.values()\n",
+    ")\n",
+    "train, test = ts[:-4], ts[-4:]\n",
+    "model = NeuralProphetDarts(\n",
+    "    yearly_seasonality=False,\n",
+    "    weekly_seasonality=True,\n",
+    "    daily_seasonality=False,\n",
+    "    n_lags=len(test),\n",
+    "    n_forecasts=len(test),\n",
+    ")\n",
+    "\n",
+    "model.fit(train)\n",
+    "preds = model.predict(len(test))\n",
+    "\n",
+    "preds.plot()\n",
+    "test.plot()\n",
+    "train.plot()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Past covariates"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "ice_heater = IceCreamHeaterDataset().load()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -323,12 +514,1411 @@
     }
    ],
    "source": [
-    "air_year = datetime_attribute_timeseries(series_air, attribute=\"year\")\n",
-    "air_month = datetime_attribute_timeseries(series_air, attribute=\"month\")\n",
-    "air_covariates = air_year.stack(air_month)\n",
-    "scaler = Scaler()\n",
-    "air_covariates = scaler.fit_transform(air_covariates)\n",
-    "air_covariates.plot()"
+    "train, test = ice_heater.split_after(split_point=0.8)\n",
+    "\n",
+    "year = datetime_attribute_timeseries(train, attribute=\"year\")\n",
+    "month = datetime_attribute_timeseries(train, attribute=\"month\")\n",
+    "ice_covariates = month.stack(year)\n",
+    "\n",
+    "scaler_dt_air = Scaler()\n",
+    "ice_covariates = scaler_dt_air.fit_transform(ice_covariates)\n",
+    "\n",
+    "ice_covariates.plot()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.forecaster.add_lagged_regressor) - n_lags = 'auto', number of lags for regressor is set to Autoregression number of lags (40)\n",
+      "INFO - (NP.forecaster.add_lagged_regressor) - n_lags = 'auto', number of lags for regressor is set to Autoregression number of lags (40)\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 413\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "86bf965d96784d6c8d67494053139ce4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/110 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 6.02E-02, min: 1.88E-01\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "40f6c7fc9a824898a8a1c7ef7378dde4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/110 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 5.09E-03, min: 1.88E-01\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 3.13E-02\n",
+      "Epoch[413/413]: 100%|██████████| 413/413 [00:05<00:00, 81.31it/s]\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "model = NeuralProphetDarts(n_lags=len(test), n_forecasts=len(test))\n",
+    "model.fit(train, ice_covariates)\n",
+    "preds_cov = model.predict(len(test))\n",
+    "\n",
+    "preds_cov.plot()\n",
+    "train.plot()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 413\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1e5d52c456314ce385d0d19f73f4203d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/110 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.32E-02, min: 3.40E-02\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "238f395a39f64f70a2a48a9ca58293f8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/110 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.32E-02, min: 3.33E-01\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.59E-02\n",
+      "Epoch[413/413]: 100%|██████████| 413/413 [00:04<00:00, 89.92it/s]\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "model = NeuralProphetDarts(n_lags=len(test), n_forecasts=len(test))\n",
+    "model.fit(train)\n",
+    "preds_no_cov = model.predict(len(test))\n",
+    "\n",
+    "preds_no_cov.plot()\n",
+    "train.plot()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "12.828873303132173\n",
+      "12.877951456793664\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(mape(test, preds_cov))\n",
+    "print(mape(test, preds_no_cov))\n",
+    "# no improvement in this case but it shows that fitting and training works for multivariate time series and multivariate past covariates"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.forecaster.add_lagged_regressor) - n_lags = 'auto', number of lags for regressor is set to Autoregression number of lags (40)\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 627\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fa455f96c51b4ba6af70833fdf223ed4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/106 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 7.69E-03, min: 1.69E-02\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d1fb1758fe784c05a0735623cdafc24c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/106 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 7.69E-03, min: 1.69E-02\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 1.81E-02\n",
+      "Epoch[627/627]: 100%|██████████| 627/627 [00:04<00:00, 152.45it/s]\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
+      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
+      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 627\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "16160f58de794b048e7d14108e7553f7",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/106 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 2.68E-01, min: 3.98E-01\n"
+     ]
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "af53cc0bb3664f85818ef12cb4e022c5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/106 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.00E-01, min: 5.91E-01\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 1.39E-01\n",
+      "Epoch[627/627]: 100%|██████████| 627/627 [00:03<00:00, 162.89it/s]\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
+      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n",
+      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
+      "  df_forecast[name] = yhat\n",
+      "\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "MAPE with lagged regressor:  57.026634304236026\n",
+      "MAPE without lagged regressor:  12.331045940870737\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhwAAAG8CAYAAACVCnwPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAACkMUlEQVR4nO2dd2AUZf7Gn63JpjdIKCEBQpEi0ixYsCBIsdcTT8Re7uSKp56e5fRnO8vJnVc89UBP9Ozl7Ao2bAiKIp1AQggkgfS6dX5/bObdd2Znd2d2Z7bx/fzDZLOZmXc35H32+TaTIAgCCIIgCIIgDMSc6BsgCIIgCCL9IcFBEARBEIThkOAgCIIgCMJwSHAQBEEQBGE4JDgIgiAIgjAcEhwEQRAEQRgOCQ6CIAiCIAyHBAdBEARBEIaT9oLD5/Nh165d8Pl8ib4V3aG1pSa0ttSE1paapOvaUnFdaS84CIIgCIJIPCQ4CIIgCIIwHBIcBEEQBEEYDgkOgiAIgiAMhwQHQRAEQRCGQ4KDIAiCIAjDIcFBEARBEIThkOAgCIIgCMJwSHAQBEEQBGE4JDgIgiAIgjAcEhwEQRAEQRgOCQ6CIAiCIAyHBAdBEARBEIZDgiNFqKysxKOPPpro2yAIgiCIqLAm+gbSmeOPPx6HHXaYLkLh22+/RXZ2duw3RRAEQRxUNK9uQeO7TbA4LBh0Zhnyxucm5D5IcCQQQRDg9XphtUZ+GwYMGBCHOyIIgiDSjfb17aj5Zy0AIG9ibsIEB4VUDOKSSy7Bp59+iqVLl8JkMsFkMmH58uUwmUx49913MXXqVGRkZGD16tWorq7G6aefjtLSUuTk5GD69On46KOPJOeTh1RMJhOefPJJXH311cjJycGoUaPw5ptvxnmVBEEQRLLj7fWyY0uWJWH3kbIOx7Rp09DQ0KDquV6vFxaLPi9yWVkZ1q5dG/F5S5cuxbZt2zBhwgTcddddAICNGzcCAG6++WY89NBDGDFiBAoLC1FXV4d58+bhnnvuQUZGBp555hmceuqp2Lp1K4YNGxbyGnfffTduuOEG/PWvf8Xf/vY3LFy4ELW1tSgqKtJlrQRBEETq4+3mBIeDBIdmGhoaUF9fn+jbCEl+fj7sdjuysrJQVlYGANiyZQsA4K677sLJJ5/MnltUVIRJkyaxr++++2689tprePPNN/GLX/wi5DUWLVqE0047DRUVFbj33nvxl7/8BWvWrMEpp5xi0KoIgiCIVMPb62PHFkfiAhspKzjETVwNejscsTJt2jTJ111dXbjzzjvx9ttvY9++ffB4POjt7cXu3bvDnmfixInsODs7G3l5eWhqaor5/giCIIj0QRJSyU7ctp+ygkNNWAMAfD4famtrUVFRAbM5OVJW5NUmN9xwAz788EM89NBDqKqqgsPhwDnnnAOXyxX2PDabTfK1yWSCz+cL8WyCIAjiYMTbw4dUyOFIS+x2O7xeb8TnffHFF7jkkktw5plnAvA7HjU1NQbfHUEQBHEwIHE4EpjDkRwf+dOUyspKfPPNN6ipqcGBAwdCug+jRo3Cq6++ivXr1+OHH37AhRdeSE4FQRAEoQsShyOBVSokOAzkhhtugMViwbhx4zBgwICQORmPPPIICgsLMWPGDJx66qmYM2cOpkyZEue7JQiCINIRaUiFqlTSktGjR+Orr76SPHbJJZcEPa+yshKrVq2SPHbddddJvpaHWARBYPkpIm1tbTHdL0EQBJF+iCEVc4YZJospYfdBDgdBEARBpDFiWWwiwykACQ6CIAiCSGvEkEoiwykACQ6CIAiCSGsCgiOxWz4JDoIgCIJIY8QcDks2ORwEQRAEQRiAz+2D4BYAUEiFIAiCIAiDSJamXwAJDoIgCIJIW7w93OA2qlIhCIIgCMIIJA4HCQ6CIAiCIIwgWQa3ASQ4CIIgCCJtSZY5KgAJDoIgCIJIWyhp9CDC5/PhT3/6E6qqqpCRkYFhw4bhnnvuAQBs2LABJ554IhwOB4qLi3HllVeiq6sLAPDBBx8gMzMzaD7KkiVLcOKJJ8Z7GQRBEEQKkiyD2wASHIbz+9//Hvfffz9uu+02bNq0Cc899xxKS0vR3d2NOXPmoLCwEN9++y1eeuklfPTRR/jFL34BADjppJNQUFCAV155hZ3L6/XihRdewMKFCxO1HIIgCCKFkDgcCW78lbLTYqdd4UNDi4onCoDXOwQWCwCTL+LTI1FWBKx9Qp1O6+zsxNKlS/HYY49h0aJFAICRI0fimGOOwRNPPIG+vj4888wzyM7OBgA89thjOPXUU/HAAw+gtLQUF1xwAZ577jlcdtllAICVK1eira0NZ599dszrIAiCINIfSVlsgh2OlBUcDS1A/X61z07MMjdv3gyn04mTTjpJ8XuTJk1iYgMAjj76aPh8PmzduhWlpaVYuHAhjjzySOzduxeDBw/GihUrMH/+fBQUFMRxFQRBEESq4u3xsGMSHFFSVqTyiQLg9XpgsVgBUxyvC8DhcMR0renTp2PkyJH473//i2uuuQavvfYali9fHtM5CYIgiIMHcTQ9kPgqlZQVHGrDGj6fD7W19aioqIDZHN+UlVGjRsHhcGDlypW4/PLLJd875JBDsHz5cnR3dzOX44svvoDZbMaYMWPY8xYuXIgVK1Zg6NChMJvNmD9/flzXQBAEQaQu0sZf1IcjbcnMzMRNN92EG2+8Ec888wyqq6vx9ddf46mnnsLChQuRmZmJRYsW4aeffsLHH3+MX/7yl/j5z3+O0tJSdo6FCxfiu+++wz333INzzjkHGRkZCVwRQRAEkUokU5VKyjocqcJtt90Gq9WK22+/HXv37sWgQYNw9dVXIysrC++//z6WLFmC6dOnIysrC2effTYeeeQRyc9XVVXh8MMPx5o1a/Doo48mZhEEQRBESpJMjb9IcBiM2WzGrbfeiltvvTXoexMnTsSqVasinuObb74x4tYIgiCINIcafxEEQRAEYTjJ5HCQ4CAIgiCINCWZqlRIcBAEQRBEmuJLoqRREhwEQRAEEWd2PFyNVZM+xb43Gwy9Dp/DYc6ksliCIAiCOGjweXzY/mA1+vb0ofrPOw29lqff4bBkWWAy69D9MgY0C46tW7fi0ksvxcyZM3H66afj9ddfB+BvsPXwww/j+OOPx+zZs7FixQq975UgCIIgUh5nowuCWwAA9OzqhSAIhl3L1+9wWByJ9xc0l8XefvvtmDVrFp588kls27YNV155JSZNmoS1a9di3bp1ePXVV9HV1YWrrroKo0aNwuGHH27EfRMEQRBESuLc18eOPZ0eeNo9sBXYDLmWGFIxJzh/A4hCcOzbtw9z5syB2WzG2LFjUVlZiZqaGrzzzju46KKLUFRUhKKiIpxxxhl4++23FQWHy+WCy+WS3ojVCrvdHv1KQuDz+ST/phO0ttSE1paa0NpSk2RcW8/eXsnX3TXdyDs0T9M51K7L0x0IqRj1GqgdG6JZcJx//vl49913cdlll2HLli1obGzExIkTsXPnTowaNYo9r6qqCqtXr1Y8x7Jly/DEE09IHjv33HNx3nnnab0d1dTV1Rl27kRDa0tNaG2pCa0tNUmmtbVsbJV8XbtuN/Lyc6M6V7h1CYLAHA6vxYPa2tqorhGJ4cOHq3qeZsExY8YM3HHHHfj3v/8NwN+6u6SkBL29vZJR69nZ2ejp6VE8x+LFi7Fw4ULpjRjocNTV1aG8vDzuw9uMhtaWmtDaUhNaW2qSjGvrc7oA7Gdf5/TloqKiQtM51KzL5/Jhi3c7AMCR79B8Db3RJDja29vxq1/9CrfddhtOOOEE7Ny5E7/85S9RVVUFh8OB7u5u9tzu7m5kZWUpnsdutxsiLsJhNpuT5pdNb2htqQmtLTWhtaUmybQ2Z4NT8nXf7r6o7y3curx9gZJYa7Y14evXdPU9e/YgMzMTs2bNgsViwahRo3DooYdi3bp1GDFiBHbs2MGeW11djREjRuh+wwRBEASRyjj3SQVHb11viGfGBt/W3JwEVSqa7qCiogJ9fX345JNPIAgCdu7cifXr16Oqqgpz587Ff/7zH7S2tqKurg6vv/465s+fb9R9pwTHH388rr/+etx4440oKipCWVkZ7rzzTslzdu/ejdNPPx05OTnIy8vDeeedh8bGxrDn3bNnD372s5+hpKQE48ePx+GHHy4Z8PaPf/wDI0eOhN1ux5gxY/Cf//yHfe/CCy/E+eefLzmf2+1GSUkJnnnmmdgXTRAEQYSlj6tSAYCe3QYJjiQa3AZoDKnk5OTg/vvvx1//+lfcfvvtyMvLw4UXXogjjjgC06dPR11dHc4880zYbDYsWrSISmIBPP300/jNb36Db775Bl999RUuueQSHH300Tj55JPh8/mY2Pj000/h8Xhw3XXX4fzzz8cnn3yieL6uri7MnDkTQ4YMweuvvw6v14vGxkaWffzaa69hyZIlePTRRzFr1iy89dZbWLx4MYYOHYoTTjgBCxcuxLnnnouuri7k5OQAAN5//3309PTgzDPPjNfLQhAEcVAiCAL6ZCGV3t3+Xhwmk76NuSSD27JTTHAAwFFHHYWjjjoq6HGz2Yzf/va3+O1vf6vLjUVi9YlfwdXkjPg8AYDX68VOSy30eCvtAzNwzKrg9Yfi0EMPxR133AEAGDVqFB577DGsXLkSJ598MlauXIkNGzZg165dKC8vBwA888wzGD9+PL799ltMnz496HzPPfcc9u/fj2+//RYFBQWora3FzJkzWWzuoYcewiWXXIJrr70WAPCb3/wGX3/9NR566CGccMIJmDNnDrKzs/Haa6/h5z//OTvnaaedhtzc6LKkCYIgCHV4Oj3wdnslj3m7vXC3uGEv1je3MaUdjmTC1eRE377IgkPEA4+BdxOaQw89VPL1oEGD0NTUBADYvHkzysvLmdgAgHHjxqGgoACbN29WFBzr16/H5MmTUVRUpFhTvXnzZlx55ZWSx44++mgsXboUgL8a6LzzzsOKFSvw85//HN3d3XjjjTfw3//+N+a1EgRBEOHp26u8b/XU9uovOHq4SbEkOKLHPjBD1fNEh8NisejmcGjBZpN2jzOZTDE1X3E4HFH/rMjChQsxc+ZMNDU14cMPP4TD4cApp5wS83kJgiCI8PD5G+YMM3xO/37QW9eLgin5ul5LElJJ8Gh6IIUFh9qwhs/nQ21tLSoqKhJeEiTnkEMOQV1dHaulBoBNmzahra0N48aNU/yZQw89FE8++SRaWlpQUFCgeM4vvvgCixYtYo998cUXkvPNmDED5eXleOGFF/Duu+/i3HPPDRJGBEEQhP7wFSr5k/PQ+nUbAH8eh95IQipJIDiSawc+yJg1axYmTpyIhQsX4rvvvsOaNWtw8cUXY+bMmZg2bZriz/zsZz9DWVkZzjjjDHzxxRfYvXs3XnnlFXz11VcAgN/97ndYvnw5/vGPf2D79u145JFH8Oqrr+KGG26QnOfCCy/EP//5T3z44YdBTdgIgiCSAW+vF/s/PiD5pJ7q8A5H4RGF7Lin1mDBkWplsYS+mEwmvPHGGygsLMRxxx2HWbNmYcSIEXjhhRdC/ozdbscHH3yAgQMHYsGCBZg7dy7+9Kc/wWLxq9czzjgDS5cuxUMPPYTx48fj8ccfx7Jly3D88cdLzrNw4UJs2rQJQ4YMwdFHH23kMgmCIKJi/dU/4ttz1mH9VT8m+lZ0g8/hKDyigB0b0YuDT05NBocjZUMqqYBSaevrr78u+XrYsGF44403NJ23oqICL7/8cshw0TXXXINrrrkm7DkOOeQQQ0ciEwRBxErbt20AgNY1bQm9Dz3pawg4HPkT82B2mOHr9cXB4Ui84CCHgyAIgkhKvL2+/n/TKaTidzhMFhMySjOQVe4vBOit69X9QyDlcBAEQRCECsQKDm+PN20cWTGHI2OgHSaLCZmDMwEAvj4fPB36tm9ItioVEhwEQRBE0iH4BCY4IPg35FTH5/bBtd8FAMgY5Bca1txAZoOnS2/BkVx9OEhwEARBEEkHExv9pENYxdno9DeHApA5yN/TyZITEAKeTn3XSDkcBEEQBBEBfrQ6kB6Cg++OnSk6HDkBh8Oru8PBh1QSv90n/g4IgiAIQoY8hMKHB1KV7upudpw52O9w8ILD02Wgw0E5HARBEAQRjFcuONLA4Whe3cKOC6YVAACsfEhFb4eDQioEQRAEEZ50y+EQBAHNn/kFhznTzASHJce4pFFXsxuAvwTXnJH47T7xd0AQBEEQMuQCI9Xbm/fs6kFfvb8ktvCIQlj6BQDvcMjH1seCq9WFrq1dAIDc8TkwmfQYXxobJDgIgiCIpCPI4UhxwSG6GwBQclwRO7ZmG+NwtHzRyipiio8pCv/kOEGCgyAIgkg6gpJG4xBScbW6dA9riDR/HhAcxccGBICkLFbHpFE+X6SIBAdBEARBKCMvi/X1Glul0rmlC6smfIpVEz5F396+yD+gAcEnMAFgzbMib1Ie+55RZbFMcJiBoqMKwz85TpDgIAiCIJKOeIdUmt5r8rcX7/Sg8d0mXc/dubkLrgP+DqNFMwphtga2XmmnUX3W6DzgQtdmf/5G/qQ82PJsupw3VkhwEARBEElHUNKowSEVZ78gAMCSLfWi+fNmdlx8XLHke0aUxbZ8wYVvkiScApDgIAiCIJKQeJfFijNOAKBra3eYZ2qne0cPOy6cmi/5ntWAsthQ+SKJhgQHQRAEkXQENf4yOKTi4hyOzm36Ohy8eOL7bvi/5spidQqpiILDZDWh8IjkyN8ASHAQBEEQSYhPPkvF4NbmfEjF1eSCq8UV5tna8LkC9y5vwGW2mmHO9D+mh8Ph7fGie4ffock7NE/ioCQaEhwEQRBE0hHv1uZ8SAXQN6zCV9yY7cHbrigK9Ega9XQHREvGQHvM59MTEhwEQRBE0hHPPhyCT4CrWS449Aur+FwCO7ZkKgkOf1hFD4eDf90smYmfn8JDgoMgCIJIOoJCKgYKDne7G4JHkDymq+DgcjiUHA4xr0OPPhz862RWEDeJJLnuhiAIgkgZXC798hzkeONYpSIPpwD+RmB64XOGFwGiw+FzCZJ8j2jwksNBEARBpBMXXHABCgsL8eqrrxpyfnlnUT0Hm8nhK1RE9MzhYA6HyV85IkdSGtsdm8vBO0NmR3Jt8cl1NwRBEETS09TUhBdeeAE9PT1YtmyZIdfwOuUhFeOqVJwKgsPZ6IS7za3L+UXBYc4wK05tlbY3j01Y8a8TORwEQRBEStPa2sqOe3p6wjwzeuKZNMqHVCzZgU26S6d+HLzgUMKiY7dRviJGKUE1kSTX3RAEQRBJT3t7Ozvu69N30JlIPMtieYeDH3SmVx6HN4LgsGbrNzGWF2pmBzkcBEEQRAoTD8Ehr1LxGdhplHc4io4OtALv2qZPHoeYCGoJJTh0bG/OCzNyOAiCIIiUJjEOhw+CIIR4dmzwSaP5hwZGx7ubdc7hUCiJBaTtzj2dsSaNcg4HCQ6CIAgilUmEwyF4BUkDLT3hQypZwx3sONbNX4QJjhACwKrjPBW+nJiSRgmCIIiUJi6CwxlcleIzKI9DDKlYcyzIGJDBHtdDcAiCENHh0DOk4qPGXwRBEES6EJeQikIZrFETY10HnAAA+wA7zA4zTBZ/6apbD8HhFoB+YyZk0qhEcMTocPCNvyhplCAIgkhl4uNwBG+8RlSq+Fw+uNv8wsJekgGTyQRrro6zTfi25hnKAkDPslgflcUSBEEQ6YJccBiRzClPGgWMERz80LaMEv90VWtu//RWHRwOrzP0aHoRaeOvWKtU+KRRcjgIgiCIFIYXHD6fDx6PPsmVIoIgBDX+AozpNspXqNgH+AWHWDWih+DgHY7QZbGcwxFjC3dJ4y9qbU4QBEGkMrzgAPQPqygljALG5HDwFSr2fofD1u9w+Hp98LljEzn8MDZ1ORx6lsWSw0EQBEGkMOkkOPimX/KQCqCDAFARUrHoOkuFcjgIgiCINMFowREqdGJEDodTIaTCC45YBYBEcIQsi9UzaZSqVAiCIIg0wXiHQ3mTNyRpdH9wSEWsUgFiL43l1xKqL4bJbGJD42Ivi6U+HARBEESa0NHRIflad8HBfUo32QLj3A0JqfBVKgoOR8ytxlU4HEDA5Yh9lkr/9UyhQziJIrnuhiAIgkhqnE4nnE6n5DHdQyqc4LAV2gKPG12lopTDEaPgUFMWCwTyOGItixX7cJgzzTCZTBGeHV9IcBAEQRCqkYdTAGNDKnaJ4DDA4eAFR5GxDkeoslggMKLe0+WNqa+JKNbCXStRJN8dEQRBEElLNILD4/Hgtddew48//qjqGryTIXE4DAipuNv8E2GteVbW0lxXwaGiLBYIlMYKHiFklY6q6/WJg+KSK2EUIMFBEARBaCAawfHvf/8bZ511Fo488ki0tLREvAa/4doL7ezYEMHR7hcUtoKAsJH0xeiMsUqlT1tIBYgtcVRMGk22pl8ACQ6CIAhCA9EIju+//x4A0Nvbi23btkW8Bl9pYSsyLqQiCAJzOGwFgQ2fr1KJn8PBj6iP/prkcBAEQRBpQTSCg/++2+2OeA3eFbBzgsOnc9Kot9sLwePPl7DlcQ6HniEVlQ6HHt1GBUFgoizZmn4BJDgIgiAIDUQjOPiqFjWCQ+JwFBjncLjbA/dilTgcOnYadWkriwWin6ficwlAf75psjX9AkhwEARBEBqIh+DgXQFJSEXnHA5xLD0gy+Ewqg+HyhyOaEMqviRu+gWQ4CAIgiA0EGtIRc1kWUkfjgLjBIeHcziMEhxeSVlsaNfBqkPSKF/dY6EcDoIgCCKViYvD4eRHrFtg7q+40D2kwjsc+QZVqahNGuUTVTt0cDioSoUgCIJIZeKSw9Er3aTFfAT9BQfvcAREhsnCzTaJU9Ko2HQMkLZb1wLvDJHDQRAEQaQ0calScUonnlqy+gVHj75VKu4QIRUg4HLEqyzWXswJjpbIr5ESyTyaHiDBQRAEQWgg3lUqcXM48q2S77FhavFyOIoDgodvtx71tahKhSAIgkhl4l2lYsk0BxyOOOVwAIHEUU+nJ6bZJrzDEW6+icThiDqkQg4HQRAEkSbEu/GXOdPC2nQLbgE+t35hFWkfDmXBAcHfICxa1JbF2gpsbEeOVnD4JLkv5HAQBEEQKYwoOKzWQAjCyJCKJdMsaWKlp8shzlEBFHI4cnWabeKUhodCYbKY2GRcV3OUORz860ZVKgRBEEQqIwqOgQMHsseMdThkgkPHxFFPuBwOnXpxqO00CgTCKlE7HDJnKNkgwUEQBEGowuVyMfFQWlrKHtfd4XBKyzvFHA4A8PbElsTJIyaNWrItMNuk26FugkNl0igA2PoFh7fbK3Er1EI5HARBEERawOdvqBUcgiBoTxoVwyYmwGQzwcLPGYkhvCFHzOGQJ4wC+jscJpsJJrMp7HMllSpRhFUkfTioSoUgCIJIVXjBUVhYCJvNv0GGExxut1tS5aGqtXm/w2FxWGAymSSTXPl25LEiVqnwTb9E9BpRLzoc4SpURGKtVPH10iwVgiAIIg3gBUd+fj4yMzMBhBccvLsBqM3h8G+cYgjCmhcQBO4o237L8fZ6WQWJPGEU0N/hiBROAWIXHNRplCAIgkgLOjo62LGxgqPfFej/lG7jBEe0c0bkSCpUlEIqOToJDqcoOCILAGlIJQqHg2apEARBEOlANA6H/HvqymL7N+l+wWHN19/h4LuMWhVDKvoIDi8THOHzNwC5wxFFDgdNiyUIgiDSgbiFVMRNun/TlORwdOiTwxFujgqgY0hFXEuEklhAKjjcUYVU0rBK5emnn8b8+fNx3HHH4cILL0R3dzcAYPny5Zg1axZOPPFELF26NKZ2sARBEERyEQ/BIQgC2zgtCjkcuoVUwvTgAOSCI/ZOo2r6YvCCwxlVSCW5Z6kEv8oRePHFF/HVV1/hqaeeQmlpKXbs2AGbzYbVq1fjpZdewvLly5GZmYnrrrsOFRUVOOOMMwy4bYIgCCLehBMcgiDAZAoOG2gNqQhuAejfN8VN02ZASMUTZo4KIKtS6YrumoJXgODxf/BW53AE7sMdVUgluR0OTYLD6/Xi3//+N5588kmUlZUBAEaNGgUAeOedd3DmmWdi6NChAICLLroI//vf/xQFh8vlgsslVW9WqxV2uz3oubHi8/kk/6YTtLbUhNaWmtDagLa2Nnack5PDBIfP54PL5WJlsjy9vb2Sr10uV9jreHoDm7s5wwSfzwcz14fD3ebW9B6EWpuzNbAHWfOtQd83Z3PX7NR2TRFeAIhrCQefS+I84Az7fKV18SEVk4rr6YXZrE7caBIcTU1N6Ovrw0cffYTnnnsOOTk5+PnPf44zzzwTu3btwpw5c9hzq6qqUF1drXieZcuW4YknnpA8du655+K8887TcjuaqKurM+zciYbWlprQ2lKTg3lt9fX17Linp0cSNt+2bRtycnKCfmb37t2Sr9vb21FbWxvyGp7mgOBwep2ora2Fl2v21dXUFfbnQyFf24Hdzey4xdUCd630QzA/dK17f3dU1/R2BO7b6XOqOocp0wShT0B3Y4+q5/Pr6mkLiLs9jXtgaomcqKoHw4cPV/U8zYKjq6sLu3fvxptvvom6ujpcc801qKysRE9PD7Kzs9lzs7Ozg5StyOLFi7Fw4ULpjRjocNTV1aG8vFy1CksVaG2pCa0tNaG1+V1ukbFjx6KgoIB9PXDgQAwYMCDoZzZv3iz52mazoaKiIuQ1es292I6dAICcohxUVFRA8AnYZqoGBMDqtob9eTmh1taDPhyAX3QMGTUEhRUFkp8TBAHbbNUQ3ILma4o4G53YBv8H7+yCbFXn2DVgN/rq+oAOIezzldZVL+xDLwCYgYqRFYohrkSiSXBkZGQAAK644gpkZmZi1KhRmD17Nr744gtkZWWx5FEA6O7uhsPhUDyP3W43RFyEw2w2p90fCRFaW2pCa0tNDua18X04CgsLJX/jXS6X4s/Kcza8Xm/YawjOgGtiybT4n2v2J4562j1wd3iiev3la/NwfTgyCu2K57TmWuFuccPTFd01BW7plgyLqnPYi+zoq+uDq8UNEyK3Q+fX5et/7SwOCyyW5Esa1fQKVlRUwGazSVSTeDx8+HDs2LGDPV5dXY2RI0fqdJsEQRBEouGTRvPy8lgOBxC6UkVrlQofyuCbV4nNv4yoUrEqJI0CgeZf0V5TshYVnUYBIKOk/8O4T3qPqq4nVvckYcIooFFwOBwOnHTSSXjqqafgcrmwa9cufPjhhzj66KMxb948vPrqq9izZw+am5uxYsUKzJs3z6j7JgiCIOKMKDhycnJgsVhUCQ6tVSqS9tzcJm3VW3DwnUYVGn8BgeoYd4cnqjYPPiefNKpuu7UVcd1GD2grjRUbfyXjaHogirLYm266CXfddRdmzZqFgoICXH311Zg8eTIA4JxzzsGiRYvg8/lwxhln4PTTT9f9hgmCIIjEIAqO/Px8ADDE4eBLUPn24mLpqs/pg7fPG3MnTdE9MGeaQ55L7HAquAV4e7ywZmvbMiUOh4qyWACwl3DdRlu0ORzeJHc4NAuO3NxcPPjgg4rfW7x4MRYvXhzzTREEQRDJRzwEh7eb6yXBlabKm3/FKjg8baFH04vw33O3eWISHGpFgL2IExwaHQ6x8VcyNv0CqLU5QRAEoQK3242enh4A2gSH5pCKRHAENni+86cezb/Ec4QKp/i/F1tLdW9UDgcXUmlRLzgEn8AETrI6HMl5VwRBEERSIZ8UCxgUUunmQiqcw6HnxFif28eETViHo0DqcGi+jkt70qjU4VAvcnxJPpoeIMFBEARBqEDe1hyIh8PBhVT48EZ7bAPc+GFsfKhGDi9ytFaMANFVqUhzONQ7HN4kH00PkOAgCIJIG3p7ew0bmhmt4NDucAQ2TknSaJ4+01vlP8+HauRYJQ5HnARHlFUqyT6aHiDBQRAEkRasWrUKJSUlmDFjhiEzNOIlOLxcSCVc0mgseLhW6eEEB5/fEU3eSMwOh4aJsT7e4aAcDoIgCMIonn32WfT09ODrr7/Ghg0bdD9/vEIqEoeDz+GQhFTi43DY9HQ4VCaN2gpsMFn9DTVd+7WEVDiHg6pUCIIgCKM4cOAAO+YTPPVCL4fD4wkvFkJWqUgcDh1zOHJDb868yPFEITi8Tu1hDpPZxFwOZ5MGh6NXe5OxeJOcd0UQBEFooqWlhR13dnbqfv645XB0KVep8E5EzCGVTuXmYnLETqNAlFUqUTgcAJBR6p9b5jrgguBTl5MjcTgopEIQBEEYRXNzYNx6MgkOvapUJJu/noJDbUglCldFUharQQRkDPA7HIJXUN1tVNJGnUIqBEEQhFHwDkdXV5fu549fH47+jdMkzUWw5nHhjVhzOFQnjcaYw9EXncNhH8Alju53hnlmAGmVSnJu7cl5VwRBEIRqBEFIm5CKWKViybJIRrNLemJ06pnDEVpwmO1mWLL8oidejb8AIGNgBjt2NqrL4+D7cFBZLEEQhAIulws//fSTYf0jDgY6OzslyZjJJDjkj3u93rDvtehwWHOkm6bZYWbVG/rmcITfnMVk1ZgdDk2CI+BwONU6HD2c4MgiwUEQBBHE2WefjYkTJ2LJkiWJvpWUhXc3AGNCKno5HEB4l0PM4bDIBqWZTKbAuPg4lcUCgbBKNCKHdzgs0TocKktj+fuz5mueyxoXSHAQBJEw6urq8NZbbwEA/vrXvyb4blIXPmEUMN7hyMvLA2CU4PBvnHyFiojoNujqcEQUHP7ve3u8EgGhBq8zulLVDD6Ho0mdw8GHfGx5oefDJBISHARBJIx333030beQFsgdDiMFR3Z2NqxW/yYcTUgFCC04fC4ffC5/uMWiKDgCbkMsIThJ6a1KhwPQHlaJtizWPoBzOFT24uDny5DDQRAEIeOdd95J9C2kBXKHw8iQihhOAfR3OPg8BGt28KYpJo4KXkHyXK14OpXntSjBuwWaBYcrIIo0lcWWas/h4F0fXiQlEyQ4CIJICE6nEytXrmRfWyzJmeiWCsTT4eAFR0ZG4JO4FsERqtuoJ8QcFRFJt9EY8jjEkIolywKTxRT2uZJ5KhqvKZlvosHhsBXYYLL570u1w8GJIRs5HARBEAFWr14t+STu9XojlkwSyhgtOLxeL3uveMFhsVhgs/k/TesRUvF2KTf9EuEFRyzNv0TBEa6tObsmH1Jp1+pwRNf4y2QyIaO/vbnaPhy8GOJfp2SCBAdBEAlBKX+ju7s7AXeS+hgdUlFq+iUihlX0CKlIB7eFDqkAsY2oF3M4IoVTgBhzOMSyWDNgtmrbbu0DxfbmbgjeyPkqnn4xZM2xaL5WvEjOuyIIIu1Ryt8gwREdRjscSiWxIuEEhyAI2nI4IoVU8qN3G/h7CjgcKgRHDPNURIcjmkZcYi8Of3vzyGEV8fXgX6NkgwQHQRBxp66uDps3bw56nARHdBhdFhut4HC5lDdKdQ5H8CYtcTiiDKl4e7xAv/GgTnDwLdW1iRxxWqzZHj5PRAmtvTjEkEqy5m8AJDgIgkgA1dXVio/39PTE+U7SA6XGX3p2bo1WcCi5G0A4h4PL4VAId1h1EBxaKlQAWUhFa9KoKDiiGBcv7cURXnB4+7wsfJOsFSoACQ6CIBJAU1OT4uPkcESHXHB4PJ6Qm300RCs4QuV1hHY4lEfTi/AlqmqnqAZdQ0PTL0BWpRJlHw5zhvaQip13OCI0//KkQMIoQIKDIIgEsH//fnY8bNgwdkyCIzrkIRVA37CKWsEhd1VicjgUBEdWpYMdd26Kbn3Spl+RhQAfUtEqOEQXRs115EjnqYR3OPh8FhvlcBAEQQTgHY7hw4ezYxIc2vH5fEEOB6BvpYoaweHz+YKEhFbBEalKJeeQHDaYrG1tW+QbV7qGRoeD79qppfeH1+ljDoc1ilbjGVy30UilsXyoh3dkkg0SHARBxB3e4aisrGTHJDi009nZCZ8veMaHUQ6HOEdFJDs7mx3LRY7WkIq3K3yVitlqRv5k//V76/rQ16A9bKRVcFiyLKwJl5bKGE8H5zpEEebgczgijajnr0VVKgRBEBwkOPRDKZwCxC+kwn/NPw+IwuHoCl+lAgAFUwvYcdu6NsXnhIMXHEqJqXJMJhNLxNQSUokkbO55RkDFuT688olycm9GKV+lEsHh4Ae3UZUKQRBEAD6kwguOdKtS+e677/Dwww/jwIEDhl1DKZwCxC+kUlBQoPg8ILTgCNXaXJrDobxxFkwLXL9tbbvic8LBCwGbCocDCORFaOnDIRkXL3M4PB4Bf1wuYHcjcM9/lAWHNd/Kymk15XAohG/6nAKcLv2qlqKFBAdBEHFHdDiys7NRXFzMHk8nh8PtdmPevHm44YYbcMsttxh2Hd7h4IepJYPDwYdU7PZAiCDaKhVAB4eDd1FUCw7/8zydHlVdPwFp63X5dRpbAXf/t6v3QrGE2WQysamxkapUpDkcwYLj+ZVA5iwBA0714fmPEic8SHAQBBF3RMExcOBASQ5AOgmO+vp6NDY2AgA2bdpk2HV4h4Ov+NFTcPCO1IABAyTf4wVHW1ub5Hu8w5GTk8OO1fXhUBYcmWUZcJT7hVX79x3weYLzV8IhDXWoqx6R9OLoUBdWkUxvlTkceznDq6MbaA3xVtnFeSoHXBB8oYWCJ8JoevF6B9oBR0bQt+MGCQ6CIOKK1+tln8oHDBiQtoJj37597JifRaI3vODgw1N6hlQaGhoAAA6HA7m5uZLvqc3hUCM4IlWpiBRMKwDg7xratVnbOrUmjQKAVdKLQ11YJVxIZa8swrZrHxQRB7jBFz5hVeJwKAqOgFgZVBz07bhBgoMgiLjS3NzMLOSDRXAYMS5ehA+pVFRUGHJNcS2DBg2CySRt0602pKLO4fBvnGaHWTI2/qedAmZc48Pv/u53M/g8jlaNeRxak0aB6NqbuztDC459srSbmhCCw1bENTo7EE5wcDkcCiGVvVxe8WASHARBHCzw9nw6h1QS4XAYITicTidaW1sBAGVlZUHfV5s0qsXhkLsbS18W8NVG4KH/ArsbBWkeh8Z+HNEkjYqhDQDo2q7ud1RyHVkiJ+84AKEdDv664Qa48a6LUs8P3lEpI8FBEMTBAl8SK3c40qlKRS449JxtwhPK4dArpCKGUwC/wyFHbUiFD8VEyuGQ9+Co4zrht3QAeYfmwWT1OyCdG7UJK28USaPFRxex48a3ldvyy9EWUlH+3bAXcYLjQGjBwVwXs388vZx9/b8iAwsBm1X7IDm9IMFBEERc4QVHOjsc/Ebt8XhCNsGKFaMdDi2CQ540qjWkIlapyCtUmloDxz19gCXDzHIV3BqHuImhDpPFBLND3RZYdFQhcxv2rzzgnzgbAUkzrtzwgqOmAYrYS7hk1TCzY8TXwJZvg8ksFRQ+n8AEx+CSSHdtLCQ4CIKIK/KKB4cjMB8jnQQH73AAxoVVjK5S4dehFFLRK2lU8Arw9fpzNOQ9OPa3BY57+z/oi5s4H7pQgzhLxZprCcpHCYXJYkLpXH91jrfHi/0fR+6r4g5TpSLP4QgZUuEdjuZwIRX/66mUMHqgHfD066NE5m8AJDgIgogz8pCK2WxmooMEh3bE0tv8/HxJPkWqhVSkFSoBh0MQBDS1BZ7X02+a8IJDS7hKFChqE0ZFyk4tZccN/2uMfB0NIZWaBuVeHPbiyIJDEAQ240WprTl/rURWqAAkOAiCiDPykAoQmMeRzoLDiEqVjo4O7Nq1CwAwZswYiYsQL4cjLy+POQVqq1SUOo16uaZffA5HZw/g4vRJb7+GEQWH4BbYkDQ1iIJDbf6GSPGxxUw4NL2/Hz5X+GtKBAcnbtweQRIiAvxrkj8GAPZirkqlOXTei9iMTLkkNnBMIRWCIA4qlJpIpZvg8Hq9knUCxjgc33//PTueMmUKLBYLc4vilcNhNpuZe6GfwxHYOOUbcY9McADS7qHh8Hl8LGyjtkJFxGw3Y+Ac/++rp8OD5s+VZ9iwexKFTY5FUuLbEOLHlMIqahwOaQ+OCCWxJYlLGAVIcBAEEWfkIRUgfoKjvb0dixcvxq233mpY1QjgF1XyCa5GCI5169ax46lTpwIIbOx6hVQiORxAIKwSLmmUTw5WEhzSOSoBh4PP3wACIRW+E6naPI5oKlR4SucNZMctXypYEhxiDoe8THUfJwDM3A6sJDhsBTa2S4cSHJG6jO5Lkh4cAAkOgiDijPjJPycnh30aFzejnp4eQ4XA448/juXLl+Pee+/FN998Y9h15OEUwBjB8d1337FjueDQ2+Ewm80sBCZHFByxJI3yIRW+tFMp/ADIHA6VgiOaLqM8OaMDa+hrDD/fJBC6kVbc8I7DoSMDx0qCw2QxwV7oFyyhQioRHQ6u5weFVAiCOKgQHQ5+JkdWVhYAfwKcUeWjAPDDDz+w4z179hh2nXgJDtHhsNvtGD9+PIDAxq53DseAAQNgsSjPHhEFR29vr0RMRBtSkTgcskaiYkjFFoXgcEu6jKqbo8KTMTAQ4nCGERw+j485NnKHg8+pmDE+cFzTEKIXR39YRY3DQTkcBEEQ/Xg8HlbGyQuOePXi2LFjBzs2svtnPARHV1cXtm7dCgCYOHEim8YqbuwulwsuV/ix5pHw+XysCkYpf0MkVKWKlj4coUbTBzsc/o2ZT8QUS10j4e2KzeGwFdpgsvWPjG8K04hL0mVUVhLbHBAWR08M5FSEKo219QsOb7cX3r7gXBXe4VCsUul3VEwmYGBByFuOCyQ4CIKIG/wcFd6ej5fg2L59OzuOt+DQu0pl/fr17LWcMmUKe5x3EmLN42hpaWHiIFT+BhC6vbk2h0N5NP3+Nuknf3lZLBC/kIrJZELGwMgj49WWxI4fDuT6zb0wvTjCV6pEHtzm/7e0ELAmsMsoQIKDIIg4opQwCsRHcDQ3N7OZIICxA9Xi4XAoJYwC0LU0NlKFikiobqOacji6lEMqoapU+D4ank51VSr8ADSrxj4cImJYxXXAxcpR5XjCDG7jcziGlADD+1/W3Y2AV+F8/DwVt8I8FU+YwW1er4CG/iZjiQ6nACQ4CIKII4kUHLy7ARjrcPAbtVHXU0oYBfR1ONRUqAD6hFT4iae8GOCbfgF80qj2KhW+YVfOmOwwzwxNRqnf4YAPcO4PkVcRpsuo6DjYrEBxPjCsv5+Y2xNckQNIu406Feap8IPb5A5HUysgFkuR4CAI4qBCPilWJB4D3OIpOOLpcFitVkyYMIE9zguORDgcoUIqkQRHz+5eduwYmsmOQ5XF8iERr4ocDmejE00f+AVvRmkGSo6PrkaUCQ6EDqu4QzT9AgJlqoOK/SEavvvnPoUeHZHmqfBD3Xg3RH6+RJfEAiQ4CIKQ0dTUhOOOOw5nnnlmyCFb0ZJIh4NPGAXiIzj4OTF6Xq+npwebN28GAEyYMAGZmYENWs+QSqwOhyg4rFYrMjICG7VSp9HeOk5wDAu8bqFCKlqrVOpf2MtCIEMvHAKzNbrtT8zhAEILjlA5HC63wASU6DiUBQbRKguOCBNj+Xvg7w1IrqZfAAkOgiBkLFu2DJ9//jlef/11fPTRR7qeW6mtORAoiwVSP6QiCALbqEeMGAFzf3cnPXNGtmzZwhqLTZ48WfI9PUMqah2OUEmjolvlcDhgs3Gf1BWEbG+tX3DYCqyw9ZeSCoIQ5HAo9eFwRxAcgiBgz4q97OuhC4eEfX44pKWxIbp/hhAcDdzQNtFxGFRsUvy+iKS9uUIOh1gtY8m2BLkpyVQSC5DgIAhCRnV1NTvmBYIeKLU1BxKTw2FU0mhraysrRx08eDDy8vIA6CtweOeBH0kP6BtSicbh4JNGxd+fkpISWK2cQJAJDp/Hh769fiXhqAiIz7auwKRTkUCnUfUOR+/6XvTs9Iuf4mOKkD08K+zzwyFxOEL04pDmcAQEg5IAiBhS4dubH1AIqfQ7HLwQClwvkISa6MFtAAkOgiBk1NXVsWO9XYBEhVQEQYibw8Fv0oMGDWICwCjBIRcCRlWpaA2peDweNDf7d9CBAweGdTj66vtYuMNRHjp/A1B2OLwRZqn0fB8I1ww5f3DY50ZCTQ5HqCqVRs7BKCsy9f8beIzv0SEiERwyh8Pn8rGkUXk4BZCFVEhwEASRCPr6+nDcccdh4sSJqK+vl3xv9+7d7FjvTTlRDkdzc3NQ2+14CQ4jHI5woQ4jqlRycnIkQkaOkuAQxQbgf6/DCY5eLmE0i3M4lCaosuFtGmapeFoDgiS7Knp3A1ApOEKEVLoCy2T9N3jnQTmkElpweFsC61IUHBRSIQgi0axcuRKff/45fvrpJzz//POS7/EOh95hB9Hh4OeoAMYLDrm7AcRfcHR3d8PrVdcvQss15M4DLzhiXaMobMK5G4Cy4JBXJFksFjbGXi44JBUqXMKoksMhhlRMZhMTHZEEh5cTHPwGHg0ZA7gcjhDdRt2dXIkv58T0ck939OuD0sLAY0ohFUuWBWaHf6uWh1Q8BwLrVgqp7Ok3FC0WYECB4q3GFRIcBHEQwm9E/CfR9vZ2icgwKqTCuxuA8WWxoQSHEYPiQgkOID4j40NVjGjF5XKxny8tLQ37XKWkUaUSaNHlCOtwhKlQAaSbtpjHoUlwlMQmOCwOC3Mt1OVwcIKDe7ooODIzTCjs14hKDgfAzVORORye5tAOhyAI2NmfJ1tRClgsVKVCEEQC4Bsy8ZsgH04B9BUc8pg+TyIcDrfbLekToRfxEBz8NeRiQC/BESrfRomcnBxWjSMmjSr9fCjB0VMboiS2LfhaPdxsP9E98PTncAheAT63L+hnxJCKyWYK6vwZDWJYRY3gsIYSHJzuEcMq+5qhKIJFweFudkPwBb7vaeYcjlKp4DjQDnT2a/eRsaWt6AYJDoI4CAklOPhwCqCv4JDH9HmMLovle3CMGjWKHRtRqRIqaRTQ7/UUHY7i4mI2tE1EL8ERqkmbEiaTiQmrWB0OadJoYHO19e/bbg/g8fQPcGOCwwNnoxOrDv0UH0/6FH17pROHvW1+wWEvsrOwTixklAYGqikNjhPLYs0OM8y2wDar5HAAgcTRXifQofDrL85TEbyCRMzwDoddFlKp5lKzRkZfBawrJDgI4iCE/2TPb4JGOhzhPjEb7XDU1NSw40mTJrFjI/I45OEO3uHQ43p8nw+l3IpECA7+urEIDvsAO6zZVng8AlxuQRJSKeduQQyrsPwIAdjz33o4G5xwNrrQ+G7g2oIgwNPi36T5rp2xICmNVWhvLooCeVtzcdItIBUcWhJH+fbmEodDFlKRCI7BiQ+nACQ4COKgRK3DoacDEG4DM1pwiNcuKipCSUkgXd8IwSGKgdzcXGRnZ+suONrb29n7p9SMKzMzk3X15HtiaEVLSAWIXnB4nT70NfgFcNYwB1xuAZMvE5A1W8A7XwfOX8FFjgLtzQOVKp2bAhU5rtbA+T2dHqB/X441YVQkUi8OMafEmisVOJEcDiBULw6+vXlAcHjD5HBUB/qcYQSFVAiCSBS8wxGvkEoiHQ4+WVVvASBHFByiGND7emp6Y8g3/2jQ6nCIiaNOpxN9fX1hczj41uZ9e3qB/g/+jmEOfLcN+GkX4PUGchAy7cBArpqD9eLgmn91bgr8Hrs5wcFXdtgH6CQ4wpTG+tw+TnBYJN/jE16zApEjSbfRiO3NuXkqEodDtradewNuCoVUCIJIGLzDEa+QSqgeHABgt9thsfj/OOtdpdLb28tE1cCBAw0VHN3d3exaSoJDD8dIniOiRCIEhzyUo/R+i91GeYdDXhLbLU2/AABUDZFu0D0Kzb+6tgeEqkRwNAd2+QzdHA6+vblUcHRu6mICKmuEtOdHpKRRQDmkYskOCBdvT8DVEHM4bEU2mO3S7ZwcDoIgkoJEJI2GmqMC+JMORZdDb4dDfl0jBYeSGNA7aVSLwxFL6W+0IRVAKjgKCgpYYqtSSEVeEstvyrOmAZfOA5640YQsLmKgNDFWcAfW6W4LnN/NCQ7dQioSh0Oaw9G2ro0dF04tkHxPXUgl+P2yZHGCo7e/IkcQWB8OpaZfYg7HwEIgN4tyOAiCSBBKIRWfzxckOHp7exUne0ZDpA0sHoJjwIABus4akaPUkEtvgaPF4fD5fFF3G43V4VDquaIkOCQlsRVSwTFnuglP3WzGkeNNkg1aqb05T8iQSow9OETC5XC0rQu4SgVT8yXfCyU4IjocvODo9gsOb5cXQn8Sqjyc0tMnsNBMspTEAiQ4COKgRO5wCIKApqYmxSmeem3K4UIqQKA0Vm/BId844+1wJCKHI9T0Vi2Ir5vJZEJxceRBHLzgaGpqYtflxUokh8NR7gi5KUcKqfC4QoRU9Esa5buNygTHWv+6zXYTcifmSb4X0uGIMMDN4gh2OCRj6WU9OHZy4ZRkyd8ASHAQxEEJ73AIgoDu7u6g/A0RvTblRDkc6SY4tDgcQPSVKnxlDz/pNRS8yOH7nigJDo/Hw0I9THCY/IKjT6H9NwBkZQTCAoGkUWlSpggfUpEIDp0cDnuxHab+zp28w+Fuc6N7h//3N3dCHiwZ0i2WTxrN5G6lIAfI6P9aUXAo5HA4G7nclHA9OMjhIAgikfAOB+B3MeThFBG9BYd8joqIKDicTqduM0f46wLpITjCtTUX0aMXh/i6qQmnANKOp2vWrGHHSiEVIFCp0rvHLzgyyzJgyTCHdAEcEXI4eNytbiZoXM1cSKVYnz4cJrMJtv5z8VUjbd8HXuvCaflBPyeuLcMOmM0BAWUymTCoP49DMaQicTj8nVQlDkeYktiRQ5IjfwMAYu/xShBEyiEXHB0dHRLBMXDgQPYJV69NWTxfqA1MPk+Fz7XQ47qA8WWxkZJG9axSyczMlKyFJ1bB0d3dzZwmtYLj0EMPZccffvghO1ZyOAB/WMVms+H4745Db10fcyV4wcG7AFpCKoJHgLfbC2uOFa79+jscAGDLtcLV5JJ0/mxb28aO86eGFhwOhdsoKwZqGoDmdsDlFmC3BYSCUg4Hn6wqFxySklhyOAiCSCTyGSKdnZ2SkMr48ePZsR6bssfjQUuL/6NbqIoHo3pxyEMq8UoaNbpKZdCgQSHbdMcqOMJVFIVi3LhxTFCE+nmlEfWWTAtyRmWjcHoBAKDXpdyNk9+kIyWNAoGwihhSMVlMsBXo43AAgDW/PzzU6WHzTaQJowVBPyM6M47gohLmcADBLofE4egPqbg4hyOorTnlcBAEkSxECqmMGzeOHeuxSR44cIAdJ1pwGOlwiGIgIyMDhYX+TlV2ux2ZmZm6XM/lcrGZNOFGxseaNKq1JBbwr3PChAlBj4cKqSglKAOhEyslDocYUgmRwwEEKlXEkIqtyAaTWb/wAmtbLvhnuQiCwASHvdiGrMrgsKGYw6EkOPjE0QZZHodSWazE4ZAljYo5HNkOacO0REOCgyAOQuQOR0dHh+TT+SGHHMKO9XAB1HxizsnJ0fWa8mubzWYUFRUhOzubOQNGhVTKysok7oMocmK9XmNjIzsOlb8BxJ40qrUkVmTy5MlBj4VyOEKVW4dqjiUNqUiHtynh6s/jEB0OvfI3RKz5gWt72j3o2dUDd38+R/7UAkX3iYVUlByOMN1GLVmBrZoljYbI4fB6BdT0p/mMHAxdhtXpRdSC48cff8T06dPx5JNPsseWL1+OWbNm4cQTT8TSpUujbjhDEISxKDkc4iaTn58v+VSqx6as5hOzXkPH5IjrKikpgdlshtlsZmEOPQWHy+ViTo5cDOglOJT6fCgR62tplODgq11CORyhqlSiCal4u73w9fmTLPUqiRXhB7O5Ozzo+CkgkgumBOfWCIIQVnAMLAgcH5C9ZZaswLVEh8PVFAgVidNkAaCjxz9RF5A2FEsGohIcPp8PjzzyiMR2Xb16NV566SUsX74cL774Ir788ku88cYbut0oQRD6EU5wGBF2iNSDAzBGcIj9RQDpxqeXAOAJ5z6IAkfseRItaipUAH0Fh9qQCqDN4dAjpGLOMMNkU/4E7251G9KDQ0TM4QAAT7tbkpzqGBIcTnG5AfGtV0oa5RNknbKXxmw3sd3aIyaN9l/PXiINFbm4n83Ud8kxE1WVyquvvooJEyZIOti98847OPPMMzF06FAAwEUXXYT//e9/OOOMM4J+3uVyweWStoO1Wq2s/a2e+Hw+yb/pBK0tNUmGtclDKvv372cb04ABAyT5FO3t7arvNdTa+A2spKRE8Xz8JtnS0qLL69PV1cXE1YABA9g5eYcj1rWJ1NcHmh+UlZVJnicKHLfbjZ6eHsWyYDXw1xg4cGDIe+ETVdva2iKuUb42Ne+XEhMnToTJZGKiymQyobCwkP0873A4nU7F8/Zwv5oZNgG+/oRMfvPscQbu1ZprZaEMx7BM9O72v9+uFhf69gdOZiu26vp/js8fcbW54OSmuFoLg6/VHehvhsyM4N8jG7cb97kC6xaxZFng7fLC2+uFz8cNiCuwSc7FO0Q2a3z+zpjN6rwLzYKjra0Nzz//PJYvX46HH36YPb5r1y7MmTOHfV1VVYXq6mrFcyxbtgxPPPGE5LFzzz0X5513ntbbUU2oHgPpAK0tNUnk2uQD0n788Ud2nJOTI/kwsXfvXtTW1mo6v3xtfCMoQRAUz8d/4q2pqdF8TSX4ypvs7Gx2TnF8e1dXF3bt2qX6DyYQ+n3jX0OHwyG5f3EwHQBs2rQJJSUlqq/Hs23bNnZsNptDvkZ8DkxDQ4Pq11Jc265du9hjHo9H03tRWVnJfr6wsBB79uxh3+OFbm1tLUum5WlpGwjAL8j2N9XB0+PfMFsOWAH4Sy72N3ejtrY/ETkzsDFbR1qB/rf8QG0zekoCv+e9tj5dfqdEOrwBd6xhZyN6awOuYbOzGb210v9jTW0WAP4P5PD2oLZ2v+T7HW0OAH43qKGpFbW1MvctA0AX4OxwomZnDSuP9dm9knXVNAZeJ4+rC7W1Cp3EdGb48OGqnqdZcPz973/Hz372s6Aa+Z6eHsmnouzsbPT29sp/HACwePFiLFy4UHojBjocdXV1KC8v1/RHJRWgtaUmybA2uZ3Nf6KtqKjA2LFj2deCIKCiokLVeUOtjd9oxo8fr3i+ESNGsGOLxaL6muHgcx4qKyvZOfkNv7i4OGQ/C55I7xufBDl27FjJ/fNhhcLCwqjXxr9vEyZMCHmewYMDzRdcLlfE68nXxlcJTZo0SVVrc5Hp06czwVFWVia5tli5A/gdJ6X7MnEv7eiR5SyUksltOSZrNioq/PvNnsK9cO/1C+SyI0rR+bH/OMvnQIG5EHvgrxEtGVGCiophqtcRCWulDY3wi4Z8Wz7gMaEVbQCA8nHlyKnIljzfy+WsFhVkBa19aOBXFVnZhaiokJaX1OTWobe5Fya3CUNKhmALtvufWyw9VzdnjBQW5KCiIgfJgibBsWXLFmzatAk33XRT0PeysrIkv6Td3d0hbUO73W6IuAiHmCyWjtDaUpNErk2ewyFvRc2XVXZ2dmq+T/na+LLYsrIyxfPxm1FHR4cur41YQgr4O2GK5+TDN11dXZL1RiLU+8bncAwZMkTyHF7QdHd3R702PvmWX4+cjIwMOBwO9Pb2or29XfX1xLWJ75fFYkFxcbGm+50yZQpefPFFAH5Rwf8s/3f/0UcfxWGHHYbRo0dj3rx5zAXqdQVCAFmZJtaRM9shQJz73ucM2Ph84mj+pMD76m7zwMN1Ac0sydD1/5u9ILAWb6dXMjAuszj4Wk534P6zMoPDEI6MwPddnuDvW/tLY709Xvi6A6+RNdcqea7bEzhPhk19uCMeaBIc3333HWprazFv3jwA/v+oFosF9fX1GD58OHbs2IGZM2cCAKqrqzFy5Ej975ggiJjw+XxBOVS8EyBvjpWIpNFo53+ova4RvTjCzTjRq+RXS/VIfn4+ExzRXkcuGNTAJ47K75FPGn322Wfx7LPPIicnR/IeiDkI8vbfktbmfDfSwX4LxGw3IX9S4H11t7nhPBD4PbeV6FsWa+OSRt0dnoDgMEGxwViocl+RDO5H+lzB3xebf/l6fXC3B9w0eaWOi6s2ztB3yTGjSXCcddZZmD17Nvv64YcfxuDBg3HJJZfghx9+wH333Yc5c+bA4XBgxYoVOP/883W/YYIgYkMuNuQMHDgQVquVfULWY0OuqakB4B8EphS3B/SZcCon1Aatt6CSX4ufKyK/XrTj4oGAw5GVlSUJYSuRn5+PhoYGza8lX9mjpUJF5KijjkJpaSkaGxtx/PHHS77HCw6R0aNHS3pFhGr/bbOaYLMKcHukm3fVb0fC5/Rh4JyByCjNgDnTDF+fz1+lcsDAKhWuLNbT4WYTam35VjbYjSdU9Y2IpEpFSXBwzb9cXNOvIMHBRUvtqSw4MjMzJX8sRNsuNzcXxxxzDM455xwsWrQIPp8PZ5xxBk4//XTdb5ggiNiQh1PkiBtzXl6eLoKju7ubJQ6OGTMm5POMKIsNJTiMcDj4cIc8KVSvdupahID4enZ2dsLn86l2Kjo6Opgo1dKDQyQ3Nxfr16/Hrl27cOSRR0q+F0pw8ITrVeHI8PeY6OF+hXPH5mDqMwFXxVZgg7PBaXhZrI1r/OVu98DdX6ViK1K+TiTBwbsR8rJYQCo4+KZfcsHhTBfBIefOO++UfL148WIsXrw4llMSBGEw8pJYOeJmlpeXh8bGxpi7fvL5IfLNhYfflPUSHKEajvGCQ6+upuK18vPzg3LU9AipeL1elluhRgiIjpEgCOjs7JQIunBEM0dFTllZmWJjMqUx96EEh1IPiawMoKNbGlKRYyvsFxztHvTt63+iCZLmWHrAOxzuVjcLc4Sa1yIVHMEOSAbvcEQSHI284JC2d+cdjowQPUoSRfJkkxAEERe0OByA/xNvLM2q+FLOcILDYrEw0ZHKDoeS+6BHSKW5uZm9D2qEQLQ5MdE2/VKDKocjzLwRsWKlN4zgsBf6r+Ht8aJzk1/cZVTZFcMcsWDNsQL9p+zZ3SvmaYYUNr1cmCQqh4Mb4NbXGNrh4HM47Ek2D54EB0EcZIRzOEwmEyuBFDdlr9cbssRdDWoFBxDYJPVOGrXZbJINWG/B4Xa72T1HEhzROhxaB6pFG6Liq22idThCoUZw9IURHOJjYR0O3mHoL+bIOiy6RmvfbRWw+D4fPlsfLLhNZhPb7Pv2BES8rVCNwxH8fYngiJDDIQmp5MhCKtzPplVIhSCI1COcw1FcXMxsb3liZVZWVlTX0yI4CgoKsGfPHt1DKgMHDpQkJuqdNBppGq4eIRWt802iFRxq26dHg5LgGDVqFDt2ewR4/f2slB0OUXD0+UNFSoPJlEIajigEhyAImHqFX2h8sUHAtucUrpVvhafDA8EbECQhHY5IVSpRh1RSp0qFHA6COMgIJziMKB3lBUdVVVXY54qbZG9vb8hZG2oJV22ht8MRyX3QI6SSCMERbkBcNMgFh7znS6RNWRQhgiDNVZBcQ8FhiMbhWLslcLx9j/Jz+DwOdn3Dkka5RnqNqVmlQoKDIA4ywoVUjMhzEAVHeXl5RJdEz0qVtrY21v0zmQSHHiEVLUmjQPQhFXl5b6zIBUeohFEgfA4HEDqsInc4HMMyYSvVvvP+d6U0jKKUx8T34hCxGxZS4ebQhEkadUqSRhVvJWGQ4CCIg4xwDkcowRHtJtnc3IyWlhYAkcMpgL6CI9wndb2bjEUSHHqHVLTmcGhZYzwdjnCCI1SVitJzJdeQbfiFRxYqPzEMPp+AFz+WPub2BD9P2eEIlTQaECxKgsNiMUEcuaOcNBrYrj0dYRp/kcNBEESyEE+HQ0v+BqCvEAiXi8C3UW9tbY3pOkD6hlSMThrV6nBIuo2G0M1yh6HwyAIttwgA+GojsEc6W01R4Njygnd0dWWxytcVRZZip9EsS/CDiNCHI8myNElwEMRBBu9w8FNMAelmqUdipVbBoWe3Ub7VuPyTel5eHks4jIfg4LuCJnvSqBhSKS4uVkzyjIVIgqMvQuloNCGVwiMKtNwiAOCFVcHhEyXBoeRw2NXkcIToQSaGQCIljUoez5H14eCTRuM7siwiJDgI4iCDFxzyjpjJ5HDoGVKROxxms5mJm3gIDovFwvJXkrksVhAE9rrpHU4BdHA47MrPlVyDczjsxTZkjwrfAl6O1yvgpY+DH1dyHfhuo+yaKqpUspS7+wcEh0qHw5Rpgtkq3cZdbq5ihhwOgiASCR9SkW9c6SQ4wjkcQCCsEg/BAQQco1hDKnl5ecjICOHJc0TjFnV0dDBBaoTgkHcalQ/4jNQcS+JwhAipZJYFfrDwqELF0tlwbNkNNLQEP67W4Yi2DwcQweFwBAsOc3bwFp7Mrc1JcBDEQQbvcMg3R73bf4uCw2q1orKyMuLz45XDAQQER1tbW0ydVAF1gkNMHI01pKI2ryIa8WZkwigQ7HDIB/lJww7BQiErM/BYqJBKRmkGRv2+CkVHF2L0zeHLsJU4EOKlUnY4pOsx202wZCuHPlQJjn4HR21IxaIgOFxJXKWSZIYLQRBGEy+Hw+fzYfv27QD8n2SV5mjISYTD4fP50NnZKVmvVkTBkZ2dDYdDueeD6HBEIzhcLlfYTqZK8OtRK96MLIlVcx96hFQAYNQNIzHqBr974vP5NNwh0Bbi7VHjcNgK7SEdlUjuDRCFw5GjIDiotTlBEMlCOIcjlOCIxm3Yu3cva4nOd5MMhxFJow6HQ1FM6FmpEm6OiogoONxuN5vGqha+k6lah8NqtbJk1WRxOHgxpnT+iGWxKkIqsdLGRbzyA9XMEsEgEiQ4wgyI05Q06gru+2FVcE4UQypJ3NqcBAdBHGSEEhxWq1Wy4fNhCHG8vBZ2797NjtWEUwBjkkbLysoUP3XqJTi8Xi+am5sBhBccsfTi0FqhIlJUVAQArBdKJHiHwwjBsWDBAhx99NEoKyvDBx98EPR9LVUq4RyOWOAFx6CiwLFiSEUmOEI1/QICAslqAaxWZReEryqR9/1QCqkoCQ5qbU4QRNIQKqRSUlICsznwJyE3N5dtyrx4UAsvUsrLy1X9jF6Cw+l0sk021DwQvQRHS0sL+zSqxuEAYhMcWia48oJDTZ6K0SEVq9WKzz//HPX19Zg4cWLQ97WEVMINcIsFieAoDhwrhlRkORyhEkb5nw8VTgHCtzdXCqko5YtQ4y+CSDNcLheuuuoqLFq0KKZJqmp56623MH/+fKxcuTLmc4VyOJQ+OQ8bNgwAUFdXB684VUslvOAYOnSoqp/RK2lUzSd13s2JRXCoLVeNpfmX1rbmIuLkX5fLhe7u7ojPNzqkAvgnEvPClkdLa/Nug/7btXYGhBkvOFQ5HOFCKmGm4IrwYSR5aazZEfyaRapSSTaHI8lSSggiNXjmmWfwr3/9CwBw3HHH4bLLLjP0eldffTXq6+uxZ88e/PDDDzGdi3c4xHCDIAgYPHhw0HOHDRuGH374AR6PBw0NDRgyZIjq69TV1bFjtYIjJyeH3U8sDgefMGq0w6FWcCQypAL4XQ7+HpSIh+AIR6T23wXc7bd1CQC0lbyqQZPDoZA0GgqtDodc4JhMJliyLPD2BIS/YkiFHA6CSC/eeustdvzTTz8Zeq3W1lbU19cDAKqrq2Mu4ZQ7HL/73e8wfPhw/PrXvw56bkVFBTvWGlaJJqRiNpuZyxGL4FCzccZbcOgVUonG4QDU5XGI17FYLJKfjReREisLAy8hWqOrLo4ILzjKigKCRklwWDItMGcEttGwDocoOMJ0/9Qyoh5Q0YcjySwFEhwEoRGn04mPPvqIfc03tzKCHTt2sOPu7u6o+ziI8IIjMzMTDzzwAHbu3InZs2cHPVcMqQBAbW2tpuvwDoeSexIKPQRHpJJYQCo4YgnfxDukEk0OBwCW2BoOUagNGDAgqO19POA/1StVqcRbcEQKqQCAles2qqZKJdocDkA6wA0ALBHKYqm1OUGkOKtXr5bEw40WHGIvCxF+M40GPqQib7wkhxcc0TocpaWlqjpjiujtcFBIxU8kh8Pn87Hcl0SEU4DIORwSwRFdw9aIiILDbpNeL2QrdW6Amz1ESMXjEeDpj4SoFhwq2ptHCqlY468Zw0KCgyA08s4770i+3rVrl+a+ClrQW3DwDkckIRBtSMXj8bD7VJu/ISIKDqfTKblXLWh1OGIRHGorSGIJqfCiRkuog39uJIejvb0dHo//43GyCg5HRiAvoSW6bvsREQVHYa70HvpcyqFMPo8j9Gj6wHFsDof6kIrdBs1t3Y2GBAdBaOTdd9+VfO31erFr1y7DricXHHv37o3pfNE6HFpCKvv27WMdHtXmb4joURqbjA5HLCEVsfFXQUGBpgmuWhwOfh3JKjhMJhOK+l9Go0Iq4nkLcqT3ENLh4EMqMYymB1TkcGSrdziSrUIFIMFBEJrYtWsXNm/eHPS4kWGVRDocZWVlbIPT4nBEU6EioofgEF8jk8kUMgTBXyfZQyqi4JBP942ElqRRfh1G9OBQg5qNudBAweH1Cujoj5YW5EjzSEIJjpz+abSWbAscQ5QFvGrBoTGkojhLpT+HI9kqVAASHAcdb775JsaMGYMHH3ww0beSkvDuxiGHHMKOjRQcfNIooJ/gsNlsIfshiJjNZuZQaHE4oqlQEdGjvTmf/BhqhovFYmGiQw/BkZGREbbsNFqHw+PxsKRWrYJDS9JoUjgcKkIPouDo6gXcntgqtuR09ASO5Q5HqKTRkb8diarfjcTUZybDmqv8u6amrTkAZNgCIRBVIRWFpFFRqJDDQSSc//u//8O2bdtw2223aW7kRABfffUVO/7lL3/Jjo0SHC0tLUGfTPUKqUQKp4iIYZX29nbVAkAs4wVicziiqR4RBEHS1jwceoyoFwXgwIEDw8bMo83h4N9/raWqWkIq1dXV7FhLVZGeRKpSAaSJnG06J47yg9uCQiohBEdGiR2jb65CyfGh3xu1DkemDmWxzOFIspJYgATHQcfOnTsB+DedWMd/H4zwn/JPP/10dmyU4JCHUwD9HA61lSN84igfKgkH/7x453A0NzfD7fb/tQ6VvyEiuimtra1R9TdpaWlhzkFVVfhR6NGGVPjBbUY6HB9//DE7PvbYYzVdRy/EjTnDHjrhkRcceieO8gJGHlLpi6GVelQ5HEohFRVJoy4uaTTZIMFxENHT0yP5o6OmLp+QIuYxFBcXY/DgwSw/IJUER7QOB6A+rBJNW3ORWAWHlm6ZosPh8XjQ09MT9rlK8O/P6NGjwz432pAK//9Uq+Cw2+1M6IRzOPbs2YNNmzYBAKZNmxZRqBmFmuZYRvbi4AWHvEollMOhhmiqVJRCOHzSqMlugtkeukqFQipEQpFP/CTBoQ2v18teQ3ETFjeZvXv3xtyQSwklwRFrSCUWh0Nt4ij/u6bVno9VcKhpay4Sa6UKLzQjCQ49HI5oun+KPxNOcLz99tvseMGCBZqvoRdqmmMVGSg4WiUhFZOqpFE1SB2O0GE3LWWxofJFKGmUSArkmwUJDm3s27eP5b3IBQegLA5ihU8YFSsHOjs7VQ3iCoUoOOLhcGht+gXELgL4cE4yCQ6bzcZeCy2CIxaHAwiEVZqbm0OGjVJJcBTmBjZsIx2OghzAZjVBbLgaKmlUDaqTRjXkcCgJDq9XgJiaRzkcREKRx99JcGiDF2xKgsOIsIooYkwmE44++mj2eCxhlVhCKmocjliafgHSTZWvnFALL/xGjRoV9rnxFBxAIKyiJaQSq8MhCg6Px6N43d7eXqxatQqAX6BNnjxZ8zX0Qs1E1XiFVMRBcaJA0M/hCP28iGWxXGtza06wokjmtuYACY6Ding6HE6nM+YhY8lGvAWHIAhs8xw2bBgqKyvZ96INq3i9XtZNUq3zwAuO6upqNDY2hnxvu7u7sXXr1qibfgHSXhbRCA4tIkAvwWGxWDB8+PCIzxcFR7ySRoHI3UY//vhj9Pb6Z73PmzcvYqm0kfQlXHAEfq8L+q8jhlV0czhi6TSaFRAZ1tzgvuWSSbHkcBCJJF4Ox8qVK1FUVITp06dr7qiYzChVXhgpOJqbm1klUVVVlSQXIlqHQ0uXURGHw8FEwJo1a1BWVoapU6cGtR1/+eWXUVxcjFNPPZU9Fo3DoZfgsNlskvwTJWIRHIIgsGuNGDFCVQdQMY8jESEVQDmPg598PH/+fM3n1wu3JxAOCFUSC8iqVDr1/VCj6HD0C4S4OBySkErw2qzZ4UMqkkmxlMNBJJJ4ORzPPPMMenp6sG7dOjzwwAOGXCMRKDkcI0eOZI+JJcd68fXXX7PjsWPHSvIRohUcWrqM8kycOFHy9ffff4+VK1eyr1taWnDNNdewclSRCRMmaL5Hh8OB7Gx/90atgsPn8zFXaOTIkSGbfonEIjj27t3LKlvUhFOAgMPR19fHnKZI6JU0CigLjm+++QaAv8nbSSedpPn8eqE2zyGeVSpAQPzEJDiimaWi4KiY+ZCKguDgHQ6qUiESSrwcDn5jfuihhzSPNU9WlARHZmYm+wQZa7mqHP6T5+zZsyWCI9qQinw0vVr++te/4pJLLsGMGTPYY+vWrWPHt99+O9sYR48ejdNPPx0333wzLr744qjuU3Q5tAqOuro65uKoEQGxCA6t+RtAdKWx/P9T3q1QS6ReHOLvdWlpadhOqUaj1gWIX5WK9F5iC6kE3IpYGn9ZIySN8jkcyehwJGGU5+Cjr68Pvb29kj9+eiMIQkIER19fH2688Ua88MILhlwrnojrslqtkv4OgwcPRktLC/bt2wdBEHSZ0CgIAhMcGRkZOOmkkySlpvEMqQDAuHHjsGzZMuzYsYMlYn733XcAgPXr1+Mf//gHACArKwvLli3DkUceGVMuwIABA1BTU4OWlhZ4vV5YLMHxaiW0ioB4Cw55aSzfxj0UopDLz8/XNLhNJFxIpbu7m51/yJAhms+tJ2oFR7wcjvxs6b30uRD1/2/1rc0Dx0qCI2tENkwWEwSvgJyxOQB8ku/zrgg5HEQQjY2NOOSQQ1BWVibp9Kc3ra2tQaWURggOn88XJGxefPFFrFmzRvdrxRtRcAwdOlSyAYrOg57dW3/44QfWHvzEE09EdnZ2QkMqIiNGjEBeXh6AgMPxm9/8hiWJ/uEPf9ClaZTocAiCoOn3NBbBofW9468VqRpGJBqHI9rBbSLhkkb5/6uJamcu0qcy7JCZEeiPobvg6AxcP8PuFxYS1yFKl0OvabGOIZmY9sIUjH/wEAxdGPx+SRyOJLQTSHAkmAsuuAA1NTVwuVxYsWKFYddRaklthOBobGxkcXx+U/7oo490v1Y86erqYp8O+aoNALqEOuTw4RSxL0Jubi6ysrJiuk60DoeI2WzGlClTAPh7bXz//fdMKI8cORK/+tWvorovOdEmjmoVHLzDEO+QiprE0VgGt4mEczh4NzLRgkPtpgwYNzFWdDgKuMiSHt1Go6lSCRXCGXBCCSouHQZLZoQqFXI4CJ7vvvsOn3zyCfta7ZyKaFDqn2CE4OCvM2nSJHasd35DvOHfG7ng0KN6RI5S5YDJZGLiJlEOBwAmOADgnnvuYccLFy6M+pxy9BAcY8aMifh8PQSHw+FQHY7Q2m2Un/ESTcKo/OfkgoPPr0qmkEq4KhUgIDhaDBIcfNiGD4FEmziq1r2JlDQaCScljRJKCIKAJUuWSB4zsv5dScz09fVFNT8iHLzgOOKII9ixkYLD5XLhq6++wueff45vv/3WkCm4SgmjInqEOngaGxtZCGrixImS0k5R3LS1tbHeCVqI1eEAgKlTp7LjV199lR3PmzcvqvMpEavgyM3NZZ1Zw2Gz2ZgI0CI43G43q0oaNWqU6v+7WkMqsZbEAuGTRpPK4VC5KQMBQdDTB7gUykejweMR0NX/X4p3ODJVjKiPhF4hlUiQw0Eo8vbbb2P16tWSx4xsxMULDjEGb8Q1+T9ghx9+ODs2SnB4vV4cddRRmDFjBo477jgcfvjhOPvss3W/jlrBoUdI5d1332WfauVtpmMVN9FWqfDwgoP/9D1t2rSozqdENILD6XSipqYGgD/EoTa5L5oR9TU1NaysVW04BdAeUom1JBaQ5qmEC6kkk8PhsId/74yoVFHqweG/l8BxtA4HL6bCuTeRkkYjQTkchCKff/550GNGCg7+D8thhx1m2DX564waNYr9kdQrt0HOli1bWLWEyJtvvgmXK4YaNgXCCQ69Qyo//PADO5b3RdBTcEQb/hg1alRQ+eQpp5yiupJEDdEIjp07d7LkVS0iQPwdPXDgAPv5SPAzbtQmjALSzV/NumLtMgr4XRzxQ4b8/zsfUknUhFiRaHI4AOMFh2REfZR/VuIWUuGrVCKItkRAgiNBKG3A8XI44iU4hg0bJsk5MKLVOd8LQkQQhKDJuLESz5AK/17xjcWA2MWNHiEVs9kcNG9Dz3AKEJ3g0DIqnkd8/zweT9iJqjzV1dXsWP4ehYNv9a4mZ0uPkAoQCKuEcjgKCgok7ksiULspA/EVHJKk0WgdDpX5KRZLYFgcORwHCT6fD2vWrEFHR4dh1+A3i0MOOQSAfxS32u6DWhH/sBQXF0v+6BklOCwWCwYNGmRIySgPLzj4JFW1Y9TVwp9PPh9Eb8EhXstkMgXZ3LGGb/RwOABpWMVkMmH27NlRn0uJaARHNFUjACQ9VRoaGlT9DN9VdsSIEaqvpVVw6BFSAaSCQxT+Xq+XCfNILeDjgTaHI/DpXa/EUYng4JNGdRQcGXZEDPWJgiQawUFJoynIgw8+iCOOOAJTp041TACIG5PD4ZD8wVL7CUsLXq+X9XQoLy+POMwpFsTNcsiQIbBarYaUjPLw4ZTTTz+dHevd3VTcHPLz8yU5MID/PRSrHfRYo3itQYMGBTV6ilXc6OFwANJKlSOOOCKmT99KRCM4tm7dyo6jcTgA9a8pLzi0OBxDhgxhG44aUayXwyH+n/d6vWhqagLgF1diCXs0Q/b0JplCKoU5AVGQyYUmYg2phGv6JSIKhWhCKpQ0moK89NJLAPxxWr0/KYuIf9gGDRpkqAAA/P0SROFUUVFh2PV6e3vZJzIx7GBEyaiI1+vF999/DwCorKyU2Px6vm98MzN5OEVEr9CR0+lkn7KVNoFYX089kkYB4IQTToDd7v/recEFF0R9nlDk5OQwB0at4Pjiiy8A+N01NSWxItE4HGJIxWazaUq2tNvt7HpaHY5YBMfYsWPZ8X333QdA+n8kGRyOTq5gLtLGXMRpfkNCKgY5HJGEFMAJjlhDKiQ4kh+Xy4UNGzawr8VPA3rS29vLMuIHDx5suOCQW81GXU9pmqre4Qaebdu2se6pU6dOlfzR1FNwNDU1sSTUSIKjp6dH0yRQOaITBSgLjmQJqQwdOhSrV6/G888/j1/84hdRnycUJpNJ0zyVPXv2YPPmzQD81VFa8hG0/o4KgsAcjsrKSs3JsuL72tDQEDG5Wa+Qyu9+9zs4HA4AwGOPPYaNGzeGzUtKBFvrAkK9MkL+qhEORzsnOPKyAsd6JI2KgiNSfxEgIDiiuRa1Nk8xNm3aJPkjYITg4D9FxcPh4JPp+MoRva+n9AfMyJAKH06ZMmWK5I+mniEVNX+YeechlnWGazAG+JP7RKGQyJAKAEyfPh0XXHCBrtUpPKLgUFM9wneyPfnkkzVdR6vD0dTUxPrXaAmniIjvqyAIEoGpBP//MxbBUV5ejt///vcA/M7gkiVLJP9HkiGksnGX/1+zGRgT4XYKuaTOVp1G1HcHtDhyHIFjPRwOFlJR43DEkMNBSaMphrzEUuu0SjXwG0U8BId85kM8BYeRIRU+YXTq1KkoKipiY831dDjUCA69nJxwyamA/5O/+Jomqiw2XoiCw+v1Rkw4jkVwaH3v+AoVLQmjIloSR8UPPHl5eVENbuO54YYbUFlZCQBYuXIlnnzySfa9RIdUvF4Bm/v1z6ih/nkp4TDC4ejpCwiXLE6L6xJS0SI4YgmpUA5HaiHmBIgY4XDwn4DjEVKRlwvyvQDi6XAYKTimTJkCk8nErrt7927dynDjKTgiORz8tZqbmyWOhRr0dDiMRm3iqCAITHDk5ORIOtyqQavDEW2Figj/voYTxg0NDazfR1VVlebryHE4HHj44YfZ1/zfhUSHVHbuDbgA4ysjP58XHC06FRPyDkc2918j1pCK2yNAbH6sJaTidEHz3zAn13WVQiopgNzhMEJwxNvhEP+wiBNHrVYrq6pI1ZCKz+dj4rC8vJxtTuJ1+QTWWNEqOPQKqYSyuflrqU1yFNEraTQeqBUcGzZsQGNjIwDg+OOP1+wEZGdns5wPNWJx165d7DiakIpah+Odd95hx3r1OTnzzDODmsnZbDaJ6EoEG2sCx+OHR36+xOFQN3Q3IhLBESqk4tL+IUZLfxFA2t7crbFIkhyOFMLr9WL9+vWSx1JdcMhnPogleeI1jRYcfMmong7Hjh07WHIm3xPCiMRRrTkcRoZUYr1WKoZUgPCCgw+nzJo1K6priRtuPBwO/n0N9zuqNDE4VkwmE5YuXSrJuykvLzd0jpMafgq8pJgwPHKHzAx7YER9u06Co4cTHFncfw2JwxFFSEXasj3y8yO1N/9orYDH3xAU74VyOFKIbdu2BQ0zMyKHI54hlZqaGjbMjG/BLF6zra1Nt2Fn4h/P3Nxc5Ofns8eN6DbKO1G84DAicVRcl9lsDjngSu+Qit1ux8CBAyNeS6ubwrs+6SA4BEHAyy+/zL7Wmr8hIr6m7e3tEYfi8YJj+HAVH8dl8L+joRwOp9OJDz74AID/dZg+fbrm64Ri/PjxuO666xTvJ1FsrAn8XVDjcAABF6InyrwKOSEdDn6WShQhFUmXUQ05HEBwL476/QJO+Z2Aqx8W8O93EIS0tbm2+4wHJDg45OEUIPUdDnmFioh4TUEQNI/lVkIQBLYxDxs2TNJNjy8Z1at7K5+4N27cOHZspMMhNjNTQq+QinitoUOHhvzUGa242bJlC958800A/iTEZNhowqFGcLzwwgv46quvAPjH0Ytde7XChxQivaai4Bg4cGBU7cAHDhzIwj6hfkc//fRTVvI9f/583R2IO++8kzkt0Yo0PREdDqvFnzSqBtGF4J2JWOjmdCbvcMSaNCoJqagQAbyjInc4ttWB5YNsUvg8RQ5HCiFPGAWMFRwZGRkoLCyEw+FgNfJ6C45Q7Z71Fjn79+9nCYlGDzcDQidXqk3IU0tvby/b7MJt0Dk5OZryAJTo6OhAe3t7xGtF+3reeuutrLz0pptuSvkcjq6uLvz2t79lXz/yyCOqJ8TKUZsX09fXxwRlNOEUwO+UiZt9KIfDiHAKT2FhIdatW4cvv/ySlcsmCrdHwNb+l2HMMMBuU/ceipUksTgcfc6As8I7HI5QIZUYHQ6tORxywcHfo5L4kbQ2T0KHIwk1UOLgHY7Bgwdj79692L9/PwRBiPoPmRLiH6xBgwZJcir27NkTd4cD0EdwaBluxnc91ON6fExc75CKmqoRkUGDBqGzszNqwaEmYVS8jkgkN+X777/HK6+8gs7OTrz66qsA/J/mlyxZEtU9xpNIguP//u//2PoXLFgQU2JlJIejo6MDTz31FNauXcsei1ZwAP73d+fOnWhra0NnZ6fEKREEgQkOm81mmAMxYMAA9hobMVhRLTv2BJIj1VSoiMTqcFx6vw8rPgQevwG4ZK6JnScrEzCbA3/vY3U41A5uEwkXUuFdGKV7kSSNJuHunoS3lBgEQWCCY/DgwZgwYQL27t0Lj8eDtrY2SSlpLLhcLrbB8xsHLzj0FDjyHhz89UT0qOYIt1kaUakiXi8jI0OyMQ0ZMgRmsxk+n08Xh0OtCAD869y2bRs6OzvR1dUVNMJdy7XCiRu1IZW+vj7Mnj076P294447WL+SZIZ/Xzdt2iT5f7F161Y88sgjAPz5Lo8++mhM14rkcPzxj39k1xOJpkJFRJ7HwYcFN2/ezCphZs6cGTS7J934KVD0g/EqEkZFRIfD7fG7JDar+p/tcwpY1p8D8e+3BVwy18TcgyyZCxGrw6G5SoUTHPLr8W6OkrNDVSopwq5du5idPWXKFEnCnp5hFXmXURFRALjdbnR16ZR2jYDDUVBQIJnFoLcICOdwGBFSCZXrYLPZ2PX0cDi0tH8eOjQQfBancGpBrbgpLi5mOQDhXs/6+vogsTF58mRcdtllmu8tERQUFLCNeP369XjuuecA+D8cXH/99Wzw2I033hjT5g9Edjj4ShjAHxbhhwVqJVxprNHhlGRj4y4uYbRS/c9lxeA8dHC1Ae3+VBkmOPiEUUBfh8MRoaEZEL5KJZLDQdNiUwR5m2xecOhZqcJv7kqCA9Avj6Ovr49tmHxJLKB/6EFLSCVW2tvbWfKpkggQE0f3798fseIgEloEh9pyRzXXCic4TCYTe03DiUX+9+iss87C6tWr8cUXX8TcsTJemEwmSaOqG264AR0dHXjjjTdYBcewYcN0yUEI53D09vZi48aNAPxVKatWrcKePXsk1VFaCfe7ctAJjprA8QQNUSq+G6jWsEpXb/CxeI5sWWoT73DEJaSiNodDYc00vC1FkAsO3s7V0+HgN1z+k78RgqO6uprFZvlwCqB/NYdawaHn+HZAeWNWU3aoFi2CI9bras0XAfyiSvykL4f/PZowYQKOPvpolpycKpxyyik444wzAPiFwKmnniop6XzkkUeQlZUV4qfVE87h2LBhAysdnzJlCmbOnCn5nY6GUL8rLS0tbOrt2LFjY3ZuUgGxQsVuA0YqV50rIhEcGoUALzg6e/yuWTeXw8FjsZhg608+iHdIRZ7D0cMluSqV6CZ7DgcJjn74CpXJkycbFlKRl8SKGCE45C3NeWL9NC5HPIfJZAoa163XYDMRLYIjVvcmWocjGsHBl/qqyRcREbtsytFr8FeieeSRR1hFzWeffcZ+h2bNmoWzzjpLl2uUlJSwZlhyh4NvoT9hwgRdrse/v2L7cgB47733WCXRweBuCIKA6YcAE0cAk0YCVg15GHxIRavD0dkjPe5zAWLerNzhAAJCISqHQ7PgCLwG4UIqSmt2Ug5H8iMIAvujUlxcjPLy8rQIqYSqUAGArKwsltOhZ0hl0KBBsNulvmF2dja7Fr+pRkskJ0BP90b8+ZycHNYxNRSxiLj29nZ8/fXXAPzVD3zjNCXUiLh0ERzDhw/HvffeK3msoKAAf/3rX3VLrjabzSgtLQUQ7HAYIThGjx7NEnc/+ugjJjIOtnCKyWTCM7ea8eNyM755XNt7qVdIpc8FdHQHvg4nOGIti9U1pBImadRmha6VlXpBggNg5a9AYAhYOoRUQvXgEBE35vr6eng8Gpv2czidTvapMJQLIAqe+vp61tAoWiLlOujlcIRrZqZELCGVDz74gL0HajYaNXkx6SI4AODXv/416urqsGHDBmzYsAF1dXW6lFfziK9pU1OTpPuuGG41m81RNxaTk5GRgdmzZwPwf6D59ttv4fF48O677wLwC6oZM2bocq1UQesGKXE4tIZUpA2l0cT1PpSHVICAUIim02hMjb/kIRVOcChWqfT/GU/GhFGABAeA4PwNAGkXUpE7HEBgg/T5fKivr4/6OnxFRiTBAUgt5GiIl8Nx4MABNntETVfOgoIC9qlV63W1frI92AQH4K8CmjBhAiZMmKC55FgNYh6H1+tl1T1OpxM//fQTAH9OhR75IiL8+/zWW2/hyy+/RFtbGwB/7kqqJPcmiqzMgEDRHFKR5ZI3tASOFR0OUXDEXKUS+flhq1QiNf7qFyjJGE4BSHAAiCw49AypiJuz1WqVbAJGOhwDBgxQtOj12pjV5DnwDgsvhGK9XiSHI5Z18e5IpJwKwP8JTbx2XV2d6mZKXq+XTQbNycnBcccdF/FneHcslFhMN8FhNPxrKiZubtiwgSXlin8b9IJvVPbWW2/hX//6F/t6/vz5ul4rHdHT4WjkHA4lwZEZU0gl8HdAc+OvMIKjxxnIOxERHY5kTBgFSHAAkCaMin9UsrOz2acZvRwOp9OJLVu2APDPfeD7R+gtOLq6utgnX6VwCqDfxqxGcPAOR6yCQ3Q48vLyFJsi5eXlMYEVS0iFd2KqqqpU/YwoTHp7e1W/j2vWrGGfqGfPnq1qqFplZSU7DpUXQ4JDG7wAuOGGG9DX1xdySKAelJWVYdq0aQD8fUZWrFgBACgqKsKpp56q67XSET70odV5kDscjZzDoRRSER0OlxvwerV1ZY1lPH24kIrPB7hlczdFgZKMbc0BEhwAAg5HXl6epF2xmMehl+DYuHEj+7Q0efJkyff0Fhz8ZqkUTgH0y3VQ0z+Cvwc+t0QrPp+PCY5wYQ7Rvamrq2MJeVqJlAOjRDR5HNEkCo4YMYIJ1lACTvw9slqtad+tUg9OP/10nHDCCQD8jQAffPBBScKo/P+sHii933fddVfEpGEi1qRRqWhobA18HS5pFFAeGR+OWEIqckelW7bOPqc070VMGiWHI0nZv38/2xgmT54scR3EsEpzc7MuI9zlpbc8BQUF7Np6CI5QLc154hlS0cvh2L9/P1wu///CcGEO8T7cbnfYYVzhiEZwRFMaywsOtfNAMjIy2Bq3b9+uGL4Rf4+KioqSMmM92TCZTPjrX//KymPvuecevPjii+x78RAc48ePx1VXXaX7ddIRPUMqkhwOR/D/lViaf8U0SyVMWSwA9LmlWzgLqVAOR3LyzTffsGP5HxRRcPh8PrS0tCBWwgkOs9mMoqIiAPrkjITrwSGiV0hFTcOqnJwclugYi+BQ241TDzElCg6TyaR6UJfW0tg9e/bgxx9/BAAcfvjhrDRTDWKYp729XXEejig4KJyinvHjx+P6668H4A+BikmcY8aMMSRRdfLkyZIE4D//+c+wWpP042mSEYvDETakouBCxNLePKaQilv6QUIurHplDocYgqEqlSTlvffeY8czZ86UfE/v0lhecBx22GFB3xcbZu3duzdmRyVShQrgX5/YUEmPkIrD4Qi7uYnCp6mpic2t0YrabpyxiilBEJjgqKioUD3KXWtIhZ/Rccopp2i6x3CVP06nk5Ufk+DQxp133okjjjiCfZ2Tk2PYCHez2YwHHngAJSUluPHGGw2bDJuOSB0ObXkVXXLBwSeNKjTjjWWAW0xVKpFCKu6A4PD5BHj6t420cDhcLhf++Mc/Yv78+Zg5cyYuueQS9ukMAJYvX45Zs2bhxBNPxNKlSxM68lgNgiCw6gCbzYaTTjpJ8n09S2O9Xi9++OEHAP6EP6Xps3wYIFT3SLXw4YBQCY98VcXu3bujer+09KrQI6yidsBZrPkp+/fvZ6JIbThFfk9qhM6HH37IjrVuNuFeT0oYjZ68vDx8/fXXcDqdzOW4+OKLDbvez3/+c+zfvx8PPPCAYddIR2LK4ZBXqUQqi43B4dDeaZS7r1bg8TcE7G70/20OCqlwDoebn6OSpCaZJsHh9XoxePBgPPXUU/j444/xs5/9DL/+9a/R09OD1atX46WXXsLy5cvx4osv4ssvv8Qbb7xh1H3rwrZt29gY6GOPPRa5ubmS7+tZGrtjxw72iTNULFjPltziBjRkyJCwo8jFa3Z1daG1tTXk80LR2trK1hWpV4UeiaNqW43HGlKJJn8DCM7h+NOf/oT58+crVpL4fD7mcOTm5ko+VauBBIex2O122O12ltNBJBd6hlT2c4arYpUKJxS0Ohz889XkcPDPWfEhcPXDAub+ToDPJwRdm3c4kn1SLABo0kEOhwNXXHEF+3rOnDn485//jNraWrzzzjs488wz2Yjuiy66CP/73//Y4CUel8vFEv/YjVitQS2x9UCsUFCqVBDdDQCYO3du0HP4kMr27dujrnYApO2RDzvsMMVz8RtoTU1NxA0o1NpaW1tZTH/UqFFh71t+zUjtu+WITZEA/2Yb7lq807Jt27awzw21Nn7jHjJkSMhz8KPia2pqNL93mzdvZseRXkOejIwMlJSU4MCBA/jmm29YPweLxYLXX38dQGBNP/74I3POZs6cCYvFouk++cFe8teTF8hFRUUx/e5qIdz/t1SH1pZcZHKbandf6HtXWpvc4eAj2A67f3Pn4TfwjTUCpo5R7waLU10tZsBiDj63HJvCrrypBmjvCv65XqeJrauPc15s1vi+l3yxRThiMl52796Njo4OlJeXY9euXZgzZw77XlVVVcj+AMuWLcMTTzwheezcc8/FeeedF8vthEUpnv7qq6+y40MPPTTIVeA34zfeeAMXXnhh1Nf/9NNP2fGQIUMUHQx+kuePP/6II488UtW55WsTQzeAvyNlOLeEL5dct26dYqgnHP/973/Z8dixY8Nei3da1q9fr8rFka9t7dq1APyzYEwmU8hz+Hw+2Gw2uN1uVFdXa3aMxOsAQH5+vqafLysrw4EDByTt4j/66CNs27ZN0mPjlVdeYcdTpkzRfI9msxlmsxk+nw+bNm2S/LzY70V8nh7zcrQQ65TeZIbWlhy0tFoA+D9YHGjtRm1tcOI0D7+21o7BAJRtgI62faitlX4gLi/MAuD/AHrNwz7kWJowdZS62EpH1yAAdmTafaitjfz6HjgQWBfP2g31AKSDMfvcJrauRu718Hoivx56Mnz4cFXPi1pw9PX14bbbbsMll1yCnJwc9PT0SDaU7Oxs9Pb2Kv7s4sWLsXDhQumNGOhw1NXVoby8XKLCuru7sWbNGgB++/2kk04Kyj+oqKjA2LFjsWXLFqxbtw45OTlR29M7d+5kx3PmzJF0NRThGwt1dHRIwgJa1rZ69Wp2PHny5LDnmThxIjvu7e2NeE054rVMJhMuuuiisFUWpaWlMJlMEAQBe/fuDXstpbU1NzezrpqTJ0+OWDkydOhQ7Nq1Cw0NDZrXxefQHHPMMZp+fsSIERLnB/C/trW1tTj55JPZ2nhRc95552m+R8D/O7pr1y7s3r1bkkPD/z6MHDkyqnNHQ6jfyXSA1pZc5BdxX5izUVGhHDpWWltfmF4aIysHQf7f5ZqhwOotwEsfA30uM654tAzrngAqBymfg0c0TxwZZlX/DzNCFEP5bEOCHutzmtm6fNz2WZAX+vVIJFEJDo/Hg5tvvhnl5eUsxJKVlSUZytXd3S35xM4jxkbjifhpUOSTTz5hYZ25c+eGjNMuWLAAW7Zsgc/nwwcffBAklNQgCAKrUBk4cCCGDBmimFzJd4+sq6tT/R9fvja+YmH06NFhz8MrUy3XBPx5Jhs2bADgL+nky/uUyMrKwrBhw1BbW4vt27fDZDJF7A/Br239+vXs8alTp0a812HDhmHXrl0sz0SeoxMOMSfCbrejsrJS0+sSKrfkvffeYy6g0+nE559/DsDveI0bNy6qXhmjRo3Crl270NHRgebmZpZ3xOfjDBgwIO6biPx3Mp2gtSUHOVkCAH+YodcZ2dbn19bVFzrckJtlgtks/b9oNgP/uVVAa6eAj9YCbV3Asx+acPslkf/P9jr913JkqAs9FOYJyLALQRUq+5pNENcr0uc2sXV5vIHXI8OmPswRTzTfkc/nw2233QaTyYQ777yT/ZEcPny4ZKOrrq6WxJiTjY8//pgdz507N+Tz5AOWomH79u0siW/y5MkhN5aysjI2sCkWC5zPP4iU8Mgrbq1D1d5++212rLZDppjo2NbWprnBGZ8Ho6bNdLSJo16vl70WVVVVmpMG+d/7yy67jP08nzP03HPPMQdw1qxZUTfmCpU4SkmjRLpjswLif00tjb98PiGo2oNHqSwWADLsJtx3ZeD/6b5mdXkcYlWLmgoV//NMWHazCYtOAS7gCif3KBRK9rpCJI2mS2vze++9F83Nzbj//vslDWrmzZuHV199FXv27EFzczNWrFihumtiIuDt7HBjoGfMmMESKd977z3WmlwL/MYsL73lMZvNrMohlkZcYqt2h8MRsgeHSGVlJQuF8Q6CGmIRHID2ShWtgiPaqp/du3cz90tLhYrIwoULMWfOHJx77rn485//zH6/tm3bhurqajQ2NuLPf/4ze/61116r+RoiJDiIgxWTycR6cWipUpH3spCj1PhLpJAzSVs71V1PrCxRU6Ei8rNZJiy/xYxpYwKCYs/+YIHTxwkOF7c1pUVZ7L59+/D6669j48aNmDVrFo499lgce+yx+P7773HMMcfgnHPOwaJFi3DOOefgyCOPxOmnn27UfceEz+djIY5hw4ahpKQk5HNtNhtryNTW1oYvv/xS8/W0zMoQN8m2tjZ0dHRovlZbWxtL1p00aVLEroVmsxmTJk0C4K/mUFsa293djZUrVwLwhwTEc0QilqmxvJAaM2ZMxOdH63BEWxIrMmDAALz33nt48cUXkZubKxHe7777Lm655RZ0dXUBAC699FIcfvjhmq8hwlf+kOAgDjbEElYtgqOzJ/z3wzkRvOBo64p8LUEIlLKqdThCXa9eIQc0pOBIh7LYQYMGSZwBOYsXL8bixYtjvimj2b59O/uDr+aT8oIFC1g1xltvvRXUkTQc7e3t+OyzzwD4kwnHjh0b9vnyTXLChAmqrwVIu5mqnW45efJkJqTWr1/PBliFY+XKlXA6/V7hggULVIcEom3+1drayhJvDzvsMFXtn6PtNhqr4JAzb9481qlyyZIlrFwtPz8f9913X0znVuNwiC3zCSLdYA6HhpCKvMuo5HyZCMrf4Mnn8jDlDocgCPjvSr+Dcuk8/3m0tjWXwwuOPQqtoPpChVSSVHAkX1ZJHNBqzZ9yyilsQ+VzP9TwwQcfsPJINRtzrC25+bVNmTJF1c/wjch4wRIKQRDwyCOPsK/VhlMA6QauJaQSzZhwPglXrbgRBAHPPfcc+/qQQw5Rd4NhmDhxIguV8bXxd911l6S5XDQMHz6cldq+++672LdvH4CA4MjNzY17gjZBxItoHA55Dw7J+SKIAovFhPz+KhK54FizGbjwLgFX/EnAm/72O5qbfsmRCA7FHI7AFu7iO42S4Ege+M1LzaZcXFzMNsoNGzZoyuPQOnpcT8GhxeEQUSM43nvvPdZXZNSoUZL+K5GorKxkSZRaHA6t7xngT94UK6XUrAvw91v5+uuvAQDjxo3T3P1TCZPJhCeeeALHHHMMJk+ejMmTJ+OSSy7B1VdfHfO5bTYbLr30UgD+brE333wzABrcRhwc8A6H2tEM8i6jPKESRnkKQwiOzVya2PfbA9UzIo4YBccBhfFToXM4knM69EEpOGLZlF0ul6QKJBxer5dVJuTk5OC4446L+DN8SCWaShVxY87IyMC4ceNU/cz48eNZiCLSxuzz+dimBvhHeIuVNWqw2WysFDfUWHU5fM8UQP17ZrFYcOihhwLwV01FGhjn8Xhwyy23sK/vvfde3dpaz5kzB59//jm+++47rF27FrfffrtuU0Hvvvtu1rDtmWeewZdffsmmG5PgINIZ0eHw+aQbbjjChVSU5qjIEUVAa5dU5PCVL+JsFq2D20JdKxQUUklyfD4f25SHDh2q2tLW6gIAwJo1a1iL8dmzZ0u6TIYiFoejo6ODhSkmTZqkWghkZGRg/PjxAPwdKsVyzebmZkybNg1Dhw5lFSwrVqxgA/umTZuGc845R9M9AoGwSnd3NwsBKOHxeDBnzhzk5OTg5ZdfBgBkZmaqFlKA9H3jO7Aq8eyzzzIxOWPGDJx22mmqr5NIiouLcffdd7OvL7/8cha6IcFBpDPSibHqfiZsSEWD4PB6peKFr34Rp8/qGVJRItWSRg86wbFz505W/aH2kzIQneB477332PH8+fNV/UwsA9z4+1IbdhAR1+f1elkzrz/84Q9Yt24d6uvrcemll6K9vV3ibtx///1RN6sSCRdW+fLLLyXj2wF/gzEtzoCW923ZsmXsONq1JYqrrrqKdY3lHTgSHEQ6E80At7AhFQ2CA5CGVbr7Am6HXg5HjiPQa0QJvg8HX31DDkeSEE1SJRCd4OBHj8+ePVvVzzgcDjY0TqvDEU1ipYh8fT/++CP+9a9/SR475phjsHfvXgB+ARWup0g41PbiWLVqFTuePn06zjnnHPzlL3/RdC2171tLSwsbtDZ69Ggce+yxmq6TaKxWK5YuXRr0OFWoEOlMNIKDdzgKZG3EYxEcPQoOR6yCw2QysZwRJXiH47MfAoJnXKX2a8WDg1pwaNmUS0pK2ATS9evXR5zE197ezvIOxo4dK5leGgnR5aivr5cMAItEtGsDpBvzF198gV//+tdBaxTng9jtdsXNTS3henH09vZCEAQIgsAEh81mw8qVK/HSSy+p7vchMnHiRJaHEU5wvP/++/D2j4zUUnWTTJxwwglBIS5yOIh0JpqQCu9wlMn0uJaQCiB3OALHeoVU5NeTIwoOn0/Ah/0dK3KzgCPHR3ctoznoBMe3337LjqPdlDs6OrBr166wz/3kk0/YBnbyySdruo6YOOrz+diwskgIgsB6pNjtdpaToRZ+I//Pf/7DNvvhw4cHTfG98cYbY2pbHyqkcvvttyMrKwuXXXYZNm7ciD179gAAjj/+eE1zUHgyMzNZaeumTZtY7xA5WquJkpWHHnoImZmBv5okOIh0JjqHI+AEDJL991DncARcBYng6JUed/cKMocjuhBteMHh38J/2AHsb/M/duIUwEZVKonnu+++wyeffAIAKC8vR1lZmaaf1xJW4cMpWgUHv5mHa7Qmv97WrVsB+O9Ta++FvLw8HHbYYUGPP/TQQ3j00UfZhj9s2DDWxCpaysvLWQKtGFL5/vvv8X//938A/LkUS5YsYc+PVQCI75vH4wma4io+/u677wLwvw7HHHNMTNdLJBUVFbjpppvY13wnUoJIN6JKGuWEQXSCI3AcyuEA/C5HrCEV+fVExKRQMYfjg8DnaMyenpxiAziIBIcgCPjtb3/LvuaP1aJFcIjJjhaLRVNnUkA6b0XcCMMhL1X91a9+pel6IsuXL8ell16K8847D+eddx7+/ve/48wzz8SgQYPw/vvv47rrrsMHH3yArKysqM4vYrFYmKiqrq6G1+vFkiVLJCVmojAE1CfchiLS+/bVV1+xlu6nnHKKpjLfZOTWW2/F3Xffjfvuu0917hBBpCJZmYHNNZqk0cGyqRaqQipcTkVYwdFiXEilJN//r5MJjsDfztnTo7tOPEjSES/689Zbb7FNbOTIkbjmmms0n4PfuFavXo3vv/8ew4cPZ8PdfD4ftm3bhvr6euY2HHnkkcjLy9N0nZkzZ8LhcKC3txfvvPMOBEEIWzHxwgsvsI108uTJQSEQtUyaNAlPPfWU4veOOuooHHXUUVGdV4lRo0axEMfdd9/NRrXLOeSQQ2KeOsy/b6tWrcK0adMwevRoJpzSJZwiYrPZ8Ic//CHRt0EQhiNxOKJIGh1ULB35rt3hEAD4/zbLJ9A2tgK9MbY2l18PAMxmf7Lr3gN+h6O7F1jtLyzE8EHAyCHRXSceHBQOhyAIkjDAAw88EFW752HDhrEGS5999hmmTJmCiooKbNq0CQBwxRVX4JBDDsGsWbPYz2gNpwD+vIMTTzwRgH9gXrj+EU6nE7fffjv7+oEHHoDZnPxvK584+sc//pEdK00hjhU+VPT8889j8uTJGDNmDJqamuDz+fD6668D8GeEz507N+brEQQRHyQ5HFGEVORJo9mOyOGIkFUqsus3tsTeaVR+PcAvisR197lM+GR9oAfH7OlI6nL+5N+ZdGDjxo2sN8GRRx6Js846K6rzmEymoMFmHR0dWLJkCVatWoV///vfQT9z6qmnRnUtfuMLF1Z56qmn2FCzk046KSqBkwiOPPLIoMdOPvlk3Hjjjbj++uvZY2effXbM1yooKJC4HACwZ88e3HbbbVi2bBnLIznmmGPCTg4mCCK5iLUPR1CVigoXQiI4uImxSg6HLiGVHKmAyM4M3KdPMGFloDgRJ09LXrEBHCQhFb6fw8UXXxyTAvzLX/6CQw45BC0tLXjzzTdRX1+Pjz76SJIbcNZZZ6G0tBQnnHCC5gZcIrzgeOeddxQTNXfv3o2//e1vAPx5EfxAtWTnjDPOwN///nfWZKygoAC/+c1vYDKZ8MADD6CyshJWq1WXWSYA8Nxzz+Gpp55Cd3c3nn32WXR2duKJJ57Aiy++yJ7DOy0EQSQ/sXQazcoEG8QmomqWisqk0YYWgeVaAPqFVLIypefawrVrmpTkOeIHneCINQFxyJAhrJrixBNPxLnnngsgMCzr8MMPx0svvRRzWEMcZb9lyxZ89dVXePvtt4Naoy9dupS1Ib/uuuvY3JBUwGw2h8yjsVqtuO6666KaJROKsWPH4sEHHwTgHyB30003QRAEtLW1AQDOP//8IPeKIIjkJhaHI8fh71khOZ8KUVCgIWmUzwnRS3DwDgcAVHOdE+RVN8lG2guOhoYGNvvj0EMPlbQOj5Wzzz4bxx13HD777DMA/pDLY489plsOxdy5c7FlyxZ4vd6wyYwDBgygT+caWLJkCf71r3+huroaAJCVlYWHHnoowXdFEIRWHJKk0UACZzjEHI7cLL/o4FHjcFgsJuRlC+jojlwWy1fB6FWlInc4ahr8/+bnqMtBSSRpn8MhTmsF9K9AMJlMWLp0KUtyvPLKKzF9un41SaJ7Eon777+fVcoQkcnIyJCEn+644w5NnWAJgkgOYunDoeRwqKlSAbiJsf2Cw+sV4HRJnxOUNKqnw8Hdp8ffXxKDk9zdAA4Ch8NIwQH4KyA+/fRTbN68GRdffLGu5z7qqKPwv//9D998843i9wVBQGlpKRYtWqTrdQ8GTjvtNLz77rtobW3FBRdckOjbIQgiCrSGVNyegDDIcQQLDDUhFcAvAmobgLb+EfVydwMwrvGXPKQiIu8pkoykteBwOp2s42dJSQkOP/xwQ64zY8YMzJgxw5BzL1iwIKRQ8vl8qK2tTeoyqGTmlFNOSfQtEAQRA1odDr4kNjcLMJtNyHYIrMJETUgFCDT/cnv8QkdJ7HT2SEMuRoVURFJBcKR1SOWzzz5DV5e/bmnu3LlsiBdBEASR+mh1OPimX2L+Ri4nMtSGVOSJo0oOBwDUNgaOo3U4crOkI+r9IZXgD5mpEFJJa8HBd5CMtTqFIAiCSC60Nv7ie3CIgoNPHFXT2hwILo0NJTjEhE4gesFhMpkkAifboXyuIQOS3+lO65DKnXfeiSOOOAIvvvgizZQgCIJIM/junVodDjFhlE8c1Zo0CvgFhy3ETsrfU0YMI5oKc4Dmdv9xVkaIHI4UcDjSWnAUFhbiggsuwFFHHYX8/PzIP0AQBEGkDFarCXabAJc7eofjkArg++3A0AHqXQj/iHr/DJbWTmnux4CCwKh4kQy7P18kWniBk+0wpWwOR1oLDoIgCCK9ycrwzxLpVZM0KnE4/ALgwWtNGFcJzD1CvSgI1d4cAEYMDhYc0c5RUbpeVoZy6CcVBEda53AQBEEQ6Y24+aoJqSg5HINLTLj1YhOmjFHvQITL4Rg5OPj50VaoKF0v26EsYORzYZIREhwEQRBEyiLmM6gJqYh5EABQlBv6eZGQj6jnrz1CQXCUxBjRlwiOzGCHoyQfyLAnf9IoCQ6CIAgiZdHicLR0Cuy4OAYREORwcM7J8EHSjX9AAfDIL2ITA/KQijyHIxXCKQAJDoIgCCKFEQVHrxPw+YSwzzXG4ZCGVIrzgKtO8yeKXjoP2PwfE06eHpvgmNYf7jGZ/BNh5VUqqSI4KGmUIAiCSFn4zXfjLmDiyNDPbe4IHOvqcPQFhE62A/jnDWb87dcCLBZ9whxnHAu8fo8JJQVA1VATNtdIhVUqlMQC5HAQBEEQKcyw0sDxcb8U8PF3oV0O3uEozov+mkGdRrmQitjLQy+xAfirZ04/1oSjJ/rPKc/hSBWHgwQHQRAEkbLcd6UJk0f5j9u6gDk3CKiuVxYdLf2zTSwW/zj3aLFZTazKpUUWUlE7jyUW5DkcqdBlFCDBQRAEQaQwZcUmfPZXE+Ye4f/a7QFe+0z5uaLDUZiDmIdeDijw/7u/TSY4VHYrjYWgHA4KqRAEQRCE8eRkmfCHiwMCorZR2eEQczhiyd8QKe3ve9HSAbRzzb/iITioSoUgCIIgEkTloMAxPzRNxOX2j4wHYsvfECktVL6e2gFwsWCxmGDnZrOQ4CAIgiCIOFFWBLYJ1yoIjha+QkVnwbFzb+A4Hg4HEAirmM3AwIL4XDNWSHAQBEEQKY/ZbMKwgf7jmgZAEKRhFTFhFNA3pAIAfS7/vxl2fatTwiGW5g4p8Q+xSwVIcBAEQRBpgRhW6ezxl6vy8D04Ymn6JVJaGLzJx8vdAIDfXwRUDHTjjsXxu2asUOMvgiAIIi2oLAsc1+wDCnMFbKkFbF5ZD4782B2BUoVhafEUHJfNB2ZN2IuKior4XTRGyOEgCIIg0oLKMr5SBbj5nwLGLwIWPzxQ2mVU5xwOkXgKjlSEBAdBEASRFvAOx659wPL3/MdfbHRIEjv1zuEQiUeFSipDgoMgCIJIC/jS2JXrBDS1Br5e/WPgWJ8cjuDHyOEIDwkOgiAIIi2o4OaqfPCt9HtrtgSO9XA48rL9VSk88WhrnsqQ4CAIgiDSgkHFgK2/FMLjlX7P5Q4c65HDYTKZglwOcjjCQ4KDIAiCSAssFpNkemwo9HA4gOCwCgmO8JDgIAiCINIGPnFUCUcG4MjQp1GWPHGUQirhIcFBEARBpA2RBEeRDuEUEbnDIZ/iSkghwUEQBEGkDXwvDiAwRl5Ej/wNEXI4tEGCgyAIgkgbKmQOx6JTpF/rKjhk7c2zM1NjpkmiIMFBEARBpA18SKV8IDB7uvT7eiWMAsBAShrVBAkOgiAIIm0YMwywWPzHJ0wGxldKv69H0y+RoCoVCqmEhYa3EQRBEGnDwEITnroR+PQHAX9cbEJpkYD8bC/au/0qRE+HQ57DQUmj4SHBQRAEQaQVi+aasGiuP5/C5xNQNdiNddv7BUeefnkW5HBog0IqBEEQRFozanCgzaieDkdhLmC1BL6mHI7wkOAgCIIg0pqpo53seFyFfuc1m02SxFESHOGhkApBEASR1px+VDfM9hIU5ppw+Dh9S1dLC4G9B/zHFFIJDwkOgiAIIq2xWoBfnet3JPSGTxwlhyM8FFIhCIIgiCg5e6ZfxEyq8vf9IEJDDgdBEARBRMnlC0yYNRUYXGKMg5JOkOAgCIIgiBioHERCQw0UUiEIgiAIwnBIcBAEQRAEYTgkOAiCIAiCMBwSHARBEARBGA4JDoIgCIIgDIcEB0EQBEEQhkOCgyAIgiAIwyHBQRAEQRCE4ZDgIAiCIAjCcHQVHK2trViyZAmOOeYYnHXWWVizZo2epycIgiAIIkXRVXA88MADKC4uxkcffYQlS5bg97//Pdrb2/W8BEEQBEEQKYhus1R6enrwySef4I033kBmZiZmzpyJkSNH4tNPP8Vpp50mea7L5YLL5ZLeiNUKu92u1+0wfD6f5N90gtaWmtDaUhNaW2qSrmtLpnWZzeq8C90Ex+7du5GVlYXS0lL2WFVVFXbu3Bn03GXLluGJJ56QPHbuuefivPPO0+t2gqirqzPs3ImG1paa0NpSE1pbapKua0uGdQ0fPlzV83QTHL29vcjOzpY8lp2drRhSWbx4MRYuXCi9EQMdjrq6OpSXl6tWYakCrS01obWlJrS21CRd15aK69JNcDgcDnR3d0se6+7uRlZWVtBz7Xa7IeJCCbPZrFp9pRq0ttSE1paa0NpSk3RdWyquSzdZNGzYMPT09KCpqYk9Vl1djREjRuh1CYIgCIIgUhTdBEdWVhZmzpyJxx9/HH19ffj888+xY8cOzJw5U69LEARBEASRopgEQRD0OllrayvuuOMOrFu3DqWlpbjppptwxBFH6HV6giAIgiBSFF0FB0EQBEEQhBKpkdpKEARBEERKQ4KDIAiCIAjDIcFBEARBEIThkOAgCIIgCMJwSHAQBEEQBGE4JDgIgiAIgjAcEhwEQRAEQRhOWggOj8cDAEjHliJKw+8IgtAft9sNAPB6vQm+E/05cOAA+ztJEIkiZQWHIAjo7u7GH/7wB7z88ssAAJPJlOC70o+Ghgb84he/wAMPPICOjo5E346uNDQ04G9/+xteffVVtLS0JPp2dGXfvn14/PHH8c4776TV+9bQ0ID//Oc/+Pjjj9HV1ZXo29ENQRDQ29uLW265BY899hgAwGKxJPiu9KOhoQHXXnst7rnnHjQ2Nib6dnSloaEBDz30EJYtW4bq6upE346u7Nu3D88++yy++OKLRN+KrqSs4DCZTDhw4AA+/vhjfPvtt9iyZQuA9HA5tm7diquuugrl5eW4/vrrkZ2dDSA91vbSSy/hZz/7GTo6OvDyyy9j6dKl2LFjR6JvSxc+/fRTXHDBBWhqasKzzz6L+++/H59++mmibytmHn/8cZx//vnYvXs3/vGPf+DRRx/Ftm3bEn1bumAymeB0OrFmzRqsXbsWa9euBeAf/Z3q1NfX4/rrr8ewYcNwxx13YNCgQQDS4+/I//73P1x44YUQBAHff/89/v73v+Obb75J9G3FhPi+vPPOOzj//POxdetW3H///bjvvvuwYcOGBN+dPug2nj4R1NTUYOTIkaioqMD777+PsWPHpoXLsW7dOsyYMQM33XQTAKC5uRnFxcUpvzaPx4M1a9bghhtuwPz581FdXY377rsPPT09ib41XVi7di3OO+88XHfdddi7dy/ef/99LF++HBMnTkRRUVGiby8qPvnkE2zZsgXLli3DiBEjsGPHDvzlL3/B3r17MXr06ETfni7U1tZizJgxmDp1Kl544QVMmzYNZnPKfhZjrF27FmPGjMHNN98MwC9AhgwZkvJ/RwBgzZo1uPzyy3HhhRdi3759WLp0acqHn8X3Zd26dbjyyitx0UUXYceOHXjllVfw/PPPo6KiAnl5eQm+y9hImf9VYnwVCORsDBkyBBMnTsSwYcOwZ88e9mky1WKw/Nq8Xi86Ojowfvx4bNmyBRdddBFuu+023HPPPVi/fn3ibjJK+LXV19ejtrYWxcXFAICRI0fCarWm/KdJQRDQ19eH7u5u5OTkwOv1YvDgwZg9ezYGDhyIf/7zn4m+xaiprKzEnDlzMGLECLjdblRVVQFAWlnYQ4YMgd1ux6hRo9DX14fXXnsNAFI250H8/+R2uzF06FBs3bqV/R259dZb8d577yX4DmOjubkZzc3NyMjIAAAMGjQIvb297OtUg/89a2lpQWNjI8rKyiAIAqqqqjB79mz4fD68+OKLCbxLfUh6h6OtrQ1/+tOfIAgCxo0bh3PPPReZmZkAgE2bNsHj8WDOnDnYuXMn1q1bhzFjxsBqtaKkpCTBdx6ZUGvbu3cvGhoasH37dixYsADTpk3DM888g//85z/Izc3FyJEjE33rEVFaW0VFBSZMmID3338f27Ztw7p167B161Y8/fTT2Lp1K0466aSUed9eeeUVzJkzB0OHDoXJZEJmZibMZjNqampYDsDgwYNx6qmn4vHHH8eOHTvYZp2syNcF+AVHZWUlAMBqtbJ/xd9BQRBS4hOz0tpEtm/fjqKiIhx11FGoqanBZ599hqlTpyIjIwOlpaVJv0b52kR3pqmpCa2trXjjjTewYMECnHjiiXjllVfw3HPPYejQoZgwYUKC7zwySu9bcXExDjvsMPz000+or69HdXU11q9fD0EQsH79epx11lkoLy9P8J1HprW1FX/7299gMplQVVWFM888E0VFRfD5fFi3bh1mzZoFAJgwYQJ27NiBr776irlUqUpSOxxdXV24/vrrYbVaceyxx2LVqlX4/e9/z5LxKisrYTab4XA4cPjhh+Obb77BhRdeiK+//jrBdx4ZpbXdeOONAIDLLrsMb7/9NjZs2IBzzjkHVVVVuPLKK1FYWIjvv/8+wXcemVBr8/l8uOmmm3DOOefgs88+Q25uLlatWoWTTjoJP/zwQ0rkO3z66ae4/PLL8c9//hMfffQRnE4n+95FF12E999/H5s3bwbgt0grKysxfPhw1NTUJOiO1RFqXWJcWdx0u7q6sG/fPiYMk3kjFgm1NtEJKC8vR1tbG6xWK4466ig0NDTgnHPOwapVqwAk9xrD/T6edtpp+OCDD/D1119j7ty5GDhwIC6++GJMmjQJK1euTOBdqyPc2i699FJcccUV2LVrF7KysvDpp59i4cKFaG9vx+uvv564m1bJW2+9hQsuuABWqxWjRo3CW2+9hXvvvRcAcPnll+PVV19FW1sbAMBms2Hs2LGwWCwpHzZKasGxY8cOWK1W3HXXXZg3bx7+/ve/o76+Hm+++SY8Hg/27dsHANiwYQMeeeQR9PX1YcKECZg+fTqA5E6OUlpbQ0MDnn76aQwZMgRnnHEGWlpa2BqGDh2K3bt3s0+Zqbi2FStWwO12o6CgAC6XCzfeeCNMJhMWLFiAlpYWFgpL5rW1trbiggsuwD333IMPPvgAu3btAuC/ZzH88Oc//5mJ4qFDh6ZEgmWodck3261bt7L/ZwCwfv167N27N+73qwX52nbu3AkAzAmoq6tDUVER1q5dixtvvBFutxtjx47FUUcdBSB1fx+HDh2KM888E319fejs7AQAZGdno62tDfn5+ex5yUqotQH+SiKLxYKenh5cfPHFAIAjjjgCPp8PdrsdQPKuraurCzU1Nbjuuutw880347zzzsMtt9yCDRs2oLm5GVOmTMFRRx2F22+/nf3MhAkTsGHDhpTPd0sqwdHY2Iimpib2dX5+Pmpra5nSczgcuPjii/HZZ59h+/btGD58OF5++WXccsstuPLKK/HHP/4Rdrsdq1evBpBcn0zUru2LL77A9u3bcd1116GrqwtPP/009uzZg/r6eng8HmYVpuLaVq9ejbq6OuTn52PLli3MCairq0NXVxfLok/mtZ1yyik48cQTMXv2bAwZMgRvvPEGurq62D3ffPPN6OrqwuOPP46ffvoJtbW1yMzMxODBgxO1BEXUrktEXF9dXR3mzp2L+vp6XHDBBXjwwQeTLgcn0trefPNNydpGjhyJ119/HXfeeScuv/xyPProo6ioqGC5Dqn8+3jttdeisLAQ//3vf7Fhwwbs3bsXu3fvRkVFBYDUXBvgv++ioiLs2LEDO3bsYB9At23bxkIOybo2h8OBE044ATNnzmTf7+joQH5+PqtIvP3227Fx40Y8+eST2Lt3L6qrqzFs2DCUlZUl5P71wiQkgQx0u9244447sH79egwYMADHHnss5s+fj8LCQtx6662YOHEiLrnkEvb8q6++GtOmTcMFF1yAtWvXYurUqcjNzQUAbNu2Lamy57Wu7ZprrsHEiRNx7bXX4vPPP8eqVatQW1uL2tpa/PznP5c8N9FEs7ZDDz0U11xzDf71r3/h3XffRVlZGbZt25YSa5s3bx4GDx4Mr9cLi8WCHTt24KabbsJvfvMbzJgxAz6fDxaLBT/99BPee+89/Pjjj6ivr8fChQtx6aWXJnpJAKJbl8lkgs/ng9lsxl133YX//e9/yM7OxkUXXYTLL7880UtiRLs2t9uNH374AWPGjGF/R5ItVh7L7+PGjRvx0UcfMQF8wQUXJM3vIxDb2l577TW8+eabsFgsqKmpwcKFC7F48eJEL4mhtLYFCxYw4SCGKtevX48nn3wSf/7zn2E2m2GxWLB69Wq89957qK2tRV1dHRYvXoxFixYleEUxIiQBb775pnDttdcKbrdb+PDDD4Xbb79d+NWvfiUIgiA8/fTTwo033ihUV1ez53/wwQfCaaedJjmH1+uN6z2rRevaPvzwQ2HBggWSc+zYsUPo7e2N632rIZq1nXrqqezrmpoaYdWqVUJ3d3fc7z0S8rXdcccdwi9/+Uv2fY/HIwiCIDz88MPCr371K6GpqSnoHLt370669y3WdS1ZskS499570/Y9S5W/I9Gsbe/evUn3+ygIsa/twIEDwldffZWSv5M+n08QBEF46KGHhHvuuUfxHJs3bxZ6enricr9Gk7CQSl9fH4ux7dixA3l5ebBarTjppJNw+eWXY/v27Xj77bexYMECZGRk4Pnnn2c/m5+fj6qqKvT29rJzJFPdfCxry8vLw+jRo9HT08PKpUaOHMkqcxJNrGsbNWoUuru74fP5UFFRgRNOOAFZWVmJWo6EcGu77LLLUFNTg5deeglAoPT6iiuuwP79+7F+/Xq8/vrreOqpp9j5ysvLk+J902Nd//rXvwAA999/P37/+9+n7XuWKn9HolnboEGDkuL3EdBnbU8++SQAf9XKkUcemZK/k263G4IgYOPGjZg9ezYA4L333pMkvo4dOxYOhyPu6zCCuJfF7t69Gw899BCysrLgcDhw4403Ijc3FxaLBZ2dncjNzUV5eTkuv/xy/O1vf8O8efNw1lln4d5778Vdd92FU045BY899hgmTZqUdG+CnmtLlv88InquTYxTJgtq13bZZZfh73//O84++2yWmJabm4vDDz8ct9xyCzIzM1mztmTAiHUly4aVru8ZQGs72NbW09ODgoICtLW1YcmSJdiwYQOrWEw34irnX3/9dVx99dUYPXo0LrroImzduhVPPfUUqqqq8O2330p6/Z9wwgkoLy/Ha6+9hsMOOwx33XUXMjIy8I9//ANTp07Fb3/723jeekRobX7SeW3HH388RowYgVdeeQWAv2HPP//5T6xYsQJXXXUVPv/8cyxYsCBRS5GQrusCaG0itLb0WNvOnTvx+eef4+6770ZVVRVWrVqFU045JVFLMZZ4xm/+9re/Ca+99hr7ura2Vjj66KMFl8slXHvttcKjjz4qtLa2CoIgCN3d3cKdd94pPPvss5K4qsvliuctq4bW1ioIwsGxthUrVrDY62effSY0Nzcn4M7Dk67rEgRaG60t+Yj2b6QgCMLGjRuFxx9/nH0/nYmrw3H22WezUiC32w2LxYLhw4fD4/Hg8ssvx7p16/DJJ5/A6XQiKysLbW1tKCgokMRVbTZbPG9ZNbS2g2dt+fn5rOTu2GOPTco5Kem6LoDWRmtLPqJdG+DP0bjyyitRUFCQwBXEh7jmcJSWlgLwlwLZbDYcOHAAJpMJdrsdkydPxmmnnYb3338fq1atgsfjwd69ezFu3Lh43mLU0NpobclEuq4LoLXR2pKPaNY2fvx4AMmVqGw0CZmlIirWNWvWYPjw4Wz2xNlnn41jjjkGX3zxBTo7O5OqL4NaaG20tmQiXdcF0NpobclHOq9NDxIiOMRmLtu2bcPJJ58MAHjxxRfR1dWFSy+9FGeffXYibksXaG2pSbquLV3XBdDaUhVa28FLQrwci8UCj8eDvr4+NDY24oorrsDTTz+dEtMLI0FrS03SdW3pui6A1paq0NoOXhI2nn7nzp34+uuvsX37dlx44YVsAE86QGtLTdJ1bem6LoDWlqrQ2g5SElUe43a7hWeffVbo6+tL1C0YBq0tNUnXtaXrugSB1paq0NoOTpJieBtBEARBEOnNwVOPQxAEQRBEwiDBQRAEQRCE4ZDgIAiCIAjCcEhwEARBEARhOCQ4CIIgCIIwHBIcBEEQBEEYDgkOgiAIgiAMhwQHQRCGceWVV2LatGm48sorE30rBEEkGBIcBEEQBEEYDgkOgiAIgiAMJ2HD2wiCSC86Ojpw77334vPPP0dBQQEWL14c9Jy33noLzz33HPbs2QMAGDhwIA455BDcfffd8b5dgiDiDM1SIQhCF373u9/h448/BgBUVFSgqakJANDb24spU6bghhtuwMKFCyEIAsrLy2G327Fv3z709PRg7dq1ibx1giDiADkcBEHEzJ49e5jYWLRoEX75y1+ipqYG559/PntOXV0dBEHAsGHD8PLLL8NsNsPr9eKHH35I1G0TBBFHSHAQBBEz1dXV7PjEE08EAFRWVmLUqFHYsmULAGDSpEnIy8vD7t27cdJJJ2HYsGEYPXo05s6dm5B7JggivlDSKEEQcaGkpAQvvvgifvnLX+LII49ET08PXnvtNVx99dX46aefEn17BEEYDDkcBEHEzIgRI9jxJ598gvHjx6O2thbbt29nj+/fvx9tbW1YtGgRe+ycc85BTU0N1q9fjwkTJsT1ngmCiC8kOAiCiJny8nIcf/zx+OSTT7Bs2TJ8/PHHaGxshMVigdfrBQDs3LkT1113HQoLCzFgwAB0d3ejvr4eAFBVVZXI2ycIIg5QSIUgCF247bbbcOKJJyIjIwNdXV246qqrJK7FkCFDMHv2bGRnZ6O2thatra0YPXo0br31Vhx55JEJvHOCIOIBlcUSBEEQBGE45HAQBEEQBGE4JDgIgiAIgjAcEhwEQRAEQRgOCQ6CIAiCIAyHBAdBEARBEIZDgoMgCIIgCMMhwUEQBEEQhOGQ4CAIgiAIwnBIcBAEQRAEYTgkOAiCIAiCMBwSHARBEARBGM7/A8Gji1vVLP3BAAAAAElFTkSuQmCC",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "ice_heater = IceCreamHeaterDataset().load()\n",
+    "ice_train, ice_test = ice_heater[\"ice cream\"].split_after(0.8)\n",
+    "heater_cov, _ = ice_heater[\"heater\"].split_after(0.8)\n",
+    "horizon = len(ice_test)\n",
+    "\n",
+    "model = NeuralProphetDarts(n_lags=horizon, n_forecasts=horizon)\n",
+    "model.fit(ice_train, heater_cov)\n",
+    "preds_cov = model.predict(horizon)\n",
+    "\n",
+    "model = NeuralProphetDarts(n_lags=horizon, n_forecasts=horizon)\n",
+    "model.fit(ice_train)\n",
+    "preds_no_cov = model.predict(horizon)\n",
+    "\n",
+    "print(\"MAPE with lagged regressor: \", mape(ice_test, preds_cov))\n",
+    "print(\"MAPE without lagged regressor: \", mape(ice_test, preds_no_cov))\n",
+    "\n",
+    "# for some reason results vary a lot here for model with lagged regressors\n",
+    "\n",
+    "ice_train.plot(label=\"train\")\n",
+    "preds_cov.plot(label=\"cov\")\n",
+    "preds_no_cov.plot(label=\"no cov\")"
    ]
   }
  ],

From bed370071ad076e529fd1b80bb7e4166a78d24ca Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Mon, 19 Dec 2022 14:52:30 +0100
Subject: [PATCH 06/15] Add future covariates

---
 .../forecasting/neural_prophet_model.py       |  104 +-
 neural_examples/examples.ipynb                | 1669 +----------------
 2 files changed, 181 insertions(+), 1592 deletions(-)

diff --git a/darts/models/forecasting/neural_prophet_model.py b/darts/models/forecasting/neural_prophet_model.py
index f40cb0583a..dd3eed7142 100644
--- a/darts/models/forecasting/neural_prophet_model.py
+++ b/darts/models/forecasting/neural_prophet_model.py
@@ -3,6 +3,7 @@
 ------------
 """
 
+import warnings
 from typing import Optional, Sequence, Union
 
 import neuralprophet
@@ -17,6 +18,7 @@
 class NeuralProphet(ForecastingModel):
     def __init__(self, n_lags: int = 0, n_forecasts: int = 1, **kwargs):
         super().__init__()
+        # TODO improve passing arguments to the model
 
         raise_if_not(n_lags >= 0, "Argument n_lags should be a non-negative integer")
 
@@ -36,7 +38,7 @@ def fit(
 
         raise_if_not(
             series.has_datetime_index,
-            "NeuralProphet model is limited to TimeSeries index with DatetimeIndex",
+            "NeuralProphet model is limited to TimeSeries indexed with DatetimeIndex",
         )
 
         raise_if_not(
@@ -48,11 +50,16 @@ def fit(
         fit_df = self._convert_ts_to_df(series)
 
         if past_covariates is not None:
-            fit_df = self._add_past_covariate(self.model, fit_df, past_covariates)
+            fit_df = self._add_past_covariates(self.model, fit_df, past_covariates)
 
-        # TODO add future covariates to df
+        if future_covariates is not None:
+            fit_df = self._add_future_covariates(self.model, fit_df, future_covariates)
+            self.future_components = future_covariates.components
+        else:
+            self.future_components = None
 
-        self.model.fit(fit_df, freq=series.freq_str, minimal=True)
+        with warnings.catch_warnings():
+            self.model.fit(fit_df, freq=series.freq_str)
 
         self.fit_df = fit_df
         return self
@@ -68,12 +75,24 @@ def predict(
 
         raise_if_not(
             self.n_lags == 0 or n <= self.n_forecasts,
-            "Auto-regression has been configured. `n` must be smaller than or equal to"
+            "Auto-regression has been enabled. `n` must be smaller than or equal to"
             "`n_forecasts` parameter in the constructor.",
         )
 
-        future_df = self.model.make_future_dataframe(df=self.fit_df, periods=n)
-        forecast_df = self.model.predict(future_df)
+        self._future_covariates_checks(future_covariates)
+
+        regressors_df = (
+            self._future_covariates_df(future_covariates)
+            if self.future_components is not None
+            else None
+        )
+
+        future_df = self.model.make_future_dataframe(
+            df=self.fit_df, regressors_df=regressors_df, periods=n
+        )
+
+        with warnings.catch_warnings():
+            forecast_df = self.model.predict(future_df)
 
         return self._convert_df_to_ts(
             forecast_df,
@@ -98,19 +117,49 @@ def _convert_ts_to_df(self, series: TimeSeries) -> pd.DataFrame:
 
         return pd.concat(dfs).copy(deep=True)
 
-    def _add_past_covariate(
+    def _add_past_covariates(
+        self,
+        model: neuralprophet.NeuralProphet,
+        df: pd.DataFrame,
+        covariates: TimeSeries,
+    ):
+        df = self._add_covariate(df, covariates)
+        model.add_lagged_regressor(names=list(covariates.components))
+        return df
+
+    def _add_future_covariates(
         self,
         model: neuralprophet.NeuralProphet,
         df: pd.DataFrame,
-        past_covariates: TimeSeries,
+        covariates: TimeSeries,
+    ):
+        df = self._add_covariate(df, covariates)
+        for component in covariates.components:
+            model.add_future_regressor(name=component)
+
+        return df
+
+    def _add_covariate(
+        self,
+        df: pd.DataFrame,
+        covariates: TimeSeries,
     ) -> pd.DataFrame:
         """Convert past covariates from TimeSeries and add them to DataFrame"""
 
-        # TODO add checks if past covariate Time series has enough coverage and the same frequency
+        raise_if_not(
+            self.training_series.freq == covariates.freq,
+            "Covariate TimeSeries has to have the same frequency as the TimeSeries that model is fitted on.",
+        )
+
+        raise_if_not(
+            covariates.start_time() <= self.training_series.start_time()
+            and self.training_series.end_time() <= covariates.end_time(),
+            "Covaraite TimeSeries has to span across all TimeSeries that model is fitted on",
+        )
 
-        for component in past_covariates.components:
+        for component in covariates.components:
             covariate_df = (
-                past_covariates[component]
+                covariates[component]
                 .pd_dataframe(copy=False)
                 .reset_index(names=["ds"])
                 .filter(items=["ds", component])
@@ -118,15 +167,13 @@ def _add_past_covariate(
 
             df = df.merge(covariate_df, how="left", on="ds")
 
-            model.add_lagged_regressor(names=component)
-
         return df
 
     def _convert_df_to_ts(self, forecast: pd.DataFrame, last_train_date, components):
         groups = []
         for component in components:
             if self.n_lags == 0:
-                # output format is different when AR is not used
+                # output format is different when AR is not enabled
                 groups.append(
                     forecast[
                         (forecast["ID"] == component)
@@ -152,5 +199,32 @@ def _convert_df_to_ts(self, forecast: pd.DataFrame, last_train_date, components)
             axis=1,
         )
 
+    def _future_covariates_df(self, series: TimeSeries) -> pd.DataFrame:
+        component_dfs = []
+        for component in series.components:
+            component_dfs.append(series[component].pd_dataframe())
+
+        return pd.concat(component_dfs, axis=1).reset_index(names=["ds"])
+
+    def _future_covariates_checks(self, future_covariates: Optional[TimeSeries]):
+        raise_if_not(
+            self.future_components is None
+            or (
+                future_covariates is not None
+                and set(self.future_components) == set(future_covariates.components)
+            ),
+            f"Missing future covariate TimeSeries. Model was trained with {self.future_components} "
+            "future components",
+        )
+
+        raise_if_not(
+            self.future_components is None
+            or future_covariates.freq == self.training_series.freq,
+            "Invalid frequency in future covariate TimeSeries",
+        )
+
+    def uses_future_covariates(self):
+        return True
+
     def __str__(self):
         return "Neural Prophet"
diff --git a/neural_examples/examples.ipynb b/neural_examples/examples.ipynb
index 091a5d0eae..a3a67b78d9 100644
--- a/neural_examples/examples.ipynb
+++ b/neural_examples/examples.ipynb
@@ -39,7 +39,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -48,30 +48,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "<matplotlib.legend.Legend at 0x7fbcda1f37f0>"
-      ]
-     },
-     "execution_count": 3,
-     "metadata": {},
-     "output_type": "execute_result"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "series_air.plot(label=\"air\")\n",
     "plt.legend()"
@@ -79,7 +58,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -95,73 +74,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 514\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "9dec31c451db4ddea6feb12df9a7dd36",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/107 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 3.30E-02, min: 1.07E-01\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "cf05800d08644e4b83b9865a345b29e2",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/107 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 3.30E-02, min: 8.78E-02\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 4.78E-02\n",
-      "Epoch[514/514]: 100%|██████████| 514/514 [00:03<00:00, 161.81it/s]\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 88.889% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 88.889% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "model = NeuralProphetDarts()\n",
     "model.fit(train_air)\n",
@@ -177,106 +92,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 1000\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "3bc3d5351b6240d29c4d2b5b41249e08",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/103 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.97E-02, min: 4.44E-02\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "640c57be3d854c4da0ece842b0b53b78",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/103 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.97E-02, min: 4.44E-02\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 3.38E-02\n",
-      "Epoch[1000/1000]: 100%|██████████| 1000/1000 [00:03<00:00, 266.20it/s]\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.741% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.278% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.278% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "model = NeuralProphetDarts(n_lags=36, n_forecasts=36, n_changepoints=20)\n",
     "model.fit(train_air)\n",
@@ -285,20 +103,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "train_air.plot(label=\"train\")\n",
     "test_air.plot(label=\"test\")\n",
@@ -315,20 +122,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjIAAAGvCAYAAABB3D9ZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOydd3gU5fbHv7O7yaZseiekEAJJCL03CR0EERREEFSK/eq1Xmw/VBAVLhd7uRaaiqigF0URlY5UgRBaAqT33jdl2/z+mJ13Zza7yfYC83mePNkys/vOTjvvOd9zDkXTNA0BAQEBAQEBATdE5OwBCAgICAgICAhYimDICAgICAgICLgtgiEjICAgICAg4LYIhoyAgICAgICA2yIYMgICAgICAgJui2DICAgICAgICLgtgiEjICAgICAg4LYIhoyAgICAgICA2yIYMhai0WiQl5cHjUbj7KE4jZv9N7jZtx8QfgNh+2/u7QeE38AVtl8wZAQEBAQEBATcFsGQERAQEBAQEHBbBENGQEBAQEBAwG0RDBkBAQEBAQEBt0UwZAQEBAQEBATcFsGQERAQEBAQEHBbBENGQEBAQEBAwG0RDBkBAQEBAQEBt0UwZAQEBAQEBATcFsGQERAQEBAQEHBbBENGQEBAQEBAwG0RDBkBAQEBAQEBt0UwZAQEBAQEBATcFsGQERAQEBAQEHBbBEPGDVmyZAkoisLatWt5r+/atQsURZHnn3/+OQYMGACZTIbAwEAMGjQIb731FgBg7969oCgK5eXlUCgU+Pbbb3H+/HlERUUhPj6e97n5+fmgKAr79++3+7YJCAgICAiYg8TZA3BHampqsHfvXhw9ehQzZszA7bff7vAxeHl5Yd26dXj44YcRFBTU4f1Nmzbhqaeewvvvv4+0tDS0t7fjwoULuHTpEgBg7NixkEgkOHToEMrKyvDMM8/Ay8sLnp6eaGlpQX5+PjFoDh48CKlUijFjxjhyEwUEBAQEBLpEMGQs4OrVq1i8eDEAQKPROMWQmTx5MrKzs/HWW2/h3//+d4f3f/75Z8yfPx/Lly8nr6WmppLHMpkMw4YNw6FDh1BQUAAAaGtrQ//+/RESEoJDhw5hyZIlAIBDhw5h5MiR8PLysu9GCQgICAgImIlgyOgxdOhQlJeXd7oMTdPk8ZYtW7Bnzx6rvzcyMhJnzpwxeXmxWIw333wT99xzD/75z3+ie/fuHT7v8OHDKCgoQFxcnMHPmDBhAnbu3Inm5mbyWvfu3TF8+HAcPHiQZ8gsW7bM/I0SEBAQEBCwM4Iho0d5eTlKSkpMXl6pVJq1vC254447MHDgQLz66qvYuHEj771XX30Vd955J+Lj49G7d2+MGjUKM2bMwLx58yASMdKoCRMm4M033+St5+vri3HjxuGTTz4BAOTm5qKwsBATJkxwzEYJCAgICAiYgWDI6BEZGWnScjU1NWhrawMAREREQCKx7qc09Xv1WbduHSZOnIjnnnuO93pUVBROnDiBS5cu4ciRIzh+/Djuv/9+fPHFF9i7dy9EIhFGjx4NDw8PKJVKsl5LSwuGDh2Kqqoq5OXl4dChQ/D29sbIkSOt2j4BAQEBAQF7IBgyepga3nnttdewatUqAMCHH36IefPm2XNYRhk3bhymTZuGF198kYSCuPTt2xd9+/bFY489hkceeQS33HILDh8+jAkTJsDHxwcxMTHIzc0ly+fl5cHDwwOjR4/GwYMHcfDgQYwZMwaenp4O3CoBAQEBAQHTENKvLaR///7kcUZGhhNHAqxduxa7d+/GiRMnOl2uT58+AAC5XE5eCwgI4C2Tl5cHgDGQDh06RIweAQEBAQEBV8RsQ+aNN97AtGnTkJaWhrvvvhtHjhwh723ZsgWTJ0/GxIkT8d577/FEsZcvX8aCBQswZswYPPTQQygrKyPvtbW1YeXKlRg3bhxmzpyJvXv38r5z9+7dmDFjBtLS0rBq1SpeKMRZDBgwgDx2tiHTr18/LFq0CO+//z557dFHH8Xrr7+OY8eOoaCgACdPnsR9992HsLAwjBo1iizX2trK+6y6ujrU19cjLS0Nu3btQlFRkWDICAgICAi4LGYbMosWLcLu3btx+PBhvPLKK1i5ciXq6+vx119/YceOHdiyZQu+//57HD9+HD/99BMAQKFQYMWKFViwYAEOHDiAAQMGYOXKleQzP/30U9TX12PPnj1Yu3Yt1q1bh/z8fABAdnY23n77baxfvx6//vorKioq8MUXX9hm660gLi4OMpkMgPMNGQBYvXo1NBoNeT558mScPHkSd911F3r37o25c+fCy8sL+/fvR0hICAAmdZxNveaSl5eHESNGQKlUkjRtAQEBAQEBV8RsQyY+Pp7oJSiKgkqlQlVVFfbs2YM77rgD3bt3R2hoKBYvXkzSks+ePQsPDw/MmTMHUqkUy5cvR2ZmJsn22bNnD5YvXw6ZTIZ+/fohLS0Nv//+OwCmAu3EiRORmpoKmUyGZcuW4ddff7XV9luMSCRCUlISAKCwsBD19fUO++4tW7Zg165dvNfi4+PR3t5OvGBz587Fr7/+itLSUrS3t6OkpAQ7d+5Ev379yDq5ubkdPDLs61KpFK2trWhqarJayCwgICAgIGAvLLpDsZqM9vZ2jBkzBomJicjLy8O0adPIMomJicjJyQHA3Bh79epF3vPy8kL37t2Rm5sLPz8/1NTUIDExkbfuhQsXyLrDhw/nvVdeXo6Wlhb4+Ph0GJtCoYBCoeBvpERic7GqRqNBSkoKzp49CwA4f/48xo0bZ9PvsDdcT1Lv3r1x7do1AEBOTg7Pu2MMdhlTlr0Rudm3HxB+A2H7b+7tB4TfwN7bz5YL6QyLDJkXXngB//rXv3D27Fnk5OSAoii0tLTA19eXLOPr60tm+62trbz32PdbWlrQ0tJCnpuyLhvOMWbIbN68GZ9//jnvtbvuugvz58+3ZFM7JSUlhTw+dOiQ0cJzrsrRo0fJ4/HjxxND5sKFCwZDTsYoKiqy+djciZt9+wHhNxC2/+befkD4Dey1/T169OhyGYtjBmKxGMOHD8f27dsRExMDHx8fXjaMXC6Ht7c3AMDb25v3Hvu+j48PMUbkcjkxUjpbl61Ca8iIAYClS5di0aJF/I20k0cmOTmZPC8uLnY7Q6awsJA8Xrx4MT777DMAQFVVlUnbotFoUFRUhJiYGJOs5huNm337AeE3ELb/5t5+QPgNXGH7rRY/qNVqFBcXo0ePHsjOzkZaWhoAJjzRs2dPAEBCQgJ27txJ1mlra0NxcTESEhLg7++PkJAQZGdnY+DAgQbXzc7OJuvm5OQgMjLSqCHj6enpsJonSUlJoCgKNE3jwoULbncQsw0kvb29MXr0aMhkMjQ3NyMvL8+sbRGJRG637bbkZt9+QPgNhO2/ubcfEH4DZ26/Wd/a3NyMvXv3oqWlBSqVCvv27cOZM2cwaNAgzJgxAz/++COKi4tRU1ODbdu2YcaMGQCAIUOGoL29HT/99BMUCgU2bdqElJQUREdHAwBmzJiBTZs2QS6X49KlSzh8+DDR20yfPh0HDhxAZmYmmpubsWnTJsycOdPGP4Nl+Pj4EG3PpUuXoFarnTwi05HL5cRATE1NhVgsJi68goICt9oWfTIzM7Fx40Y0NjY6eygCAgJOoLa2FkOGDMGQIUMcmogh4BzMNp/+97//YcaMGZg0aRK2bNmCNWvWICkpCWPHjsW8efNw//33Y968eRg5ciRmz54NgPGSrF+/Htu3b8eECROQnp6O119/nXzmww8/DH9/f0yfPh3PP/88VqxYgfj4eACMuPfpp5/GM888gxkzZiAsLIzX0dnZsFlAra2tuH79upNHYzpXrlwhGU7sNiQkJABgBNOlpaVOG5s1tLa2YsKECXjggQfw4osvOns4AgICTuDrr7/GuXPncO7cOV40QODGxKzQkkwmw6effmr0/aVLl2Lp0qUG30tNTcW3335r8D0vLy+sWbPG6OfOmjULs2bNMmeoDmPAgAH48ccfATAiWa5uxpVhs8IAXZVirqgqLy8PMTExDh+XtRw9ehQVFRUAwCvWKCAgcPPAvb6Zk7gg4J7cvAE9G+FKrQrMwZAhw3pkAPD6L7kTf/zxB3l8/fr1mzYlUkDgZubixYvkcXFxsRNHIuAIBEPGSlypVYE5cE90/dAS4L6GDFtIEQDa29tv+pRIAYGbDY1Gg8uXL5PngiFz4yMYMlYSGxtLGi86wpCZNWsWpk+fbvC9o0ePgqIoXLhwARRF4fz587z3f/jhB4wfPx4BAQE4ePAgACZcKBaLAXQMLbkbpaWlJBOLxZ10SwICAtaTl5fHK9khGDI3PoIhYyUURZHQTHFxMWpra+36fcuXL8eff/5p8OTcvHkzhg4dCn9//w7vvfzyy7j77rsxbNgwfPXVV+R1Hx8f8pwVWAPu6ZH5888/O7zGFvkTEBC4OeB6mwGmUBu3gbHAjYdgyNgAbniJqz2xB7fddhvCwsKwZcsW3uvNzc3YsWOHwYyu06dP480338SGDRuwfv16SKVS8t69996L+++/HwBj1ERGRgJwT48MN6zEIhgyAgI3F/qGjFwuF0ox3OAIhowNcKRORiKR4L777sOWLVt4s4wdO3ZArVZj4cKFHdbZtm0bZDIZHnvsMQAdhb6BgYHkOauTKSsrI+0j3AGNRkM8MtyCiIIhIyBwc6FvyABCeOlGR2hrrMfQBzUoNyU6RANqdTTEYkChXAwMnwoAeGmXD9YfMT9TJjIYOPO5aXblsmXLsH79ehw+fBjjx48HwISV5s6di4CAANTV1fGWv379OhISEuDh4QHAsNCXJSEhAcePHwcA5Ofno0+fPmZvizM4f/48qqurAQC33nor9u/fj+bmZsGQERC4yTBmyKSmpjphNAKOQDBk9CivBUqqTF2a/fk8AWl3AECLGmgxeX3LSE5OxujRo7Fp0yaMHz8e2dnZOHr0KFavXm1wef34MOuREYvFvMaXQEfBr7sYMtyw0rRp01BUVIRz584hLy8PCoXCYW0rBATcFZqmUV1djbCwMGcPxWLa2toMCvwFj8yNjRBa0iMyGIgOM+EvFIgMUiE6lHkuUZcD7cWAogTRobRpn8H5iww2b5zLly/HDz/8gKamJmzevBk9e/Ykfa706d27N3Jzc6FUKqFUKpGZmUle9/Ly4i3rrinY3PoxU6dORe/evQEwISd32g4BAWfx0EMPITw8HC+88IKzh2IxmZmZpL1KaGgoeV0wZG5sBI+MHqaGdzQaDQoKShAXFweRSIS7734a33//PQDgj8uX7e7JmD9/Pp588kl88803+PLLL/Hoo4+CoiiDy95zzz14//338fHHH2Py5MlQKBQAdIXw6uvriU7GHVOwm5ubcezYMQCMIdazZ09iyACMTsZdKi4LCDiD1tZWbN68GQDwxRdf4K233jJ6PXFluGGladOmYdu2bQAEQ+ZGR/DI2AhuhV97Zy4BTP2Xu+++Gy+++CLKysqwZMkSo8uOGDECK1aswLPPPsubbfn6+uKuu+7C1q1byWvu6JE5fPgwlEolAJBmo/qGjICAgHHOnz9PPBk1NTWkzYe7wa0jxa23JRgyNzaCIWMjnFHhd/ny5airq8O0adPQrVu3Tpddt24dvvnmG97YDhw4gP79+5P0awDo1q0b0ZO4iyHD1cdMncqIrm8UQ6ayshJ33nknnnnmGaEWhouhVCqxevVqbNq0ye33zd9//817bkgw6w5wxz1+/HhSasLdK3y3tbXh0KFDvEJ/AhxoAYtQq9V0bm4urVaraZqm6cLCQhoADYC+9dZbnTw648ycOZOMMz8/3+AyvXr1ogHQMpmM1mg0Rj9L/zdwFklJSTQAWiwW0/X19TRN03RdXR3ZzrS0NLt8ryO2f9myZWQ79u/fb7fvsRRXOQacwYYNG8i+2bdvn7OHYxX33nsv2RYA9IYNG0xaz9X2f3R0NA2ADgoKojUaDd2zZ08aAB0QEGC373TEb7BgwQKXvbe4wjEgeGRsRPfu3YnOxJV7LrEzFn9/f8TGxhpchg0vNTc3k5RmV6WgoABXr14FAIwcOZK0iwgMDER4eDgA921TUF1dTWL8AHDo0CHnDcZGvPfeexg4cCB+++03Zw/FarhFKU+fPu28gdgAfY+MfqsPd6Curg4lJSUAmLISFEWhe3cmm7ShoQFNTU3OHJ7F0DRNzpfff/8dra2tTh6R6yEYMjaCoigSXiotLXVJA6C+vh6FhYUAdCe6IbiCX1cPL3GzlVh9DAsbXiotLUVzc7NDx2ULPv/8c7S3t5PnR48edeJorKepqQnPPfccMjIy8NJLLzl7OFZx8eJFXhjDHW/8LI2NjWQywOKOoSVD9bFYQwYAMXLcjdraWjQ0NADo2BBTgEEwZGyII1sVWAL3YqtfCI8LV/Dr6plL+mnXXLg6GXfzyqhUKnz88ce8106ePEkyztyRU6dOQaVSAWBuOu48s/zmm294z9355nLu3LkOGp/Lly8T8a+70JUh466CX/3JpCveW5yNYMjYEGcIfs1BvzWBMdwlc0mtVmPfvn0AgKCgIAwdOpT3fq9evchjdxP87tq1q8OFt62tDWfPnnXSiKznr7/+Io/VarXbXpA1Gk0HQyYrK4sYae4GN6wkkTAVOVpbW11+EqPPjWrI5OTk8J674r3F2QiGjA3hGgeueLCZasi4Sy2ZM2fOoL6+HgAwefJkiMVi3vvunLn0/vvvk8fz588nj905vMTW+mFxV6Ps2LFjJETL0t7e3uGG4y6cOXOGPL7tttvIY3cLL3HH27dvXwA3piHjrhMAeyIYMjYkNTUVIhHzk7riwWboRDeEu3hkDKVdc3FXQyY9PZ0YLH369MGrr75K3nNXQ0alUuHkyZO819zVkOEKsLmFFt1VJ8N6ZLy9vXH33XeT193JkKFpmvz+cXFx8Pf3B3BjGDL61+CMjAy3T/e3NYIhY0O8vb2RlJQEgIkxs0XaXAGapsmFKS4ujmT3GCIwMJBkYLmyIdOZPgYAevbsSQTN7mTIfPDBB+TxP//5T6SkpJBy68eOHYNGY35TUmdz8eLFDoJrdzRkFAoFduzYAQDw8fHBihUryHvuqJOprq4mXtdBgwZh4MCB5D13MswKCwvR2NgIgK//uxEMGX2PDDc7S4BBMGRsDKuTUSgUHTIBnElBQQFJP+xM6MvCemWKiopcyiBjaWhoIDP85ORkg6nk3t7e5PVr1665xSymqqqK6C8CAwOxePFiUBSFsWPHAmAuYu54w9QPKwHMjb+trc0Jo7GcvXv3ora2FgAwZ84cjBgxgrznTjd+Fq4xOWzYMCQmJpIicu7kkTGkjwGA8PBweHh4ALhxDBnANaULzkQwZGyMq+pkTNXHsLCGjFqtdsmqmAcOHCBZFYa8MSxseKm+vt4lU+L14aZcL1++HL6+vgCAW265hSzjjuElriGTmpoKgAk3uWIItjO4YaVFixYhMTGRVMJ2R0OGK/QdNmwYJBIJUlJSADCZfu5iaBoLm4tEIkRHRwNwT0Omra3NoPfF3c4beyMYMjbGVVOwzTVkXF3w21VYicWddDJKpZKkXFMUhX/84x/kPXc3ZNiMJR8fHyxbtoy8fu7cOWcNyWwaGxvx888/A2A6K0+ZMgUSiYQY/devX+fV/XEHuIYMm/XHejTUajWysrKcMi5z6ay0BBteqqmpcbuU//z8fOJJdtVJsisgGDI2xlVTsI25Xo3h6oJf1pDx8PBAWlqa0eXcqZbM//73PzL7uv3223nG5KBBg4h35ujRo24RJmMpLCwks+ERI0bwwjHupJP53//+RzwU8+fPJyEL9hhTqVQubyzrw2Ys+fv7k3IFXI+Gu4SX2HFKJBKiU2Rx56J43LDSjBkziPfPlSbJroBgyNiYbt26ITg4GID9DZmioiIsW7aMNHqMi4vDk08+iZqaGrLM+PHj8dRTT5ED39PTk3dzpyiK/Pn6+qJXr15YsmQJryaGqxkyOTk5ZExjx46FTCYzuqw7eWS4Kdf//Oc/ee9JJBKMGjUKAHMxLigocOjYrIEbVhozZgwGDhxIsvvcyZDRDyuxcI8xdwovlZaWorS0FADjjWH3CXei4w7bo1QqiecoOTmZ3OxZ3Fnwy732JiUloU+fPgCAq1evup13yZ4IhoyN4bYqKC8vR2VlpV2+Jzc3F0OHDsX169exfft2ZGdn47///S/279+PUaNGEUEiwJ8p9unThxS9Ytm8eTPKyspw+fJlfPTRR2hubsZTTz1F3ne10FJXaddc3MWQOXv2LLnhp6amYsKECR2WcdfwEteQGTt2LHx9fUna8qVLl9wiHFNeXo79+/cDYMKurFEJgOcBcCchNrd+DLeYJNeQcQePzNWrV0lCgiFvszsbMlyPTEJCArm3aDQaXLlyxVnDcjkEQ8YOOEIn849//AOenp74448/kJaWhtjYWNx6663Yt28fSkpK8PLLL5Nla2trScquIX1MYGAgIiMjER8fj6lTp2Lnzp1YuHAhed/VPDKm6mMAJtWcDQG4siGjn3JtqA8Wm7kEuKchQ1EURo4cCQAYMmQIAGY27Q43y2+//ZacQ/fccw9v/3ArSLuDB4NFX+jLEh0dTcozuMO+6SpsfqMYMj179hR0MkaQdL3IzcVfE09AUdn1DJEGI4bLFRdA/5YzsWUKBgUyF2r5A204IDvU5ed5hksx9sCoLpcDGMPk999/xxtvvAFvb2/ee5GRkVi0aBG+++47IhzlZuuYIvQFgGeffRZff/01ANfyyCiVShw4cAAAEBYWxqt7YQixWIzExERkZmbi+vXr0Gg0xIXuKlRWVmL79u0AmFYL3LAFl5EjR0IikUClUrmNIdPY2EiM+X79+pEb5JAhQ/DVV18BYLxR+u0lXA1jYSWAuVH6+PigpaXlhjBkKIpCv3798Ndff6G4uBj19fWkrpQrYo4h44oZmJ3BGjJeXl6Iiopy2WQSZ+NaV3QXQFHZjrayrv/ay9qhqlSh3cB74gYxQkVhCBWFwaPJw6TPM8V4Yrl+/TpomiZpkvqkpKSgrq4OVVVVAPiGjClCX4BfsbS6upoUm3I2J0+eJPVwpkyZYpJRws6Y29raXHJG9tlnn5FmkA888AAR9erj4+NDPBlZWVlk/7oyJ0+eJJ6MMWPGkNfZ7QBcXydz7do1EoYZNGhQh/NOJBKRlPLc3Fy0tLQ4fIzmQtM02abQ0NAOdZjcSSdzo3pkNBoNmUT26NEDIpFI8MgYQTBk9PAMl8Irqus/aZQUknAJpIbei/REtaYa1ZoqNIgbTPo8z3Cp2WPtKnOFFb1Z4pHR/2xX8cqYE1ZicWWdDDflWiQS4bHHHut0ea5OhtuE0VXRF/qyDBw4kIRnXD0FuzNvDAsrwqRpGpmZmQ4ZlzXk5+eTpIBhw4Z1CGW6U+YSOz5/f3+DhTEjIyNJHzZ3MmTKyspIllzPnj0BMF7oqKgoAIxHxp2yF+2JEFrSw9TwjkajQUFBAeLi4gx6Bfr27YvLly/Dw8MDzcXNHZT01pCYmAiKopCZmYk77rijw/uZmZkICwsj7mD2ghUaGoqIiAiTvkP/YpyXl8dzazoLWxgykydPtvm4LOWHH35AWVkZAGD27NmIj4/vdPlbbrkF//nPfwAwOhlD+9+V0Bf6sshkMiQlJSErKwsXL16EQqGw6TliK2iaJoYMRVFYsGCBweW4N/5Lly7xPE6uiKH6MVzcxSPT2NhIMvj69u1rUFsmFosRFRWF4uJitzJk9PUxLP3790dZWRlqa2tRUlLC8zjdrAgeGTvB3vS5qYG2IiQkBFOmTMHHH3/cIQWvvLwc27Ztw5IlSwAwrRJYV3f//v0NnuiGePfdd3n6G1cQ/NbW1pILcL9+/cjMpCtc2SOjL/LtCq5Xw9V1MtxGkdHR0R1my+zNXqFQuOzN8vTp0+SGMmHCBFIlVh82tAS49o2fhZuxxNXHsLiLR6azQnhc2Jt9ZWWlW2TJAfxrLteQEXQyHREMGTth71jmhx9+iPb2dkybNg1HjhxBUVER9u7diylTpqB379545ZVXAAByuZysExkZifPnz+P8+fOoqKggr9fX16O8vBwFBQX4888/MW/ePHzzzTe8hniuYMjs27ePuFJN9cYArmvInDlzBsePHwfAXIQ7K+zHEhISQm6a6enpHRoxuhIXLlwgx9+YMWM6GNHuoJMxJawE8A0Zd0jB7sojExQURIy2ixcvumwIw1hrAn24Xgu2do6ro596zSLoZDoiGDJ2wt5Wc69evfD3338jISEB8+fPR1xcHG699Vb07t0bx44dI0XiuIbMN998g0GDBmHQoEH4/PPPyetLly5FVFQUkpOT8eijj0Imk+H06dN45JFHyDKuoJHhhpWmTZtm8nqRkZHk93AlQ4brjXniiSdM9paxOhm1Wo0TJ07YZWy2wJg+hsXVDRmVSoXvvvsOACCVSjF37lyjy3JTll3dI6PRaMjvHR0dbdSzyXo46uvrXfbmb2rFcncU/BoLLQkemY4IhoydcESrgvj4eGzZsgXl5eXQaDR45ZVX8Mcff/AObq449O+//wZN06BpGv/3f/8HAOQ5TdNobW1FdnY2tmzZgsGDByMiIoKEl5ztkaFpmhgyXl5ePL1FV1AURbwy+fn5JEPImVRUVODbb78F0HnKtSHcpTAeV4xsyJAZNGgQMd5c0ZDZt28fKWh52223EUPFEBRFEY9AUVERGhoaHDJGS7h27RrJ/DMUVmJxh/CSuaElwH0MGfaaS1EUTzuXlJRE9GSCR4ZBMGTsRGRkJEJDQwE47mBbtWoV3n//fV7KK3sBoiiKZFaYCkVRpN9Pfn4++UxnkJWVRWpAjBs3rkP9nK5gDRm1Wu0S3iVuyvWDDz4IHx8fk9d1B0OGpmnikfH19TUoFPfz8yP75cKFCy5hYHIxNazEwg0vuXLVVWP1Y/Rx9Qq/NE2TcXFbwxjCHQ0Z1iMTHR0NLy8v8rqHhwevVYG7dCi3J4IhYye4rQoqKytRXl7ukO9dunQpnnrqKYhEIqjVahKv79Wrl1k3SxY2NtvW1uawbTCEJdlKXFxJJ6NQKPDJJ58AMC3lWp+YmBjExcUBYOq0uJoBADCNItkGfWwhP0NwBb+udPOXy+X43//+B4CpfD1jxowu19HPXHJVutLHsLh65hKbuQN0XR8rJiaGPHYHQ6axsZGUzeCGlVhYnYxGo3ELTZa9EQwZO+LsWGZ2djax1k0thKcPtwOzM8NLlupjWFzJkOGmXM+ZM4cYJebAemXa2tpcsg5LV/oYlsGDB5PHrhRe+vnnn4m+bN68eZBKu67z5C6GjLEeS/okJyeT0hKu6JExVR8DuJ9HxljGEosjpAvuhGDI2BFnH2xc48nUQnj6cNXyzgrJtLe349ChQwCAqKgongvfVFzJkOmsy7WpuHp4yVRDxlUFv+aGlQC+IeOqs2SlUon09HQAzA2ys3CMt7c3qYp95coVqNVqh4zRVMwxZKKioogeyx0MGWMZSyzc67kg+BUMGbvi7DQ5WxgyruCROXbsGKmFM3XqVJOze7hwG/s505A5ffo0qa3Sv39/jBs3zqLPcfUGkqwhIxKJSKNIQwwaNIg8dhVDprq6mnRY7969u8n7KCwsDGFhYQBc1yNz5coV4qU1pb8VayC0t7cjOzvbrmMzF3MMGQ8PD0RGRgJwP0NG8Mh0jWDI2JGUlBSiDXCG1WzOiW4MV/DIWBtWAhidA3uTcaYhY0qXa1NISUlBSEgIACY7yJlCbH0aGhp4jSL9/f2NLhsQEECMzIyMDCiVSoeMsTN27NgBlUoFAFi4cKFZTUZZr0xFRYVL9sIyVejL4sqZS+x4xGKx0b5zXNjwUllZmUscZ53RVWgpLCyMGGZCqwLBkLErUqmUnGCZmZkOryjJ3kx8fX15nhVzcAWPDNeQsaa9ABteKi0tdUohufLyclKXJDg4GPfcc4/Fn0VRFPHK1NXVuZRQ9uTJk+TCakqaPBteam9vd4ntsCSsxOLq4SVThb4srir4ValU5Fjp1asXL6vHGKwhQ9O0UxMXTKGr0BKg88qwrQpuZgRDxs6wB5tKpXJoM7mmpibiQenbt69Zs0ouMpmMeDKcYchUVFSQmP7gwYPJWCyBq5O5fv261WMzly1btpCZ4EMPPWR2Crk+rqqTMVUfw+JKOpn8/Hwy/tTUVLNDsq5e4ZcV+lIUxRNaG8NVU7Czs7PJxNBUbzNX8MuWcnBVWEMmICDAqI5J0MnoEAwZO+MsnQx39mSpPoaFnRGUlpY6vGbBvn37yGNLw0oszhT80jSNr776ijx/8MEHrf5MwZCxPd988w15vGjRIrNDf66cudTW1kZueCkpKfDz8+tynYSEBGJwu5IhY2prAi7ukrmkVCpRWFgIgAkrGTsGBZ2MDsGQsTP2SsGuqqrCo48+itjYWEilUkRGRmLatGnkJsL9ri+++AIzZ860+LvY8BJN06TTrKOwtn4MF2caMufOnSOu8LFjxxp1F5vDoEGDSG2go0ePukScXKVS4dSpUwCYG4d+o0hDcAW/zkwl53a6BmBR6M+Vm0dmZGQQ7Y8pYSWA0Z+wxdeys7M7NKl1Fpbo/9zFkCksLCQZYp1dJwSPjA7BkLEz9rKa586di/T0dGzduhXXrl3Dzz//jPHjx6OmpgYA/0S/8847ceTIEYv7pThL8KtSqbB3714AjM5n1KhRVn2eM0NLXG/MfffdZ5PP9PDwIL9JcXGxw41MQ2RkZPAaRZpCYGAgETRyb7aOJiMjg2dsWlLfJzAwkDRbvHz5sksYlyxddbw2BuvxoGnaJTRMwI1tyHSVscSSnJwMDw8PAIJHxixDRqFQYNWqVZg5cybS0tKwZMkSYgnu3r0bI0aMwC233EL+uIKqy5cvY8GCBRgzZgweeughUhAMYFyeK1euxLhx4zBz5kxy82LZvXs3ZsyYgbS0NKxatcrlFedcIiIiEBERAYA52GxxYauvr8fRo0exbt06TJgwAXFxcRg+fDhefPFF3H777QD4Fvq//vUvzJw5E1u2bLHo+7iGjCN1MgcOHCC9bqZOnWpSUbLO4LppHemRUSqVJGQhlUpx11132eyzXS281FV/JWOw4aXW1laHasm4WCPy5cLe+Ovq6njXOWdjbsYSiyvqZNhx+Pj4mOzdvNEMGaFVgQ6zDBm1Wo1u3bph48aNOHjwIBYuXIinn36a1PgYMmQIjh49Sv7Y9DCFQoEVK1ZgwYIFOHDgAAYMGICVK1eSz/30009RX1+PPXv2YO3atVi3bh3y8/MBMO7Mt99+G+vXr8evv/6KiooKfPHFFzbafMfAugCrq6ttcmGTyWSQyWTYtWuXwUwobg8SDw8PjBgxAosXL8amTZssMqSclbn09ddfk8eLFy+2+vO8vb1JqOPq1asOmy3/8ccfJBX39ttvR2BgoM0+29UMGa4+xpzGns7WyajVamzfvh0AIJFIrDI2XVUnwxoyEonEYO8rY7ha5pJcLifXodTUVJMTGbp160Yeu7Ih01XqNRd2P97srQoMN0Axgre3N0+kOG3aNLzzzjtdurTPnj0LDw8PzJkzBwCwfPlyTJo0CSUlJYiOjsaePXuwbt06yGQy9OvXD2lpafj999/x8MMPY+/evZg4cSKJPS9btgyvvfYaHn30UYPfpVAoOvSekUgkpFtoV0zafxqVbSb0rqEBtVoF8cUioAs9YMN9TyDgjiUAgFuOXYKXV8fCUuFentg/abhJYxSJRNi0aRMefvhh/Pe//8XgwYMxbtw43H333ejfvz+KiopQX18PgDkRNBoNpk6dioaGBhw8eBDjx4836XtYuJ1Xc3NzSc0S/f+2RC6X48cffwTAuOtvvfVWm3xPr169UFBQgPr6elRVVZHGnpZg6vZv3bqVPF60aJFNf6/hw4dDIpFApVLh6NGjDq8nw/0NuI0iZTIZUlNTTR4PVydz5swZm4XfTOXdd98lKazTp09HUFCQSWM3dAxwm7NeunTJqpIBtqK5uZl4uvr16wdPT0+T9w13ey5cuMBbz57XAGNcunSJTEL69u1r8nd7enoiLCwMVVVVKC4uttmYbf0bcAsPxsfHd/q5XCPz/PnzvPPIUdj7GDDFUDXLkNGnsLAQjY2NiImJQXZ2Ni5evIhJkyYhODgYd999N+bNmweAuflxK6t6eXmhe/fuyM3NhZ+fH2pqapCYmEjeT0xMJKGR3NxcDB8+nPdeeXk5WlpaDDZB3Lx5Mz7//HPea3fddRfmz59v0jaVNbegUmlGKW5TlvX0giiEqXNQRwNo7ehFUatUZmkchg4diuPHj+P06dM4f/48/vjjD6xfvx5vvfUWL11v2LBh5HNvvfVWfPDBB2bXlFGr1RCLxVCr1bh69WqHcdojlXHXrl1Ea3HrrbfarO5DVFQUeXzkyBGeJ8BSOtv+xsZG/PTTTwCY2jEpKSk217L07dsX58+fR1ZWFs6dO0cK5TmSoqIiFBcXEx3WgAEDzKptwU2rP3HihEP1Phs3bsQbb7xBnt9+++1mfz/3GOCefydPnnQJ7dLp06fJzT85OdmsMdE0jcDAQNTX1yMjI8Pguo5MZ2bblQBMZ2hztiU8PBxVVVUoLS1Fbm4uxGKxzcZlq98gKysLAONN12g0nW4fK1sAmLDuxIkTbTIGS7DXMWDK/cpiQ4bVtSxZsgQymQyDBw/Gd999h8jISFy5cgXPPfccgoKCMGnSJLS2tsLX15e3vq+vL1paWkhYivu+r68vUcfrryuTyQDAqCGzdOnSDvFtczwyUdkVEJvjkRFLuvTIKJVKVFZUAAC8fXwM1gUI9/K0SFzYu3dvEnZ58MEH8eGHH+KRRx4h72/bto3oM2iahlQqRWBgIAICAsz6nri4OOTm5qK4uBixsbGgKAoajQZFRUWIiYmxuE6NMdgS8QDw8MMPW/TbGGLIkCFEeNvY2GjV55qy/V988QXxEN5zzz08g91WTJo0CefPnwfATC5MqQ9iK7i/AVcfM2nSJLN/24SEBOTm5iIzMxPR0dFGO2bbknfeeYdnxKxcuRLLli0zOe3a0DHA9fIVFBTY7Ni1hh9++IE8TktLM3tMAwYMwOHDh1FZWQmZTEaMZUuuAdUHa9B4oRGxy2Ig8TN/H3PD8+PGjTNrW3r06IHLly9DrVbDy8uLF26yFFteB2maJmGv+Pj4LvU/3FpUzjrW7HkfMBWLrhQqlQovvPACYmJiSKiJVeoDzAxxwYIFOHjwICZNmgRvb28yu2aRy+Xw8fEhxohcLidGilwuJztIf122IqshIwZg3IemGi2GODjFeF8YLqylHBcX1+XOUygUkMlkUCqViO7Tx26xzNTUVPz00088BfuOHTt42Tpz5szBd999xzN2TIG9yTQ2NqKhoYFnjIlEIpsewBUVFfjzzz8BMAbULbfcYrPPT0pKIo+zs7Nt8rmdbT9XQHr//ffb5US/5ZZbsGHDBgCMRuXOO++0+Xd0hUgkwokTJ8jzsWPHmr2tQ4YMQW5uLlpbW3H9+nWLmoOaw3/+8x/861//Is9fffVVvPbaaxZ9FvcY8PPzI+cLm+XjrAs8C1d3NGLECLPH07dvXxw+fBgA068pLS2N976p14D2ynacW5wOjYKGqkmN5Fd6d7mOPlydzoABA8zaFm45gNLSUp4A2FpscR2srKwk97iEhIQuPy8yMhKRkZEoLy9HRkYGKIqyuO2Jtdj6PmDWd5u7gkajwcqVK0FRFF577TWjPxpFUcSVmZCQwIv7tbW1obi4GAkJCfD390dISAjv/ZycHCJy0l83JycHkZGRRg0ZV8TT09Om6vKamhpMnDgRX3/9NS5cuIC8vDzs2LED//73vzF79mxyQxGLxbjtttvQt29f8jd37lxs3LjR7O90pOD322+/JfHWxYsX2/TkcGTzyLy8PBw5cgQA4863RRjLEK7SQJL1yHTVKNIYjhT8rl+/nmfEvPbaaxYbMYZgjbDm5mZS3MyZsEJfLy8viwxEW2UuNV5ugkbB3Bdqj9Va9Bns94eHhyM8PNysdV09c8nUjCUubDJJbW2txSU23B2z7xBvvvkmampqsHbtWp7r9/jx46irqwPAxPi+++470jV2yJAhaG9vx08//QSFQoFNmzYhJSWFeHFmzJiBTZs2QS6X49KlSzh8+DCp4jp9+nQcOHAAmZmZaG5uxqZNm6wq7uYsWHW5Wq22uhaDTCbDiBEj8M4772DcuHHo27cvVq5ciQcffBAbNmwgsco+ffp08E7NnTsXZ86cMbuAkiNryXBrrliTBmuIuLg4UnvB3oYMN+vqvvvus9tMKSQkhBjK586dc0ofqfr6ejJTHjBggElVY/XhhsTsacisW7cOK1asIM9Xr16NV1991abf4UqZS3V1deQGOXDgQHL8m4OtMpdaclvI48bLTdCozBOIVlZWkpIMljTCdXVDxpyMJRahwq+ZhkxZWRl27dqFy5cvY/LkyaReTHp6Ok6dOoX58+dj7NixeOmll3DfffcRY8TT0xPr16/H9u3bMWHCBKSnp+P1118nn/vwww/D398f06dPx/PPP48VK1aQTJnExEQ8/fTTeOaZZzBjxgyEhYVh+fLltvsFHIQtWxVIpVK89dZbOHv2LOrr6yGXy5GVlYXXX38dBQUFxBNm6EQfPnw4aJo2u22BozwymZmZ5CY2ZMgQk7ramoNEIiEXiOvXr9tNaa/fksDWBpk+bBq2Wq3GyZMn7fpdhuA2ijSnfgwXRxgyb731Fl544QXyfM2aNbxSELbClQwZSwvhceF6cazxyLTk6wwZTasG8uyWTpbuiCWtCbi4uiFjSrNIfexVPd6dMEsjExUVxTspuAwaNAhPP/200XVTU1Px7bffGnzPy8sLa9asMbrurFmzMGvWLHOG6nIMHDiQPP7tt9+wdOlSu3wP90C2tscSF0cVxeNqSu699167fEfv3r2RlZWFtrY2lJSUICYmxubfcerUKVI9eMKECSaV6reGW265BZ9++ikAJrzk6JTf48ePk8eWGjIhISGIj49Hfn4+0tPTSbacrXjjjTfwf//3f+T5m2++iRdffNFmn8/FlZpHmtvx2hABAQGIjY1FYWEhSX+2xMMoz+UbLo0ZjfBLlpm8viUVfbm4kyFjbmgJEDwyAnZm7NixJJ77448/2i1VjWss2rKmgCNCSxqNhhgyYrEYCxYssMv3OKLnEtcbYy+DjIuzC+PZwpABdDqZlpYWXL161epxsbz++us8I2bt2rV2M2IARlTOGmE3gkcG0BkOjY2NFl+/WvL4hkzDxUaz1rfWkOEmpbiiIcOdJJrqkeG2KrhZPTKCIeMgpFIpyRRSq9X46KOPbP4dmZmZ2LNnDwBGC2LLmgLBwcFE92Avj8yxY8dIRecpU6bwaiTYEnsbMgqFghiU3t7emDt3rs2/Q5/Y2Fji9Tl58mSHopD2RKlUkkaRMTExVnm47CH4XbVqFV555RXy/N///jeef/55m3y2Mby8vIiwPDMzkzQBdAasR0Ymk/Gy9syFG8qxJLxEa2i0FPCbTjZesMyQoSjKItGyD6f8hSNr35gK65GJiIjoULLEGEKrAsGQcSiPPvoosZw/++wzUkPHVrzzzjvk8ZNPPmnTOhwURZEZQkFBgV0uzLZuSWAMexsye/bsQW0tk5ExZ84c+Pv72/w7DMF6ZVpbWx3aRfrKlSvkWDanLYEhbG3I6Gcj6adc2xP2xt/e3s4LGTiS8vJy4nkYMmSIVRmA1gp+28raoWnja9IaLzSB1pjWKoRbhj8hIcHkG70+bHippKTE4ZWwO6OlpYXUyDE1rMRiy2QSd0QwZBxIZGQkCZfU1dXxbtzWUlVVhS+//BIAU8fCHoJoVvCrUqls7pZtb2/H999/D4ApiMi2s7AH9jZk2P0A2K7TtSk4K7zENTisCSsBfEPGGmOMpmm8+uqrWLVqFXnt7bffxrPPPmvV+MyB6zFwVnjJVmElwHqPDFfoy6JqUnXw0hgjLy+PGMyWhJVYWENGqVSSHmiugCUZSyw3u05GMGQczJNPPkkev//++zZrXPjJJ5+QBpIPPvigXbwA9hT87tmzh/SHuvPOOy2ebZlCZGQkKb5oa0OmtrYWv/zyCwDGPexI0e2NYMiEhISQ6qTp6ekWz5hfffVVrF69mjx/5513Ok1GsAeukLlkacdrQyQnJxPdj0WGDEcf4xmmKwthanjJWn0Mi6sKfi3Rx7Dc7JlLgiHjYIYMGUIu9pcvX8b+/fut/sy2tjaiuRGLxfjnP/9p9Wcawp6CX6441p5hJYAJk7Femby8PJvqSb777jsolUoATMq1I8rss6SkpJDS8ceOHXOI25ymaWLI+Pn5WXWDYWHTsJubmy0yNLdv384r7/Dee+/hqaeesnpc5sI1ZJyVuWSLjCUWqVRKNDZZWVnkODcVriETOVOnf2vIEAwZwLKMJRbBIyPgcLhemXfffdfqz/vmm29Ikah58+bZrd+GvWrJ1NbW4tdffwXAeEsc0fiMNWTUarVNjTJHZytxoSiKaFRqa2tJt2N7kp+fT469kSNH2iRd2hqdTHZ2Nh5++GHy/N1337WbYd8ViYmJpCClMzwyNE2T0FJISIjZzWINwRpnCoWClBcwFTnHkImaozNkGk3MXBIMGeOEh4cjMjISAOORsZWn310QDBkncMcdd5DMjl9//dXsCwIXmqbx9ttvk+fPPPOM1eMzhr08Mjt37uQ1VnSEF8MeOpnr16+T9hD9+vXjuXsdhaPbFRw7dszgd1uDpYaMQqHAggUL0NTUBIDRJ3EnDY5GIpEgOTkZAHOMOTKTDGAaiLIakKFDh9qksrQ1rQpYjwwlphA0IgieIUziQ+OFJpNuvOz3SaVSq5qvuqohY01oCdB5ZWpqam66VgWCIeMEJBIJHn/8cfL8gw8+sPiz/vzzT+K2HjNmDIYPH271+IzBVlsGbOuRcVS2Ehd79FxyVEuCznC0ToZryFirj2Gx1JB54YUXyPK9e/e2S4kDc2E9GCqVyu4tMfSxZViJxdLMJZqm0ZLHiHq9untB5CmCf39Gx6eoVqCttL3T9dva2siEr0+fPlZNdlzVkGE9Mr6+vhaVnriZdTKCIeMkHnjgAdLhe/PmzWhoaDB53f/+97/w8/ODSqXieWNOnjyJ8ePH85Y9dOgQKIoiJ8mJEycgFost6lfFbXtviSFTc7QGp+48g9L/lZHX8vPzyQ23T58+vArI9sTWHhluSwKRSIR77rnH6s+0hMGDB5OGqo4wZNhCeGKxGCNGjLDJZ4aFhRGPpamC319++YWUH/D09MR3331HBN3OxJmZS7bMWGKxNHNJUaOEqkkFAPDtwRyfrCEDdC345dbisaQ1ARdXNGS4Ie6EhASLJkE3s05GMGScRHBwMEnNbW5uxubNm01ed8KECWhubsZ3332H33//HQCTIRMVFYVTp07xCiIdPHgQsbGxJOa6ceNGPPHEEzhy5IhF7kfW5VlZWQm5XG7WuldezkLN4RpceuoyaRbHbUmwePFih3kxuB4Za0J7LMeOHSMXosmTJxODz9F4eHiQztNFRUV2NWbq6+uJN3DAgAE2NRxYr0xTUxOys7M7Xba4uBhLliwhzzf8ZwN89vji3NLzaK/qfKZvb5yZuWQPj0yPHj1IRqE5hgxX6OsTzxgyAQM4hkwXOhlb6WMARpQeEBAAwHUMmZKSEiKeNlcfwyJ4ZAScAleE+P7775tcZC4pKQlRUVG8Anh9+/bF7Nmz0aNHD17TwEOHDmHChAkAQIyfRx99FDNnzsSWLVvMHjNXMGiOTkbdokZTJtOVWdWsRnNmM2ia5oVj7N1YkUtQUBDCwsIA2MYjw60d42iRrz7c1g6PP/44VCqVXb7nxIkTRNswevRom362qeEltVqNRYsWoaamBgBTgHDxLfcie0Muyn+uwPV/O6cQHYuzDBmNRkN+t6ioKF5pfmsQiUTEy5Sbm2vyZIZnyCRoPTL9dB3Su8pcsqUhA+i8MsXFxS4hjLWkWaQ+SUlJpOCq4JERcBh9+vTB1KlTATBGAVt/xBRGjRpFCoYFBASgrq4O48ePR1paGg4ePAiAqfB66tQpYsh8//33SE5ORlJSEhYvXoxNmzaZfRJbWkum8XITwIkQ1J9rwNmzZ5GVlQUASEtLs3tjRX3Y8FJJSQmam5st/py2tjZeMb877rjDJuOzlGXLlpEU5gsXLthNK8LVxzjLkFmzZg2OHDkCgGmPsHHjRsiv6W6uFXsqTa4caw/i4+NJqM+RKdjZ2dkkXG2rsBKLJWnl3GJ4PvHe2v8+kPgxWpfGC02drm8vQ6atrY1U4XYm1mQssXh6et60rQocV+TCTUj7cDwqmipNWJKGSq2GRCwGYHk4pG1kG3wT/UDLNXjvvfcwe/Zsk9ZTKBTECFmyZAk++ugjpKWlQalU4r///S8AZsbc3t5ODJmNGzcSMe306dPR0NCAw4cPd9DVdAbXkMnPzzc5M0ffdVx/tgFfpzte5Muld+/e5GacnZ1tsT5n9+7d5KYxd+5cuxbzMwWxWIyPP/6YhJheeeUVzJ8/H1FRUTb7DpqmeTWQbCX0ZWENMcC4IXP48GFS9E4sFmP79u0IDg7G9ULdTaG9vB0N5xsRODjApuMzFZFIhD59+uDMmTPIyclBS0sLMWzsiT3CSiz6gl9ThKncrte+Wo8MJaLg398Ptcfq0FbahvZqBaShngbXZw2ZoKAgm4Rt9XUybP0lZ2FNVV8u/fv3R0ZGBmlVwD2PbmQEj4weFU2VKG0sNeGvDJXySpQ2lpm4vOG/WkUtRP4iUL4iHDx40KTYZmtrK7kBi8ViDB48GL1790ZYWBjS0tKITubQoUNISEhAbGwsrl69itOnT2PhwoUAmMypu+++Gxs3bjTr97G0lkzjRf6Mq+5MPbZv3w6ASaecN2+eWeOwBbYS/OpnK7kCI0aMwAMPPACA6VZs6/5Cn3/+OQlhxsfH824MtiAiIoKEQ86dO9dB8FtdXY177rmHvL5q1SpiTLXqlbyv+M2UiYn9YD0YNE07pLYPYB+hL4slmUu80FKczpDz79e1Tqa2tpbo+fr162cTHZ2rCX5tEVoCbl6djOCR0SPCL9zEJW3jkQEY7UqdnHFvvvfee10aF19//TXq6uoAMBfJixcvIi0tDQDQrVs3xMTE4Pjx4zh48CApLrdx40aoVCrebIamaUilUnz44YdE/NYVltaS0c9KkF9rRlMNY9zMmjULgYGBJn+WrbCFIVNTU4O9e/cCAKKjo83ybtmbt956Cz/++CNqa2uxbds2PPjgg+Q4sYbMzExepdwXX3zR6s80xJAhQ1BSUoLGxkbk5OQQgTZN01iyZAm5uU2aNAkvvPACWa+lgN/Tp+K3SiS93AvOQj8Uww2b2QONRoM///yTPLe1R8YS3U9LPmNcSiOlEPvoiiYGcDOXMhoRNiG0w7r79u0jj20RVgJc15ARiURWFTS9WTOXBENGj8OPHzJpOY1Gg4KCAsTFxVnVURZgDBn2xNq2bRvWrl1LhKiGvpcr8vXy8sKhQ4d4M+5x48bht99+w+nTp/Hoo49CpVLhyy+/xIYNG4gmh2XOnDnYvn07HnnkEZPGGhUVBalUivb2dpMNGY1SQ4S+BJpCL0kvXFBdcEpYCbCNIfPLL78QMe3ixYttUtnWVoSGhuKtt94ilW7/8Y9/ID09nQgCLaG9vR0LFy5EaytzY3rkkUcwZcoUm4xXnyFDhuDnn38GwISXWEPm3XffJZWgw8LC8NVXX/F+d/0mhM2ZzWjJbyHZMo7G2hRsRa0CRV+XIHhUEIKGBXa5/Ndff020K0OHDkVoaEfjwBoiIiIQFhaGqqoqk7ZH2aiCopopBsiGlVj8OZlLDQZSsFtbW/H888+T57feequlw+bhaoYM692OjY0l1aAt4Wb1yAihJRdAJpORbtXt7e347LPPjC67d+9e4p7u3bs3Lly4gPPnz/Nm2mlpafj000+hUCgwYcIE/PLLL6irq8Py5cvRt29f3t/cuXPNCi+JRCJSGC8vL88ksXDzdTk07UwIQOSlO+SSJMkIDg622cXJXHr27Enc1JYaMj/++CN57OxsJUMsX76chBYuX75sVfFFgCk6x870+vTpg/Xr11s9RmMYEvyeOXOGd2P78ssvedofjVKDttKOIkdnhpeszVy6uuY6rq66htNzz0BR23l14JaWFrz00kvk+VtvvWX295kCu00VFRUkY8zomPI7pl6z+Cb6QOTNXBP0w88AsH79euTn5wNgPG8zZsywZtgEriFTVFRkk8+0lLq6OuJhtyasBDCtCljNUkZGhktkZDkCwZBxER5//HHi2fn444+NNmTjFsB79tln0draisTERJ7gLi0tDU1NTSRNe+PGjZg8ebLB8NHcuXNx5swZs6x39mRraWlBdXV1l8tzw0pRsyPJ496SJMyfP9+qGYg1eHt7k0ypq1evmn3SZ2ZmEhHi4MGDeTNvV4EV/rIG26uvvoqSkhKLPuu3334jvcGkUim2b99uV+Eq15A5d+4cGhsbsWDBAnJu/Otf/8L06dN567QWt5HsOFmKrq6NMw2Z6Ohocu5ZkrlU8xcTdlbL1SjfXdHpshs2bCD7d+bMmXbrvs4N8Vy9erXTZXn6mB7840UkEcG/D5OG3ZLbAmWj7rpXUFBADDGJRIL333/fZnWmXMkjY4uMJS6sV6ampgZlZWVdLH1jIBgyLkKPHj1w++23AwBKS0uxc+fODstkZGSQTJHExEQ88MADBgWEcXFxoGmapDbv3r2buOL1GT58OGia5sVWTRkriymzGe5Mq9vcSLSLmCJlvSVJTvdisOGK+vr6LmeW+nBFvs7ejs4YOnQoCS81NzfjueeeM/szKioqeEXn1q9fb9YxYwmRkZFE03X27Fk88sgj5KI/YsQIvPHGGx3W4Qp9wyeHwqcnc+OsPVHXpTfDXlAURYzcwsJCNDaa1iQRYMIyLTk6Q6D0R+M3prKyMqxbtw4AY8Da01vGNWS68mbyDRnvDu/zKvxyrhXPPvssSSF+4oknSGqxLQgICCDZhTeaIXMz6mQEQ8aF4AooDXXF5mpjnn76aau1OZbCdX+aZsjoLtytkW3IamcMr1BRKAbFD7L9AM3AUp2MWq0mVYnFYjHJBnNV3njjDZJi+u233+LAgQMmr6vRaLBkyRLS5XrmzJm8XmH2hE0fbWhoIFlu/v7+2L59u0GtD1fo6x3ng4hbteJ9DVD5R5X9B2wES2qvAEDTZX64hU1VNsTKlStJgbqHH34YKSkpFozUNLjb05VHhtv12rdHRw+eoQq/+/fvxw8//ACA0eS8+uqrVo1XH4qiXKYonrXNIvW5GXUygiHjQowbN44chKdPn+ZV6C0rK8M333wDgKmlcP/99ztljIB5HhmapsksyytKip2/70CWKou833DW9NmpPeAaMqa2Krhw4QLGjh1Ltn3q1KkWNXlzJMHBwWS2DjDCX1O7Mb/33nskMysyMhKbN292WCsJQxk+X3zxBe8Y5MIV+vrEeesMGTDF8ZyFpYZMhx5ENFD2U3mH5S5cuIBNmzYBYAy91157zaJxmgo3jGpNaAngV/htzGiCUqnEE088QV5bu3atyVmV5sAaMnK53Kxed7ZG8MhYj2DIuBAUReHJJ58kz9977z3y+KOPPiLagEceecSpRde4s4bCwsJOl20tbIWqkcnq8e/vj6+//hrXVLoLX/3ZeruM0VTM8cjI5XI8//zzGDx4MDEyRSIRnnnmGbuO0VYsXbqUFMnLysoy6PXTJz09nSeu3bp1q9GMOnugb8g8/PDDuOuuu4wu31qoM2S8Y70RNCwQntoia9UHa6BuNa0NiK2xNHPJUCZP6Q/88BJN03j22WeJV+Hll1+2+z7y8/MjxuT169c7bezJdr32CPaAR0BHL5osxQ+UhDGMGy404sMPPyTh8hEjRtitNpOr6GRsbcgkJycTb6XgkRFwCgsXLiQXoZ07d6KkpAQtLS345JNPADBNAR3l1jcGdzbc1QWAW3q8PbIN6enpuMo1ZNKdNxMCTDdk9uzZg9TUVPz73/8mPbGSkpLw9ddfk1o9ro5IJMJHH31EQpKrV6/u1KMml8txzz33EAP6ueee65C+b2+GDx9OxOB9+/blhVcNQTwyFOAd4w1KTCF8KnM+qVvUqD5ing7KVliaucSGWigxBVkSM3lpSG+EPFfXhuG3334jtVbi4uJ4PdzsCbtNcrkcBQUFBpdRt6pJKMxQWAkAxFIR/LTC7OZrzXjztTcBMBO7Dz74wG4hdFcxZNjQUnBwsE08T56eniSsmJWVdVO0KhAMGRfDy8uL1HTRqDTYu/B3/DJnD5pqGYNgwYIFTuuszBIQEIDg4GAAzGzsk08+wYYNG7BmzRq89NJLePrpp/Hwww/jvvvuw6ZXN5H1NuzcAACoo2uh9Gdujg3pjaDVzotPx8XFkdmLIUOmtLQU8+fPx8yZM8nFWiqVYtWqVUhPTyceDndh8ODBePTRRwEwN6Bnn33W6LJPP/00EYwPHjzYoLjW3kREROCbb77B448/jt9++w3e3h3FolxatRoZrygviKXM5Y0bXqp0UvZSeHg4maCYasio29RozmIMFlmSL6Lv1p33Zf9jwksqlYon3l63bh28vLxsNexO4Qp+jXXC5oX6jBgyAEfwqwFC5IyWa9myZTavSszFFQyZ9vZ2MpmwhTeGhZUoqNVqh1WTdiaCIeOCPProo/Dw8MAoj9GIuhwN2Vk/TJYyRcdcJYzBhpdqamrw+OOP47nnnsPKlSvx1ltv4d1338Vnn32Gr776CorrunTKYyW6tgrhI7WzZLkaTVctb9hoLRKJhFxAuC5ytVqNjz76CCkpKdixYwdZfuLEibhw4QJeeeUVSKVSp4zZWl5//XVyU92xYwf++OOPDsv88MMP+PzzzwEAPj4+2L59u9PS5OfOnYsPPvigyzYIqmYVFDXM8eYdpzN4QseHkFolFb9XOa2JJLf2iillC5oym4mR79/fH93u1NXLKf2RMWS++OILcqMaOXIk5s+fb+thG4VryBjT/XSlj2Hh6mR6ShIREBCAN9980wajNE5MTAx57CxDJj8/n4QEbWnI3Gw6GcGQcUGioqIwf/589JDoQjjJkhRMnDjR4saGtoZtRNkVCRLm5GzSNKFSUwE/Pz+sXr0aEaN1s+SGc64RXmptbUVJSQnOnz+P0aNH4/HHHyepsqGhofjyyy+xb98+XjjKHQkKCuKl5j7xxBNob28nz4uKivDggw+S5x988IFbbHMLRx/jE6szZMQ+YoSmMbN8RaUC9Wedc7xxdTKmCH65Ql///v7wjvFG0PBAAEBzVjPKTpfhlVdeIcu8/fbbDhNhA/xw2Q8//GCwfIGhrteG8OMaMuJErF69GuHhpraLsQxX8MjYqlmkPjdb5pJgyLgoTz75JLqJosnzRHGiy3hjAKZJ37Zt27B69Wps2rQJ3333HXbv3o19+/bh+PHjSE9Px+UTVxAqYsqjx4ztDrVajcbGRrz00ksIHKKLBTvrxsLCvUk/8sgjGDp0KE6fPk1eW758ObKysnDvvfc69EZhT+69917SZPHatWuk0KJarcbixYtJpdG77roLS5cuddo4zYFbQ4brkQGAiBnOz14yVyfD1ZexPYmi7tQVlPzpxd2oqmJSyufPn49Ro0bZaqgmkZSURIyB9PR0DBs2rIOBZqjrtSF+St8FDc14Q/v69cVjjz1mhxHzcQVDxlbNIvXhemT27t1LWorcqAiGjIsybNgwJPknkeexkjhMTXOs0LIzvL29sWDBAixevBj3338/5s+fj9tuuw2TJk3CqFGjMHDgQIS16DInggYG8kR7/v39ydHnSplLe/bsIWLelJQUHDlyBF988QWpwXKjoC/8ff3111FYWIi33noLR44cAcD0ffn000/dxnjjeWT0DJnwaeHkeHNWlV9zU7C5GUt+qYzHImp2JNkOz3NMaNPT0xNr16614UhNw8PDAzt27CC9nPLy8jBy5Ejs3r2bLGNKaKm+vh4vvPoCijWMVqQ7HQORxv63puDgYKIncgVDxpYemYiICGLMZGZm4tFHH72h2xUIhoyLQtM0Imjd7EsEEZoz5Z2s4XrwXON9/XnvSWQSkqnQlNkMlVzl0LFx0Q+bSKVSrFmzBufPn8ctt9zipFHZnwEDBpAMuNbWVsybN4/UHxGJRNi2bRuCgoKcOELzaOWGMeL4N01pqCcJy8ivy9Gc7fhzyZwUbI1Kg6YrjEfGp6cPPPyZ/r7ScClCxzFGdaQoEkniZDz55JNG6+rYm+HDh+Onn34ihQubm5sxe/ZsrF27FjRNk9CS2FdM0uD1efXVV1FVVYUclfamrgKaHaCb0y+K5wzsFVoCmD5kbAuRrVu3Wt1nzRhyudyiHmK2RDBkXJS20nZQCv5MuOF8AyiK6vSPWwgrOTkZUqkU5eUdC2jl5eXhnnvuQbdu3eDl5YXu3btj9uzZJEvFFnDLjfv39+vwfuCQQOaBBmjMcF5hvKFDhyIykjEap0yZgkuXLuHll192mrjVkaxevRoRERGIFcdhwpVJGCliwhMrV67E2LFjnTw682jRqyGjj7OzlwIDAxEdzYSLL1261OkMWX5dDk0bE2oJ6MefBLQO1Bls0wKm85pEOoOoqCgcPnwYd999NwBmEvbiiy9i8cLFpK6Pb4KPQc/epUuX8NFHHwEACkW6FO6GC00dlrUHrCHT0NCApibHfCcX1iMjlUptno06YMAAbN68mTx/5plncOjQIZt+R2NjI2bMmIGFCxfyCrg6GsGQcVHkOR1njA0ZjSgrKyN/7777Lvz9/XmvsamYf/31F5llb926lfc5SqUSU6ZMQUNDA3788UdcvXoV3333Hfr164f6+nqbbUODtgaGyEsE314dC/gFDnYNnYyvry+uXLmCjIwM/P7770hMTHTaWBxNQEAA/vOf/+Axn38gTToez8texO1DZuP//u//nD00s2FTfUWeFLyiOmaU8ar8Ojm8VFdXhzVr1qC52bDngauP4U4CaJrGql9fg5JmqjJPkk1GgJ/tq96aC5vZtmbNGvLagR0HQWvrD+p3vQaYbXniiSdIKHfUQp3Gp0NFYzvB1clY2kzVUmiaJh6ZHj162KVezvz58/HCCy8AYPRvd911l9GaP+ZSX1+PqVOn4q+//oJcLseiRYuMNju2N4Ih46LIDbi+GzMaERkZSf4CAgJAURTvNZmMCdds3LgR99xzD+69915Supzl8uXLyMnJwccff4yRI0ciLi4OY8aMwZo1a2xWF0XVpGt255cig0jS8VAL4Ap+nZy5FBQUhP79+7uNHsSW3D37bvTxYMIeYkqMZ6Keg0QicfKozIOmaTL7947xBiXquB99e/pC1psxqOtO16O9sr3DMvaGWxfllVdeQUJCAt55550OYswGvYwllh9++AEHTx3E38q/AQCSZgnpju1sKIrCyy+/jF27dsHX1xfdxDoPQ71XfYfld+7cSTwECQkJWL5qGXnPGYaMKX3jbEl5eTnZ77YOK3FZs2YN6RJfXV2NO+64Ay0tLV2s1Tk1NTWYNGkSTp06BYDxNu7YscNg/zNH4F5XKwfQ/ulo0M0VJixJI1ythkIsBmD9zY+SRUD68HHyXM7peEuJKdBqGs1X5VC3qiH2Fnf6WU1NTdixYwdOnTqF5ORkNDQ04OjRo0TvERYWBpFIhJ07d+Kpp56CWNz551lC4xXujNLf4DJ+STKIfcVQy9VOz1yyhMLNRai6Xo2Y12Ig8nLfOUH93w0QQ3cMyI/LUXWwGmETQp04KvNQVCugljMze/2MJS7ht4aj+VoeQDNNJGMWd16bxtb861//QklJCbZu3QqNRoOqqio888wz+M9//oP/+7//w/Lly+Hp6clrtOqvDS21t7djxYoVAIAjisMY7clknZX+WEbSy12B2bNn48SJE1g/fQOgLSr79tcbUDq2GIsXLwbQsRjjO++8A79IP3jHeaO1oBWNl5pAq2lQYvtOLJyZuWSvjCV9xGIxvvnmGwwbNgw5OTlIT0/Hgw8+iK+//tqiiVtlZSWmTJlC0rrDwsKwdetWopNyBu579bUTdHMF0FRiwl8pxC0VQFOpict3/qdvPHFDSyFpTBVdWk2j8VLXcdxvv/0WvXr1QmpqKsRiMRYsWICNGzeS96Ojo/H+++/jlVdeQVBQECZOnIjXX3+dJzyzFp7Qt59hQ4YSUwgYyLzXVtKGtnLHz5AtpfZkHa6syEL157XI+yDf2cOxCkMz+qxXrjq14rK5cFOvfWKNp/k6O7zk7++PTZs24cqVK1iwYAF5vbS0FI899hiSkpKwZdMWcv54dfOCVCuS/fDDD5GXlwcA8B7tBbEvY3yW766ARmG815Ez6NevH55coOsbV9ReiHvvvRcrVqyAWq3G2rVriQdk+vTpmDVrFgBdmrm6RW0wvG5rXMWQsadHBmA8zj/99BPp0ffNN9+QkgvmUFZWhvHjxxMjJioqCgcPHkRycrJNx2sugiGjByWLAPyiTfjrBrVPBODXzcTlO/+jZPzuyWxoSewrJr1iAEYn0xWbNm0iMx8AWLx4MXbs2METs/3jH/9AeXk5tm3bhlGjRmHHjh1ITU3Fn3/+ae1PCIAv9A3o11Hoy8LVyTi7MJ451HB69hR9WQyNyrVuJOZQe0xnyLChl6YrzSj+xrGaAWvgCX078cgEDgmANELbRPJQjdOy5ZKSkrB9+3ZkZGRg9uzZ5PX8/Hy8+OBLUDUx3iVWH1NdXY3XX38dABPCWffOOmKUqRpUqDrQdaVgR0OX6wzhUk0pAGD9+vWYNm0aKcjo4eGBd999l3gGuN5bQw0zbc3NYsgATNbcl19+SZ6vWLHCrOt9UVERxo0bRypJd+/eHYcPHyZ9nZyJEFrSgxve6QyNRoOCggLExcXZXKSlUWiIcNG3pw8CBupu9l1l91y5cgUnT57E6dOneV2L1Wo1vv32W17FVj8/P8yaNQuzZs3CmjVrMG3aNKxZswZTpkyxehuIa1wE+PXpxJBhM5fA1JPhFi5zZepO15PHbaXtqNpXjYjp7jF2LqomFRrSmX0l6+2Lvm+n4uRtTDHAa29dR9ScSEj8XP8ywevp00kFWUpEIXxaOGN8tmlQfagGkTMjjC5vb/r3749du3bh9OnTeOWVVxixuVgnNv/m2DaU/FSEffv2oaGBMfSXLFmCgQMHouLOSpTuZDphl/5Y5nLHH1tDRiQVYdV7q/DPp/4JtVqN/fv3k2WeeuopJCXp6mVxhc2NF5oQPc++Y3SmIcP1gNsztMTlzjvvxMqVK/H6669Do9Hg7rvvxpkzZ7r8/vz8fEycOJF4BOPj43HgwAH06NGj087njkLwyLggLXktgPbY8E30hX9fPxIrbsjo3GuxceNGjBs3DhkZGTh//jz5e+aZZ3jhJX0oikJycjLkcuvduRqFBk2ZTDaGrJcvxD7GNTgBLpK5ZA60hkb9Gf5YC7c6r3uuNdSdriMhpOCxwQgeFYSI25gbYnuFArkf5DlzeCbDNosE+O0JDOHs8JIhhg8fjr179+Lw4cNI6zGevH689DjmzJmDDz/8EACTHcRmBoVNCIVHIGNkVv5W5dRaTPrQGhot+Yxx6RPnjccefwx//PEHaTYLMGGJlStX8tYL4HhkHCH4DQsLIwJVSwyZ+jP1ODLiGEpfLTc7FMv1yDiyDtBrr72G2267DQCTPTdnzpxOr/vZ2dkYN24cMWISExNx5MgRp9UuMoRgyLggXKGvb09fiL3FkCUxLv/mLDnUbWqD6ymVSnz11VdYuHAh+vbty/t74IEHcOrUKVy+fBnnz5/H7NmzsXPnTly5cgXZ2dnYuHEjNm3axHNzW0rz1WbQSm2zOyP6GBbvaC9II5lU2Yb0Bqc19DOH5qvNUDXxbxpV+6rQWux+ZcBr/qojj0PGMjeZ5Fd7g/JgDOfcj/PRWtLmlLGZQ0sn7Qn0CRkXTPQlVX9UuZQWaNy4cbi9/+3keY46m/f+ihUrSL0RkacIkbcz9Y/ULWpU/l7luIF2QVtZOzTtzGyMreg7ceJE/P333xgyZAi8vLzw+eefw8+P762Vhkt114MLjXavRisSiUhtH0sMmZz38tCS24KGXxtR9ad5vz9ryERHR3fZ1d2WiEQifP3118QTdvHiRSxdutTgb52VlYVx48YRPVNycjKOHDnCa7jpCgiGjAvCrTrqm8hcBPwHMAYBrabRdNmw4Pfnn39GTU0N7rjjjg7vpaSkICUlBRs3bkT37t0RHx+PVatWYcSIERg8eDDee+89rFq1Ci+//LLV42+40HXGEhe275KqWY3ma65fvbju73ryWBKm9TZpgKKv3M8rwxX6Bo9mqvj6JvgibnksAEDTqsG1N647ZWzmwKZeS/wk8AjsPAVU7CVG2EQmI0tRo+SFCZ0NTdOkhoxHiAc2/riRlJpPTU0ldaJYunF6L5X+UOa4gXaBsdYECQkJ+Pvvv9HU1ISZM2caXJe9ZqgaVGS/2hM2vFRbW2tWWjJN07xjJ/8z09O3m5qaSJ8sR4WVuAQEBGDXrl3EkNyxYwfWrVvHW+bixYtIS0tDWRlzXPXr1w+HDx9GVFRUh89zNoIh44Jw1fq+PRlPTMAAjghOq5NZsmQJr4Dd3LlzoVarERFhOOZ/5coVvP322wgNDcV7772HixcvoqmpCY2Njbhw4QKeffZZm+h9+KmjxvUxLO4m+OVevCKeCSdhv6KvS9xK9KtsVBHNlSxZBmmYrohc4nMJJGxR8l0pGs677n6h1TRaixmvkXect0kppa4YXgKA9rJ2KKqZYncB/fxx++zbkZ6ejqysLJw8eZJknbAEjw6GNILZb1X7q6Gsd05BMn0667FEUVSndYoCODoZU5IbrIXrXTCnKJ48p4XsKwCoPVqLRiOTTH3s2ZrAVJKTk7Ft2zby/KWXXsJvv/0GgGkCOmHCBFRWMufG4MGDcfDgQbt3JLcUwZBxQfRDSwB4gl9HnNzWYLYh40KdsE2h/m9mjJQHBdk4X4RNZWb37eXtLuXe74q6Uzp9TMgYfk8lzyBPJD6nu8BmvnLVZZvOtZW2gVYxY9NvFmmMsKmhxACt+K3SZbatgXvuaD0TIpEISUlJpNglF0pMIeoOxitDK2mU/2JKDSz7I+cYMr49zAubcL243OxHe2Gp4LfuVH2H1/I/M61qrqMzlowxa9YsrF69GgDjYVq4cCG++eYbTJw4ETU1TGbm8OHDsX//fpdunCsYMi4Im3otjfAkzeL8U/3I3nJmX6KuoDU0ufh4dfeCZ1DX/YoCBgaQmoL15+rtODrrUdQqyP7x7+8PkVSEmPs51UHdSPTLTbsOHhvc4f245bHwSfDRLluHyt9c00hr4TWLNO2m6RnkiaBRgcz6uS1ovuoaIU1e/SUD/ckM4YrhJVO6XhuD6312hODXYkPmtE5fxl6/SneUoZ3jpTGGK3hkWF5++WUiR2hoaMCiRYuIp3/MmDH4888/ERgY6LwBmoBgyLgYygYlFFXMieCbqHMji33EkCXpukUbE/w6m5a8FlJhVb/ZnTEkfhLdtl1uhrrVNbcNAC9bKXAo40kKnRBCmhRWHajm1TRxZfj6mI6GjMhThORXdJ3Bs167Co3S9UJnXTWLNAYvvLTHNcJL3B5LASboywAm849NOa/5qxbtFc4vLMkal5SYgneMeR4Zr2gveAQxOidXriVTd7IeAOOZDZzLXAs07RoUbe1aK+Ooqr6mIBKJsHXrVvTp04f3+oQJE7B37174+5t2HDoTwZBxMbg9ltiwEgs7U6FVNJqu2L/NvSV01fHaGGx4iVbTDrl4WQpXHxM0LBAAU5sk5l7txZBmCuS5OspGJQlRylJkpHqsPhG3hSNoFBN2kue0oHCzY/vRmAKvqm+c6bN/V9TJsMe+RCY22ZNBURSi7tAKMDVA2a6O3e4dCU3TaMllDBnvGC+IPMy7zVAURcJLikqF3St+W2LIKGo4ntkB/gi5N5jcTQs2FnVZadlVQkssfn5+2LVrF4KCmHN96tSp+OWXXwyGM10Rs44whUKBVatWYebMmUhLS8OSJUtIqWIA2LJlCyZPnoyJEyfivffe48WdL1++jAULFmDMmDF46KGHiBIaANra2rBy5UqMGzcOM2fOxN69e3nfu3v3bsyYMQNpaWlYtWqV0zpsOgK+PoZ/ITMk+HU1GkxoTWAId9HJ1HMylgKH6cYcsygalITxLxd/U+ySngsudSfrSa2iEANhJRaKopCyWlew7Pq/c1xGUMrCK4ZnYmiJWdYHfn2YC3XDuQa0lTk3zVxRq0CbVrTs18/fYONLY3Sbywkv/c+5hoyiRglVM+NVNdT12hQCeIXx7Huts8SQ4U1ohgfCM9qDGMbtFe0o+7nzfcCGlvz9/V1Ge9KrVy9kZGTg1//twa+//gofH8v2nTMwy5BRq9Xo1q0bNm7ciIMHD2LhwoV4+umn0dLSgr/++gs7duzAli1b8P333+P48eP46aefADAG0IoVK7BgwQIcOHAAAwYM4BVC+vTTT1FfX489e/Zg7dq1WLduHfLz8wEwxXjefvttrF+/Hr/++isqKirwxRdf2O4XcDF4HplEPY/MQE7suIvCeM6i6ZKFHhk3yFzSqDSoT2fG5hXtBa9uXuQ9aYSUcyFTuMwM3xg1HH1MyBjjhgzA7Jtu85gZv7JOiey3bdeTyxbwasiYEVoC+F4Zc4XaGpUGtcfr0J5rG48BL6xkgkiei1+KH2QpjFFW/3e9U8Ob1uhjWBzZqiAiIoI0zjXZkDml08cEDQ8EAMQ9qMt+yv9vgVEBuUqlQkEBIwpOSEiwqHGjPaDVNMqeqgCWi/DXiOO48OQllOwodbqBbwpm1R739vbmlbifNm0a3nnnHRQUFGDPnj244447iHW7ePFi7N69G3PmzMHZs2fh4eGBOXPmAACWL1+OSZMmoaSkBNHR0dizZw/WrVsHmUyGfv36IS0tDb///jsefvhh7N27FxMnTkRqaioAYNmyZXjttdfw6KOPGhyjQqGAQsEXW0kkEnh6di06NQe2LLOtyzNza8h4J3jzPt83RcaYnhqg4Xyj00tDG/oN2IuOR7AHPCM9TR6jT5IPRN4iaFo1qD9b7/RtM0TjpUai/wkcGtBh+7vfF43y3UzWSOHWIlIh1xWpOarrFRU4MqDL37vXyz1R/ksFNG0a5H9egJgl0fCJ97HbeWAObFVfabgnKCll1ljCpociewNjmJXvqUD3+6K7XKetrA1FX5Wg+KsStJe3g5JSiNgbiYC+1mkJGi7oDHi/fn5m/6ZRd0TieiZTQK/0h1IkPGn/yquG9n9zji7s7d3D26Jjw6+fLqTRkNFg9mcoahW4+up1UBIKfdYlQ+RpfM5OURS6deuGoqIiFBcXm/RdtSd1hoz/ED+0tbciYGQA/Pr6oelSExrSG1F7uo6En7nk5+dDpWIKaiYkJLjMta7meC2qDzHXhZb8VrTkl6D4ayYd3aenD4LHBCFkbDCCRweRlH/AfvdCFlNKgljVRKWwsBCNjY2IiYlBXl4epk2bRt5LTEwkccDc3Fz06tWLvOfl5YXu3bsjNzcXfn5+qKmpQWJiIm9dNmSVm5uL4cOH894rLy9HS0uLQdfX5s2b8fnnn/Neu+uuuzB//nxrNtUobMVDW1GXWc88EAPVVBVqCvjN4DzjPaHIVaAxswl51/M6PUEdBfsbKKtVRKjsmeiBwsJCsz5HmiRF6/lWtBa2ISc9B5Jg1+rxU/tHPXlMJ2rIdrP/6TgaHtEeUJYoUXOoFtf/ug7PGNsa0LZA3aQmWiZpL0+UNZcBJkiughYGomZzLWgFjfQXM9B9bTfynq3PA1PRtGnQXskcc6JIEZnpmgodSEMSLoGqUoWaI7XIvZIHsW/Hc4rW0JCfbkH9Dw1oOtIMcPTodDuNrE+uInKFdYZr2QldOKI5tNnsbdEM11WbLviuEOI5jrs2cPd/1XndNUvuY/52AAAtoiHyoaBpoVGXXmfWZ6ib1Sh8rBhtVxhPmaq7EkHzAjtdJywsDEVFRaisrMS1a9cglUqNLqtp16DhPDNh84z1QGU7430tLi6GbK4P8UpnvpOF6Lc6Fo87ceIEeRwaGmrR72MPyr7mpO5TADgOpZacFrTktKD4S8aw8ezhCZ8h3vAd6gOfId6QBEnsdg0wpRWCxXcKVteyZMkSyGQytLS08Io1+fr6orWVcW+2trZ2KOTk6+uLlpYWUknR1HVZ8ZExQ2bp0qVYtGgRfyPt5JEpKipCTEyMzZpG0hoa14qZGZVPnA/ie8Z3WKZhaBNKc8sAFRAsD0ZAr4AOyzgK/d+g6prONR8+PAxxcXFmfV7r6Hbkn2dOar8qf4QPCutiDcdSn6NzcfeY2gP+MX4djgHNMhrXXmf2oeYAEPeKeb+BI6j8vYroYyLHR5i8n6JXRuPIL8egqFKgaV8z/CsCEDDU3+bngTk0X2vGVTC/d2DvQLOPOQBomdmKws3FoJU0fLJ9EHm7rqCkolaBkm9LUbSlGC15euEaETObp9U0mvfLEfNeDEQSy3+DwlwmrCGSitArLdFskSzigOohdWg424D26wqEtIWQbEB7Yeg6WF+nO0/iRsTBL86yMVT0r0LdyXooy1SI8ouCZ3DX13B1ixpnHj9HjBgAUJ/TIO7Zzo+LhIQEnDt3DgBzv+jsOKo7WUdasLTHtWPp0qWoq6tD7969kdA9AWN9x0Esl6DxQBOSFb0R3IsfuuVqQAcOHGjRMWtraDWNnMP5AACRlwjj029BU2Yzav+qRc1fdWg410BqNQGAIk8BRZ4C9TsZL6K0pydi7uqOxGedI1y2yJBRqVR44YUXEBMTQ0JNPj4+vMZTcrmc9I/w9vbu0JRKLpfDx8eHGCNyuZwYKZ2t29zcTL7PEJ6enjY3WjpDJBLZ7ALeWtYGdQtzh5H18jX4uQEDA1D6PSOUbrrQjKDBQR2WcTTsb9B0STetD+gfYPbvEjQ0EPlgDJnGc42InO68rsSGaNCmXou8RAjsH0AUZtxjIGZRd1xfmwNaSaNkeymSXurlEl4zLnXHOf2VbgkxeT95Bnii1/OJuPzcFQDA1VeuYcRvwwDY9jwwh7Yi3Q3LN87HojFEzIxA4WbGiKjcW4Wo2ZGoP9OAws1FKNtVTnoGsUgjpIi5rzti7u2OKy9nomJ3JZQ1StQdrUPYJMuMb1Wzigj9/VJkkEgtm2N2mxuFBq1Yvvx/Fej9kmNSZ7n7vzVfZ/DJehi+jplCQH9/kuLcfEmO0PFenS6vbtcgfWkGWYel9q860EpALDU+Dm5139LSUl6EQB+2ICYAvL/3fRxuPwwAJIpQ4l2Ke7wXA2rguUEr8Jvfr4iPjyd/6enpZP1evXo55bzRp+ZYDfGmh00OhVe4F7zCvRCWxhT7VMlVqDtdj5qjtag9VouG9EZej7L2HAVaC9qcti1mf6tGo8HKlStBURRee+01IlTq0aMHsrN1Dc5ycnJIWllCQgLvvba2NhQXFyMhIYGotk1dNycnB5GRkW6lqDYVfuq14e0LGOCPt5v/g5m10/H2F2/z3tu1axcoisKSJUtAUZTRv/j4eLuMn5d6baZYEeALfl0tc6m9sp108w0Y6G/UOJGGSRE5kzHAFNUKlP/qGpVWudQc0xoylK6/kqnE3BtNGpjWn21A+S7nbh+3GF5XzSKNETImGBI/bRfpvVU4NvEETkw/hZLvSnlGTMi4YAzaPAATMsah9wuJ8I72IiJoACjdaXm2UOPlJuLK9x9gufERNTtSV5ztxzKnVCxmxb5eUVKIvcUWfw6/wm/ngl+NSoPzD2ag+gCj8ZD4SRA0IhAA46WpO1HXydrmZS4VHdC1MbiivNTh/T1tv0JJM5l9t0pnoKGyAadPn8b333+Pf//73/jzzz/Jsq6Qeg0AZZzzOGpOZIf3Jb4ShE0IRfIrvTH695GYkjMRQ78bjIQn4uE/0B8QAcFjnTepNtuQefPNN1FTU4O1a9fy+mXMmDEDP/74I4qLi1FTU4Nt27ZhxowZAIAhQ4agvb0dP/30ExQKBTZt2oSUlBTSdXTGjBnYtGkT5HI5Ll26hMOHDxO9zfTp03HgwAFkZmaiubkZmzZtMtpszN0x1GNJH/++fgAFeMITG09tRF1dxxP0vffeQ1lZGfkDGO0Q+/zvv//usI6yQWl1ITr2YiP2ERsdf2d4dfeCZzjjTat3sU7Y3EaRhgR8XHiVfre4Vk0ZZb2SpLP6pfqZVHmZi0giQjInHfvq69c7eCwcCbepoI+ZGUssIk8RwiZrZ55NKl72kEegBPGPxmHcybEY8b9hiLo9khfyCZsUCpEf87z81wqoWyw7h7jVus0pW6CPV6SUZKG15LWiId2xZRqUjUooapibOFsV2lK4Bl1nmUu0hsaFxy+h4ldGqyLyFmHot4MR92AsWabqQLWx1QGYZsjQNI2tm7ei/ChjsDZoGlApqcR7772HrKwsXLt2Dfv27cP6z9ejPom5LvuJ/HBn5FyDmUmxsbEu0UVao9KQ1hYiLxHCp3TtVZT4SRA+OQzJryVh9J8j0Ht/T0TMdJ4H3Sz/ZVlZGXbt2gWpVIrJkyeT199//32MHTsW8+bNw/333w+NRoM5c+Zg9uzZAJhwz/r16/H666/j3//+N/r06YPXX3+drP/www9jzZo1mD59Ovz9/bFixQriNUhMTMTTTz+NZ555BnK5HBMnTsTy5cttsOmuR2ep1ywSmQQefhIM1AxCmaYUb6x5E//ZsJ63TEBAAAIC+NqZwMBAREZ2tLQBoOL3SpxdlA5ZL1+M+m1El92DDaFsVBKPhV8fGeljYw4URSFwcAAq91ZB1cC42mW9zDeI7AG/fkxgp8uG3BIM354+kOe0oOavWjRnyyEzsj8dTe2JOjLz76x+TGeETQpF6PgQVB+qQVtRG0peLoPkdg8E9AuAX4rMqlm4ufBqyFhYswQAImdFoIxTfyVgcADilsYg6o7ITrdHJBXBf5IM9buYjLaK3yvR7Y6OAs+u4HozA8woW2CIqLmRpGpz6Y9lPE+nveHqiKzZHwAg6+0LkZcImjaN0bYsNE3j0r+uoHQHM2ETeVIY8tUgBI8MgqJOQbI8q/ZV8eoh6dOVIdPY2IjHHnsMR7YfwX8DmWSSIq9CnD50Gv369UNBQQHi4uJIUkvDkAYcm3QSAPBIj0exOX8TiouLkZ+fj/z8fNTU1OC2227rtHmmo6g9VkeaX4ZPCYNEZv6YxH5iSHwdd97rY9aIo6KicObMGaPvL126FEuXLjX4XmpqKr799luD73l5eWHNmjVGP3fWrFmYNWuWOUO1mJPjx0NRaUINEJqGSq1GkVgM2KAOgGd4OMQRG8hzY6ElAPAI9oSoQYT7vJdgw0fr8dTTT/JORHPJ/k8uQAPN1+TIeTcXya8ZP+GNwQ8rWT6jZA0ZAKg/W+8yhow5HhmKohBzfwyyXrkKgKn029lF1JFw2xJ0VT/GGBRFIXl1Ev5KO84cN4fkuHwok3lTxHgT/fv5wT/VD/79/OCX6g9phKdd6mWwHhlKQvHq+phL5KwI9Hq+J5T1SkTP78Zr0toV/tP9Ub+LudGW7iyzyJBhPQ6UmIJfH+sMmchZEbi8IhO0kkbRl8UImxSKsAmhVn2mqdiihgyLSCKCX4oMDemNkOe0QNWkIiFAgDFisl69RryelJjCoI0DybZ6BnkicEgg6v+uR/NVOVqLW+Hd3bDXrjND5u+//8aCBQuQm5uLqdLp5PVZz8xCyoAkg2nHAQMDEDQyEHUnme9uPN6EnhN6ukwoiQu3eJ+hsJI74Hxz0MVQVFaivbTU5OVt2RVI08x4ZMS+Ykgjjaf/eQZ7AHnAaM8x2BPxK1599VVs3LjRou+U58h5BejyPytE3AOxRk94Y/A6XlsxowwcEkgeN5xrQPcFXdf1sDcahYa46H3ivSENN75vWLov6IZra65Bo6BRvL0EvV9KhNjLeTMWFtIo0gJ9DBf/VD8kPpvA1GDhRgA1gPy6HPLrcpT9qLtAeoZ5EsMmYGAAImaEWy2CpmmaeAG9u3tZ5AVkoUQUeq0wLvDsDJ9B3pBGSdFe1o6qfdVQ1CpMyrBhUbdr0JzFCOV9e/ta7dHyDPJEtzuiUPJ9KdRyNc7cfQ6p/05B7BL7hzH4Xa+t1zH6D/An517j5SYEj9Qds9nrc5D3UT7zhAIGfNIPETP4KfBhk0KIN7VqfzVi7zf8G0RFRUEkEkGj0RBDRqPRYMOGDXjppZdI3ZcB3gPIOhHjOk+3j384jgiP8z8tcJgxaQ4alQYV2tpXIm8Rwqa43hhNQTBk9PAMN7EWhNYjI7GVRyYsHLVXmIuyb6Jvp7NXD85F8okh/8SDW5fjueees+h79bvlato1uLY2GwM+7GfW53B1BdZ4ZAIG69Z1FcFv48VGogMJHBpo0jqeIZ6IvD0SpTvLoKxVovyXCkTP69b1inZEUadAo7bGhX8/P4tCiFx6v9gLcQ/H4vr+bPhU+6LpchOaLjehOasZGgVf36SoUqD6UA0puNV9UTT6v9/Xqu9X1iuhamJuMOZW9LUllJhC1B2RyP+4ALSKRvnPFWYZDc2ZTSS11dRGq12R+p8UqJpUqPitErSaxqVnr0Ce24LkV3tbZfB1hS09MgCTucRWJ2nMaCSGTO5H+bi+TtevqN87qeg2t6MnLGxSGK6vZZbrzJDx8PBAZGQkSktLUVxcjPLyctx///34448/yDIjRozAlLapUBYpIfKkeC1jDBExIxxe3b3QVtyGqj+rXSrEzFL7Vx3RNIVPCYPE1z1NAvcctR0ZeehQl8vUNtLYd4bGX+n1mDoqELeNtj7lrPlqM46MPgYA8E3s/ALgEajbbb2re2PatGl48cUXsWTJErO+k6ZplOxkDBkNgFaRBL4aFUq+LUWPx+Lhb4aLm/XIUGIKfimW167w8PeAby9fyK/L0Xi5Ceo2tdM9GXWcjtdsOXJTiL2/O0q1v2/R1mKnGzK1x3X6mGALw0r6eAR6wHeoD+LiYnW1dJQaNF+Xo+liExovN6HxYiMaLzVBWavr0VTxawXod1PN6iekT6uFPZbsQbe5jCEDMJMDcwwZSxutdobEV4LBWwcia9U14rXI+ygfLXktGPDffna7YXGzyHx6WL9PuJMiNvxWuKWIhG0BIOWNJF3TVj0CBvrDM8QDiholag7XQKPQGPUEdu/eHaWlpSgvL8eAAQNQqZUYUBSF559/Hi8/8X840u8v7ecGdHldEklEiH8gFlmvXQMAFHxeiNR1KSZuuWO4EcJKgND92iKyCoC7XwM++CkQPx+zzWc2d9L1Wh+RhwhiGXMSNV1pwpuvv4ndu3fzKkaaQsO5BrRoa1dc9A3Ct2HaCoo0cHX1NZM/R92mRvM1ZvyyJF+rDQ/SCVtJ8y7yzqKe0yCuK6Evl6BRQZD1ZvZl7fE6NF91bsfyWm5/JQuFvqYg8hDBv48fou/uhpTVSRjxv2GYfG0CJl5MQ8g45nuV9So0ZVn3e3D7CXmb0fXaHvj184NvL92+bi02vdcRtwEsN+XYWigx0/Cz74Y+xAtTsacSJ2f9bbf+OWzXa88QD3j4W+fxA/iJA40XG1HyfSkuaesYAUCvFxPR45F4o+tTIgqh2pCOqlnN07rpw+pkaJomRkxkZCT++OMPvPXWW2g+pzte2dTuroi5tzvEPto+TttLoGxwnYarGpWGtFQR+4hJ5p47IhgyFtA3Qff4Yo7x5cyB2/XaFPcje5HQKGjES3pg0aJFeP/99836TtZbAACHAqKwOzgGLf6MYLLqz2qeMLQzmq/KiWvcmrASC68Ttgs0kGQvfmJfMemWbAqM6Fc3Uyz80rmp2Lz6MaMcW/OBohgxbhgntbOr2h5d0ZLvOh4ZiqIQza0p86PpNWV4+jIL6i91ReySGAz9bjARyjZmNOL41FNovGTb1Gx1qxptZUyBQluElQBA7CWGLJm5HjZlNuPC45eIVzHhiXgkPpvQydoMYZN0N+iqfcbTsPUTJm699VZkZGSQDF1ex+sRpp0/HoEeiL6b8cSq5WoUbSvpYg3HUXO0lnhJw6aEum1YCRAMGYvw96XQQ3vNupgLaGxQ78SUYnhcJAG6g64xoxGrV682q2mXRqVBqTbdVElROOYfDqVIjAMpOlV91mtXTSqo1XjBNkJfFlfqhN1a0oa2Emb2GjgowOwS9NF3d4NIW1G05NsSs2v15JbS6L9Eg9kvaqBSWX6cKWoVpAeMf39/eARYP1u2BK5Ys9ZKQ4YXWnKiRoYlai63OJ5pCQO0mmaK4YEJxdjCi2GIsAmhGPXbcHjHMBOVttI2nJhxGpV/mtfxuzNslQqvD/FSaUCqycYui0HSq71NyoQLnRhKigR2Vk/m1ltvBcDoZd5++2388ssvCOdoJus4jSLNCTHHP6SrZ1PweSGvIq6pNF9tRuGXRWivtE2ndQAo/+nGCCsBgiFjMf20EwF5G5BX1vmypsAthueT0LlHZsuWLfj2Y10qe8P5RsTHx6O9vd2g4UHTNOk8zlJzpJaUpD7lFwa5mLmAfk9FEa9DQ3ojyn/uunJrk41Sr1n8Uv0g8tL2bTlbb/XnWUP9Gd33mxNWYvEM8mSqrYIJp5jye3L54AcaF3OBn48B+86a/fWEWm5bAjuGlbrCv78fxNp6E7Un6qyqPNtSaH1VX1vi28OHeBObLjejKbPrsGjzdTk0rcwExBbnTmf4pfhh9B8jyRjVcjXO3HMO+Z/bpmkhG1YCrC+Gx0VfAB19dzekrksxOZ1fGupJhLlNl5qMhtWmT5+OzMxMFBYW4umnn+aV21e3qolGxzfRF54hpmelyXrLiFeotbAVFb+ZUN5DS/25Bpy9Lx1HRh/Dpaev4MRtp6G2QQFKjVKDcm0BQbGPGOGTXauvnbkIhoyF9OeUA7iYa/3nsaElaYQUHv5du/h4IrgM870WbAEpgAkrsdQ0UYh4Rtep/Oqa69AoOz9x2EwYZlzWe2REHiIEaGdhLXmtUNQozFpf3a6xmQaA5042YxbGJWYJJ7y01bwOsWd1mkZkZBtfriu4YcLgMc4rJV5YRaGlRyAAoL28nedVMRd2XbGvGJ6hrtFlnNey4Ieuw0t8b6b9+yJJw6UY8dMwRM7WVmHVAFdeyMLlFzIt8hRwkXOEvr429MiEjAsmHpXIWRHo937XInGaprF9H41v99OgaZqn/6jStjEwRHJyssHCoQ3pDaRRZNDIQLO3If4RXWPI/E87NxxpmkbN0RqcnnsGx6ecJBWLAaYLddEW67tM1/ylCyuFTwsjOh53RTBkLISrk7lgpU5G2aAk3pGuMpZYPPwlJATVdKW5S2ODi0quIj2AWj0k+FvGF3mVxIWSvhktuS0o6kTbQatpNLGu8XjbucYDuH2X0k0z1NRtauT9Nx8H+h3Cgb6HkftJvtXj4FX0HWpZhdSg4YGQJTNerrpT9SaLXDUaGunXdc8v5lp+oyFCX5Hj9TEsGg2Nqc/S+L4iUDeu45aFl2gNjdYibQ2ZWG+7FNuzhKg5kUScWvpD172OuPoYayv6morYW4xBXwxAz6d6kNcKPi/EmcXpJJ3dEuzlkfFL8cOw74eg/wd9MfDz/iaFd//4G7hnNY2Fq2gcOAeEcTwOVfvND6fVchpRBg03//wJnRDCE4MbarlAa2hU7K3EiemncGrOGVKqAACkETpDPXtDDpSNlu8nACjbpTOyud3e3RXBkLGQ/jxDxsqZjAmtCQzhP5CZwWk4BbVMoXJvFdRyRqtx1C8CKr2OpVmFFJJf1VWivb4+x+gFTlGkJB27beka5wl+u6gno1FqUPRVMQ4P+wuZL1+FUlsX4fLr2VZlCajb9NzJZhQ540JRFGK5XhkTZ1Q5JUAzx2FhqcGsqFGg6QpzfAT097ebDqMrrhczf5d8A8lrtSctM2TaytpJrRpnC325SMOlCEljQnetha08Q9gQ3BuavUNLXCgRhaSVvRnvhoQxvKr+qMKJ206jtcQybyYv9TretvskbGIout8Tzetz1RkHz+muyccuMro7tmxF9aEaaFTmhWfqTnP0MSZmLHGhKArxDxv2ymhUGpTsLMXRccdxdlE66jnlHnzivdH37T4Yn56GqLmMp0hRo0Teh3lmj4F8n1JDvDxiX/cPKwGCIWMxidGA1IM5Gaz1yMizOS5ZM5otBgzgiGKN9CIxRAknW+mAP3NyxHDqAGYW0AgcHEDcz4oqBXI/zjf4WW1ZuoueLTMuuIZMgxFDhtbQKP2hDEdGH8PFpy6jrZR/ARa3q1G42XI3bENGI3EnBw6zrl9N9PxuEHlrRb/flUIl73pGxfXGAEBmAaBQmm8013DSroOdqI85pc2aveYdAKXWg2KpIWOLZpH2gluYrWSHcQEdTdOkkKQ0UmpSxWhbE7OoO4bvHEKSB5ouNeHsvekWhZnYYngSmfNDfdxw//ViGpSYk4bdoDKr2CatoUmI2TPU06RkDENEz48iv3PZj2VoLW5F4dYiHBnxFzIevojmTN1kVJYiw8DP+mPcqbGIvT8GYqkISS/1AuXBnDd5nxSgrdwy4W/N0Voo626csBIgGDIWIxYDvbszB0N2CSBvtdwrw+t6bWJoCQCvsmTDedMMGUWNAtVa5b46WIpLPoybdMEk3TJX8pn/Sf/Xi8zW8j7KN6iYb7uqe82WM0rvWG94hjCeg/pzDTwXPU0zLti/xh/H+Ycu8FzaGB6K12MGkNYReZ8WQN1mWSOJehvoY1g8AjxIDx5Vowrnlpzv0pg5d41/TKnUwNVC87+79phrCH1PXWG2RyES47qXVgOV04L2CvMvyC0FHKGvDfUYtiByZgQRq5f/VG407Nta0AqVNkTQVZVYexJySwhG7x0BL21GU2NGI4q/NS9NWKPUoLWImUj4JPg4PdTHNWSuaecyYRNNS8PWp/lqM1QNzH4KGh5o8bZJfCWIvY/xzGoUNA4PO4pLz1zhlREIHBqAIdsG4ZYjo9FtbhQvjOYT74O4pUyhRXWLGtnrLZtBc8NKUbe7d7YSi2DIWEFyDGPI0LTu5m8J5hTD48JNdTbVI1O2q5zUfClMiQKtPSmnDKUQonU6ZGq9nr4JvojV1kFRy9W4buDEaecZMrbzyFAUhQBt3yVlnZLM9mqO1uDE9FM4uygdTZd1M5jgsUEYtXcEri0djJP+4Tjur/UmVSp4wmZzMKdRpCn0+Ec8ucFVH6jB6TvOQFFrXMis75EBLPP+EaGvyPT6F/bgVKbu8WVfThq2BV4ZV/bISPwkiJjOuDgVNUqe1oELP6zkGH2MMWS9Zfi2Xyp5fnXNdbP0Mq1FbcSLY8vUa0uob6JRxEkMulbETH5CuYbMftMNmbpT9eSxJWElLnEPxBINFbeNR+j4EIz4aRhG7R2BiOnhHcTMeaU03vqaBrUgARJtMdSir4p59w5TYMJK2iJ4vu5dBI+LYMhYQVJ33U3ImvASm7FESSiz4v0e/h7wYQW/l5tMEvxyXd2HA3XW+IBEIEUbwi2tBhqamZMs8bmeJF226MtinveIpmm0XWVmYZ5hnp02urQEbnip6MtinLrzDE7NOcOLIQcM8sfwH4ZixK5hCBoWiKtFzLh/CNXFo3M/yANtZq0fmqaJISPxk0CWZHnbBRa/ZBmG/zAUEm1WWv3ZBpyceRqtJR0zd2iaNmLImLcd7VXtRD8VMMDfpIw4e9DaTvOyri75BJLHltST4c5iXSH1Wh9e9tJOw4Y0rz+ZAzKWOqNRTuOLwmAc89MaYJUK5LxrejqmrXssWYN+Fml9M1DTAHhFSonB2JjRaHJNFr4hY91EwLu7N6IX6lqVRNwWjtH7RmL4D0MRMjbYqLfn/jdpvPQZjbvf8UCPJxiRNq2mcW2NgYtEJ9QcqYGynjFQw6eHWd2g1FUQDBkrYD0ygOWCX1pDk9CIT5y3yWI2FtYlrWnXdFkCvyW/hYgPZSky7KthTupuoUB4EIU+uns/8cpIw6VIeDyeGauKxtU3dCdOW2k71A2s0NfP5u5kbmG83A/yUXNYN7OVJcsw+MuBGP3nSISODyHfzYZernsH4II2bCbPaTGrdgPAzPgVlYyhGjg0wKqeQFyCRwZh5C/DSRZC8zU5Ttx6Gs3X+PuupAqoqmce94nXvW5uqj/XSHBmWCn9GhMaY8n0CQRrdtdZ4JFpcWGPDMBUk2XFpRW/VRoMI/IzlpxryPydxXiWN0X2IvqlvE8KeCG8zuAalrboem0Nhs4REl7iVPmtPmg8DZsLK/QVeYlsEgLs+58+GPLNIIw7ORZDtg5C4KDO9XftChonLjOPL+YCXvNj4RnOXD/Kd1d02nZBn7JdujpWbH2rGwHBkLECnkfGwloybWVtULcwV3hzMpZYeDqZjM4LcHFnhr7To1CnXXyAtiZOSpzuZs0NlfV4NB6eYdoT56cKUqSu6aJ9My64hgyLT7w3BnzSD7ccGY3ImREdjKerHG3vztB48jj3/Tyziq/xwkpW6mP08U/1w6g9I0hTvbaSNpyYeZpX/I/rjZk1GgjSRh7MPc5qjrqI0DeT/1wu9kCBN+PlarzUZHZ2GVtDxjPUExKZ65VWF3mKEKm9UajlalTu7Zjyy4aWPII84NXdy6Hj0+ek9kZZ7umDXSHMjEbTrkHWKtN6rrmWR6bjeX5dW0GCa8hUmpCG3VbeToy0gEEBRhtOmoPIQ4SIaeGQ9TLtep9VyJ8EnM6VoNeKRPL86uprJl3bNAoNKbsh9hXzNEPujmDIWEGIvwZRIczjCzmwqEopL2PJSkOmsZPCeNxO1wBQ2ldnjQ/U1r/jzvwzC3TbIvGToNe/OK0LVjEnDrcQnj1mlB6BHghJY35gaaQUfTf0wbiTYxE9vxuJM3NpaqFRygl9n5WFoNiHuVnWn2nguYi7wtJGkabiE++DUXtGEFe3slaJU3ecQdVBZgO4hsygXhSpJF1SxXRfNxW2fgwlphDsTH3MFd2YE6OZ/xe9teOh+YUHu4Jb8NDbBb0xLLzsJb3wUlt5O/H42cObaS7c/fNdaA/IpbqJS+3xrnuu8bteO9uQ6fjatWI2+zCQ9JyqPljTZXYWL+3axhMaU9GXLRy/RCNmcTSRFdQer0PVn11rfqqP1BDRcsSt4TdMWAkQDBmrYW8wtY1AmWmeSh7m9ljSx5/nkTEu+G3MaIT8OvNdwaODcL5BdwMYmMhcRFMMhJZYYu7rTopc1R6rQ9W+al5nanuJFYd+Mwij9o7A+DO3IHZJTKeht2v6mdYUhW+D+FoZU6ljdTgUX6tjS6ThUoz4eRiptKuWq3Fm4TmU7SrnZSwN7m1ZJen2qnY0X2X2ecBAf3IBdwZs6rW3FLhnCvP4so9lfZfailtJ40BXqiGjT/CoIHh1Yzwt1QeqeRWquRV9nZmxBDCTHK7HrFUsweZQ3QF35eWsLm/4LXmM10IkFcEryvFp5Cw0TeOS9jT35JRLYq8NIg8RqfOjrFWioYtim9zJD7dPmCPR9zCduMxsR9L/6SqwZ62+1uU+4mUr3UBhJUAwZKyGe4OxRPDLT7023yPj4e9BDIzGy01GCz1xZ4Td5kXh/HXdQT9A66XsHg7ItPcF/SysDifOqmto1IayxDKx3WZhYi8xgoYFmjR7yOIYX17aMhZHAiKBMOZmUrm3qksdEcBUPmYbLMqSZXYtIOfh74Fh3w9BxK2MyJJW0kh/IAN+B5krr58P0DMa6N9TN2M39Tjjpl0Hj3FeWKmyjka+9ho6NInxMAHAZY7g15xO2NzmhK7skaFEFLppi5jRKhplnCZ9jU4qhGeIgnKgUu/n/z0gGkjQCmMvNKH4W+NNMGkNTUJ9PvHeNtOTWUJRJdCgPcUnDAIk2ssGG1oCwCsA11kTSYCv37K2lpSl6J/vZ68Bbe00ImdFkArozZnNKPne+D7SKHRF8CQyMUInhthtvM5AMGSsxNpWBZYWw+PChnU0rRo0X+uYjkeraZT9yFxEKQ8KkbdHIEM7Vh8vnaufoijilckvB1ra+BZ+5O38E4ftCu2f6ufUixcLm7EEADNGMv/VlAjlY3XdZ3M/yu/ycxrSG8nsxhZp110h9hJj0JYB6L5IuyNoYOG1LCyszMGAnjREIl1oCTBdWM7trxRyi/ML4QHAiD4647/OQ4r6AMYArk9vMLkzOK/LspM9MgfTgQt5xou/8Xsv6SYTDQ7usdQZJzn7hw0vaygKGeN7k9evvXHNaDq2qlIFjbaRobPDStxr8KBeQII2Qeh6sS70H2piPRmVXEW8zrIkX3gGOafIn74HVqEEzl1jrtfJr3H20VvZRmtmVR+qITWLwqeHQ+x144SVAMGQsRprWxWwHhmJTMzrp2EO3PBSo4HCeDVHa0nRsfDJYWjz9ECOtt5V/56AmKM3YQ0Zmu5YfE3/xGHxc3INDBbueOdP0G3T0W7RJOW55PvSLhtK8hpFOsCQAQCRRIR+76Ui4Z+6/jeLq3JxT85V0BoafXUvmxxaYg0ZSkI5Lb4P8PUXI/pQiI8EfFnPn7ZdAa2kUX/OtGqrrZxMGp845904fzxMY/LTwJ2rIo029PRL9YMsiZmg1J2sJ/2h2Buk2FdscaVYW8HdP8/foztvfmwIRsRMxlPYXqFAznuGQ7OKYp1Q29mGDPfc6JdAoTdTPw4tbSD6Oe9oL8hStNq5cw1Gazk1nGvQTWicFFaqrufr/ljYLKaQMcEIm8IYZm0lbSj4wnDFTK43MGrOjRVWAgRDxmqSY3XuS3M9Mup2DUkj9enpa7HgL6ALnUzJTp3LsdtdUbxxDujJX7ZPvG4M+joZgH/isDi7mBcLm7EkEQMzRzHVlwHgXJEEccuYKxqtpLvsPlt/pp48tofQ1xgURSH51d4ov0NnLMb9XYTzD1+Aj4RGT1Ykm8s0YOyM9op2ookKGBTg1Mwerv5iRAp4HqZT0N0gTA0vcVOvnVlD5pOfmH2goSl8f9DwMhRFods8Xd2Q0h/KoKhTkIJ+/n2d783kesxmjtLp/s5dA6JX9IbIU5uO/XE+77dn4RkyTi6Gx9WT9EsAeutanPE0dCR7iTaehs2rH+OkiQDXMBs/SPf4+CXddiat7E26g+e8kwtlPT8DUN2uQcUebVjJT4LQCTdWWAkQDBmrUJQqIVGrkayNXGQVmtcLpyWvBWwxDVNT8QzRmSGjblWjYjeTcifxkyB8ahhv9jiwF/8iyhf8Gt6WpFd0Jw7g/Bg/wNzY2QtVQjdA5kOR/ZJZAHRbFksuyIVbiqFsNJzuyy2E5xHkYVbLCFuxLzYOb0enQq39kct+LMfJ2//GHapS+KiVaGkDco2HwwEANZxMk5AxzstW0mhonNYaMlEhjA4L0N0seQ0kTTVk2JolIsDbSWnLpdVMV2WWvaeML8vqZADGkOGL5J177iiUNM5pM+QSo4GQAAoTtDdMjQY4Xe9Dmh1q2jXIeu1qh89QFuk8Gr427HptCRe1kzSJGEiKBXrF6C5UBg0ZGK/yW3uK2yjSWUJf3eMFEyn4a28Txy/pQmX+qX6IvpsxlpX1qg6es+pD1ZywUtgNF1YCBEPGIhrON+DMgnPImZ2H8p8rSMxfqTKvFw5P6GuFe9kj0IN0m2281MgT/Fb+XgVVMxM3jZwVAbG3GOezdQbKwET+Z3FTsI21XfDv44fY+xkPhyRCAlmy5UaYrSiuAlq1hTqTtO7kAZz9ktsqRfQCxqWhalKhaGuxgU9hiucpaxkjJ3CY5X1VrOHcNWB/YDes7TGAtDSo/7seEw5cxjdXD+OVwnRc3lRq1BgD+PoYZ9aPuVoINGoP8xF9QH7P/gnM/3IPb6gDmSyXur/rTepKzHozvLp5mV1A0lZ8u5+50bOczwZKqgwb/j5xPmRG33SlGSUc4ayz9TEXcoB2rR0yUtuhYOJg3TF/4ByNns/05NWR0jc4XSW0pFDSyNJef5NjAU8PiueRuV6s2z/BI4NIxfKqA9UdKn/Tahr1fzOhTs9wT5t38zYVrlxhUC/GowkA5bWMSJul94uJEEmZcyH/swJeB/NybhG8GzCsBAiGjEVolDSq99cANFC0tRj9EszPKAH0U6+tMwZYnYymVUNCCgBQsoMTVtIKD89rPTIUBZ6IFAB6RAHaEhIGQ0ssfdYlY/C2geixNdZpNxMuXAMySeuJ0c/06fFYPPEk5f23ABpFx5tmvRP0MVyaW3SepfZBYRj+w1B4x+i8Dh40jRFN1fB47xL2Jx3EmUXnUPJdR6Om9i/mZuN0fQwvrKTbH/3YkCZFoaJbIAAm/ZzrrTCEslFFOvc66+YCANv+7Gi0dOaVieLWlOFklwT0d25Yli2EB+j2T9pAQKQ9pQ+mAx7+EvR+STfjufJyFu/Gryhi9gclppzmIQOYawBbOI6dXPYyEloSeYqIAF5Rpehw3DVlNRNxc/CIIKfV+WHvJxQFpPYARvfVvXf8ku6xd3dvxD3AXPg0bRpcX8dc5NXtGlLVnAkr3ThF8Lg4/w7khgQODYCsj1Ys9ncD+ot1J4E5gl+2xxJgWeo1F0OdsBV1CuI2lUZIETI2GCoVjUtad2Wv7oCvN/8EFYt1s5jrxYBSZXh7RBIRwqeGQRLqGlVV+YYMs00DON6mjGwasl6+iJihFS+WtxvsgVPH0ccEOSHdkimsyDwe1IuZOY4/Nw4jfx2OgIWxqJHoBOEaBY3KvVXIeOwiz6hpvi4nRnLg4ABIfJ1ZP4Yr9NW9zjWgL5lRT6a1kCP0jXXO7D+rgMY5bcHbII4d8tsp4+d+1OwIXRFH7WIiT8omPbys4VRmx/0T6EdhkLbSwoUcoKqeRsyi7vBL1VZizmgkXiWapqEsYQwZ71jnecjYsbKwk8voMKZ2EQBc03PChnHTsPfxq/xy066tbRRpKWo1jcv5zOOe3Zhr9ai+uuv1icv8463nUz1IUkPx9hI0ZTWj+mA1McgiZoRDLL0xb/k35lbZGYqiEHOfztQPOa47Q8wpIc/zyFgZWw4YoLvpsjqZ8p8qQGs1O93mRoISU7hWDLRpXckDEjt8DABdeEmlBrINR2BcDm7qtX5oCdBd5BKe0KX/5H7YsZkk24uKElMI6KIHij1gb5AAMLg3c9GiRBTTo+m9ZDzSbxz+FT8Uh2NjII3QFR7jGjVHRv5FXndm/RhAJySlKGBosu71YH8K0dr7yIH2QPJ6V4JfV2gWyfXGvLQYCJIxboA//jZu+EvDpB1ElrIUP5uUvLcGdv9IPfnXg4mDdY8Pn2fOhz5v6Hbg1TXXoGpWQVGthEauTb12JaGv9twXiSjilcktBVSc/cPTyejVk+FlLjrJo5lbymRbATrDf0QKcy4BfI8MAHgGe6LnU9rrmwa4+vo1XhG8yNkRdh6x8xAMGQvpdlckKC/miGrYXYZIH+ZiZk5TP9YjI42UWl111X+AbmrYqDVkeGGluxgx2HlO6Xu2oq8+xnouuTKGQkuRIUCo1hZh6+YEDQtE0MhAAEDzVTkq/9TNxJSNSjRlMtW0/FJlTsn0Sb/Oj4lzEYsppCZQuOIbhPX+yRjxN+OpiX8o1mjncWc2imxpo4lhnxoP+Pnwjze2dMEltQziAOa3rj1Z12mrD2c3i6RpGt/sYx6LRMDCycAt/Zi7TVMLcOyi8XW5LQsA54eVahpoUihuUC9GU8IyYRBfJwMAIbeE6Dya2nRs1+qxpHvM9fixhoxSBRTo5CLwifWGrzbJov7vBl6/rzqt0FfkLXKajom7PWyoLNCPIhPNjBxA3so/V+IfiiOVlSv3VqFcm3Yt8ZcgdPyNGVYCBEPGYjz8PeA/jbkQqZpUmCdiDpiSKuYC0RXKeiUU1YxrxBaZMZ5BnmSG2nipCS35Lag7WQ8AkPX2JSnSGZ0IfVn4PZesHppDYFOvA2VAWCDzmKIoMsusrAPKa5ht53llPsgnj+vPNhC3vzP0MYCux5JIxK8azcK+RtPAlQLGU9PnrRRMvJjWwajxjvMmRpszOHcNUGs1C9ywEgvZFoqCuncgAKZsPNtWwRCt3GJ4TtDInLysyxibNJjJxBrfXzem304aP/cjZoRD5K275Do7Y+k0R780Um//jO2vKyvBzc5KXp0ESmvw5H2Uz2tK6iqGTIAMiAnXvc7WkgH4FX4BnVeGVtOoPsSkYbeWtKG1iDFOA4cEOC1cxpUpcPV+rE5GrWa6lnMRe4vR6wXdhV2jYD7jRg4rAYIhYxVBdwaSx6OKdGeIKV4ZXljJSn0MC6uTUbeocW2tLse627woIlY7z0m9NhZaMiUF25VoaaNRqJ1pJcWCJ8wzFF4KnxoGWW9tkbITdSTdup7T8dqR9WNYFEpdn5jkWMDHq6PHjCcs5xxnbPiJNWpuOTYGYw+NdmqqJb+ib+fbUtaNU0/mpPHwEq+GjBM0Ml//oTsfFk1hxj+uXytx9+85aXxdiUxCWlEAcErokot+oUIufj4UhmszZK4WMunmAODbwwc9HtGlY+ds0B2Evk40ZOqbaBQxmlb07cG/BvQ2koINAGGTO6Zh8xtFOq90AV/zo3s8mqeT6bhe9IJu5PrGcqP1VtJHMGSswKuPlHg6Aksb0bOVCemYkrnEE/pambHEwhX8lu7g91ZiYQ2Z0ACgmxFPY6/uumJyV9zAI8OdZSXF8N/jzmTY8BIlotDjca5XhrEe6jiGjDM8MpfzGPc30DGsxMJrHmlEWE6JKPgly+Dh71whtiEhKRfutmRIA8njzgS/bFVfkZfI4krYlqJU0aTwnZcncMc45nGwnwbDtfKRS3lAUYVx4z9pZW+EjA1G/CNxCBjkXI+MfqFCfdh6MgBwkOOV6flMAjxDmd+e5tTNcqZHhp0AAB0zMfmZS/x9EzwqiHjJqvZXM3WkuIXwnCT0BXQTYh8vXasFABiVqnvMLYzHIpKImFpfWpiw0o1XBI+LYMhYAUVRiFmiO0turWPuqKZkLjVzPDIyG3lk/A100Q0aHkjKuJfX0KQ53MBeMJpSKPWk0FN74mQVMOp5V8ZQxhKLfuYSS7d5USQEU7GnEs1Xm1F/Rlc3whlC0nSOfmlQL8P7htdzyQw9ljNgPTK+3oxGRp8kTlXsI83+5IZSa8QjQ9M08ch4x3g7PCX2j7+Bam0XhdvHAP6+uu+/daRuud86ScP2ifXGiJ+Goc8byU5L6QW0Ha+1+ycsEIiP6rgMt57MwXTduePh78FLx2ZxZt8r7uSRO3kBOg8tib3EREfWXt6OpivNOqEv5bwQc3MLjRxtCDM1nt9GpncMEKy91J+4DIOasvDpYYjSFmJMfDbB6aJye3Njb50DiLozkhRWSmsoh7daZVpoyUbF8LgEGDBkut3V0RsDdGxNoA8bXmpT8AVyrshVjrtY3yOTEme4hYRYKiIuctDAxWcuk+qXQUOdVQhPd0Ea3LGlFQAgNJBClHZyxaRqu6aRWV6jC/cNTeJfiFk8PXRNSi8Xi0hD0rbiNtKTiIuiUgFNqzZDxgk3TW62EhtWYpk+Qvd4Tyc6GVfhejFQp60aMbKP4UnNqL66mlJcnQwAxCzWpWMDgFc3qUkd6u2FfmsCLqEBjG4G6BhaAoAwThPJ8p/K0XiR8az7pcjgEeBh87GawuV8XRkGfa0cRVHEK1PT0NE4Y5cZ+Gl/TCuejASO9/lGRTBkrEQikxBjwUejRlpDOS7lde3FYDUylISCt42yLzyDPXnF0ygJxYuNdtaaQB93EvxeLeSkXsfy35N68lsVtCt0y8bc3x0SGXPxZYXRgPPSLbkemYFGQkuA7sJW2wiUGW4T43R4+hgDYQsW9qajUgOqlM7ryfAylhxsyDS30PhJm9Ue7M83XABgSG+dyHzfWf5x5op0pV8CAG+p7oaZVwbkl+m2iRJTSFmjS8e2lc7PUriTx756922K0tXGKqjouG+4Opn8zwpJ2xhn6mP4GUsd9w9XJ6Ofhs1CUZRTjUtHIhgyNoAt1w8AM+qK0dJKd9oLh9bQkOcysX6feB+bquK54aWwSaHwDNHpCDprTaCPO6Vgsx4ZimL6xejDhpdUapAS5gDjIo9dGtNheWcIfdVqmmh44iOBID/jhmZ/AwJmV4Ovj+lsW3TvlUZwDBkD4aUWTtdrbwd3vd71l66mx/wJ/FRlgMkyu1Vr3Mhbgb86ScN2BYwVKtSHH17ivxc6LgSJzydAmiRFz2ecN+unaZ1IPiacSVHWhw0v0TRIyIbFN8EXPj0Yw5gtHgfAqRl/XHmCvocJ4Ff4PWFAJ3OzIRgyNiCgvz9xi/dsa0Lv1sZObzBtZW3ERW7rpoTcWQQ3rAToash4enT0XOjD98i47olC0zTRyMRHAl7SjhcxnuA3m/9e/ENxJJ0UACgPymCIzt5cL2ZugIDxsBKLpS0xHAmv9H0nN0ruRfq8OACUhNm22uMdDZnWfOfVkDGUraTPrSN1r+854brnDKAT+lIUMCzZ+HJcwS9bT4ZL4nM9kbAtzqmFF4sqgQam/JPBmz4A9OpuPHMJAMImhXV4zVmNIgHjNXFYhiXrEjKMeWRuJgRDxkbE6ol+L+Qav5DJr9uuNUGHcdzXHTH3dUfPZxJ4YaWWNpqU6O7bA/CQdB5aSuYYOq4cWiqvYQqRAcaNM2OCX4BpPBh9ly4lwL+fv1PcsaYIfVl4mUudHGfOQq2mSX2L6DAgOsw071JGsZgYkfLrcrRXK3jL8lKvHRhaqqil8ecZ5nFcJH82zGXqMF2Pos4Ev86mtZ0mk5qUOCBAZnz/DE9hsmYARifjipqsrm76QOeCX4Bf5RdgipRyw/SOhKZpMkHpFsro4vTx9aaIzvFyPtDQ7Hr7xZEIhoyNiJoTCZG2Euy4hnJcvWK8M7E9hL4sEj8J+r2TiqSXe4ES6U6AS7m6br1dhZUA5kSJ09pBV/Jd8wIGdC70ZeHVkjEgxO7xj3jiCQib6Jw0RV5F3y48MsmxhgXMrkJmAdCstTk608cAjKETqBViXsgFgkYZryfjrGJ43x3QnTv3TGbK3hsi2F+nKcksAPJKXfOcSb+ma67YmbcMYEJot/RnHpdWGzYCnE1nGUssnaVgA0wFbBGnYFzQCOcI/gHmd65l9MZGDTNAl4ZN03zN082IYMjYCImvBN3nM6EcL1oD7xMdGxKyyO2Qet0V/EJ4pp2gbEZJo9x1RaWdpV6zRATrhJgZ2R2NMr9kGYbvHII+byYj4Z/OifXzeix1IvQFGAFzEkfAbKy/j7MwRUjKQlEU8cqUVAEe/QPJe/o6GdYj4xEogYe/47JJOstW0ufWEbr3XdUrY6wjuTH47QrsMSLr6CxjicVYF2wWsY8YwaN1RnSwi4SVDFX3ZuELfl3rGuBoBEPGhnCFo8PyStCkbaamjz2K4XWFKa0J9OFW+HVVwa+hZpH6cFsVVNUz4Sh9Qm4JQfzDcU7pFE3TNAktRQQDUaFd31zYC5xSxTfmXIGuCuHpw735FIdxBL8cnYxGqUFbCaO2daTQ93oRTUr5D0gEUnt0vm9m8OrJuObNxVShLwu3geRBAzoZZ3NR65GRiI2HlwNkFCK0Mh5jXqVudzITUUpC8TKZHI2hLt6G4Al+DVT4vZkQDBkb4t/HDzXRjOg3vr0Z53c3GFyOLYYn8ZPAM9wx1Um5HpnOrHwufeJ1J5Gr6mSyOOPqTMDcVXjJmRRV6lzJxir66tPfhQW/rEdGJGLSkruCGw64VOUBWQoTa2q82EiySNpK2kBrSxo4UujLNogEuvbGAEzafKT2hrn/LNDW7no3/pPa/ePj1TFV2RCDeunqsBxMBzQa19kmhZImmYjJsR2zybiwXpmyGqCppeM2RC/shiFfDcKo30Y4bIJpCK6HqbNrdVyk7lg7ecW19oujEQwZG6OaqnMLlG/r6MNUt2vQqnWR+/b0cUgcVqPhp/YaSk80BN8j45onCauRkXkbb7kAdJ655Gx4YSUTbvwA0I+Xgu06+6a5RZcK27cHIPPp+ljjVyumETxS65XRAHVn6gEALU7Qx9A0TcJKFAUsnNT1OhRFkSq/re3A4Qw7DtACKmppFDD9bTE0CZB0IfoHmGXGaXUyVfVMKw1X4WqhTu/Tr4sJWm9OeCnbSBG5iBnhCBzs3B5YFzgepuROJmcURRGvTKPcdb3mjkAwZGxMj7si0CRmwhNepyqgqNPLvMhrIR2WHVVEKrdUl9rbWaE1ffjNI207JlvQrqCRr70o6zeL1KezzCVnwxP6dpGxxNKfc/M3pZK0ozh7TSeMNSVsAQB99baFq1Vgw0utvGaRjjFk/s7UhSHGDwS6h5u2b2Zw0rA764btDPj6JdPX66yejDPhZyx1vn86ax7pKihVNLnWJsUyerjOGJXadWG8mwHBkLEx/VPE2B/ApPOK1RqUfMevvsTveu2YWL85rQm4BPvr4squaMhkl+humsb0MSyunOmTzvHImBpa6h7OyfZxoe3h3ihHdiH0ZfHzodBDW/LoYi4QyGnUx2Yu8TwyDkq9Nkfky2XyEF2Nj866YTsDnn7JBKEvywSOTsZQPRlnwfVGdhUy5wl+XTD7CmA8TGzjWFMkAFydzM0s+BUMGRsT7E8hvaeuvGzh5iJelgw/9dpBGUucGX9XrQn06aP1ylTWATUNrnWimJKxxCL11PX1ySp0rRLy57RC3wAZv8ttZ1AURUIyxVVAXZNrbI+5QlIW9qItbwXKKS9SJ6b+bAPU7Rp+Vd9Y+08AVCoa3x5gHnt6AHPTTF830I/CGO0N5noxkF3sGvsG0DM0U40vp0+/BCBEG3E5fN51GsmaUkOGhV9LxjXGr4+pQl+Wwb0BD21+ws0s+DXLkNm5cycWLVqEESNG4NNPPyWvnzlzBsOGDcMtt9xC/tLTdf7H4uJiLFu2DGPGjMGiRYtw7ZpuCqrRaLBhwwaMHz8eU6dOxbZt23jfeezYMcyZMwdjx47FM888g8bGRku31WGE95fhok8gAECe3cLLvpBn268YnjEyOCeHqRlLLK4cXjKlhgwXbqsCV9mWqnoaJVXM44GJnYfH9OEVxnMRrwyb2uvn03l8Xx/9tgusTkbTrkFDeoOuhgwFhxQq238OpFP8baNM15Wx8MJLLpKGrVbrMrC6KlSoj0hEkSq/9c18L68zYQ2ZABnTnqAzOPNLlw0tmeNhAphK5kOSmMfXioDqetc00OyNWYZMaGgoHnroIUycOLHDe9HR0Th69Cj5GzRIV9v6pZdewogRI3DgwAHccccd+Ne//gWVivGf/fDDDzh79ix+/PFHfPHFF/j6669x+vRpAEBtbS1efvllPPfcc9i3bx/8/Pywfv16a7bXIfRLAH4L1vkxC7fqzhqeRybBQaElzow/LrLzZfVx5Z5LnTWLNIQrCn4tCSuxcLfHFcJLJVU6o4wpoW76jZI7++ygkzlRR2rIeEVKIfayf+VlS8NKLLdy0rBdpRt2VqGuCnZXhQoN4Wr1ZOqbaBRVMo/79uh6EuAtpRAbwTx2VUOGV0OmCw8Ty2iOZ+3kTVoYz6yiGePHjwfAeElMJT8/H3l5efjiiy/g6emJefPmYevWrTh//jyGDh2KPXv2YPHixQgODkZwcDDmzJmDX3/9FcOHD8fBgwfRp08fjB07FgDw0EMP4a677sLLL78MLy/DszKFQgGFgi+wlUgk8PS0bZqzRivOYP9z6ZsAbPCLQIP4KgLUSpT/XIG2NW3wDPUkqdfSKClEPiKD69uSmgYm9AAw+hiaps2q0pusl7nETfHr7DdwBNzQUs9udJfph7y+Ptk07rUyXdEW23/2qu7xwF7mfVYqJ3U2I6fr7bcH3N/gxGXdvGh4innbwk0DzsimEXifLnOkan8VFFXMOe0d6233403eCvzvCPM4UAZMH278tzV2DKTGA9GhQEk1cPAc0NyiIaX+nQU39GDu/gEYwTPLgXM0nr2bduo1gGu89+1h2hh6dQcKK4C6JqCqTkPCZdZgy9+A3aYAX6BbqGnnNDeEe+wijRkjHXsdsPcxIBJ17W+xWfWviooKTJkyBTKZDDNmzMCyZcsgFouRl5eH2NhYniGRmJiInJwcDB06FLm5uejVqxfvvb/++gsAkJeXh8REXSwkOjoaEokExcXFvNe5bN68GZ9//jnvtbvuugvz58+31abyKCrqaNqHentAJeqGfYHdMLemALSSxqVPLiNwdgCUNUzrAnG0CAUF9o9vHL/iBYCZhiSEN6KgoGMzvs7wE4sBMN6lc1mtKCio7LCMod/A3tA0kFnQHYAYUcEqVFeWoLqLdYKlIgBMDOr0ZcPbYgnWbP+xjFAATIgxwrcUBQXGW1vo4y+mADCuqLOZ7Shg82qdQFFREf48GQiAuTP0CK1EAUeg2xUeGkDqEYN2pQjp15SoEldBHCyGulaNuhP1ZDlNiNru583PJ3zQ3Mo0EZw2pAkV5bVdrmPoGBibGozvDvuhXQl8/0cFJgxos/lYzeHA6WAAfgCA+OByFBS0m7W+Fw2EB0ajsl6CI+c1yM4pIvoMZ1wDDp+RAWBainQLrEFBQXOX60QG6H6DI2fKMDhR0fkKZmDtb9AgF6G4irk+9Y5uQ2FhhUnrxQbqrtEHz7ahYKpp69kaex0DPXp0XezIJoZMfHw8tm/fjtjYWOTn5+OFF16At7c3Fi9ejJaWFvj68rUgvr6+aG1lLnKtra289319fdHSwvg/W1paEBERYXRdQyxduhSL/p+9s4yO4moD8DNrUUISXEJwd3enaCkuxUrxltIWqQu0pQK0FGkpxQsUKVCkuLu7Q4BAgoUY8azN92N2Z3dDPLtJ+NjnnJzMzI7cuztz572vDhhgs81RGpmgoCD8/PxekBiLFpMcsHb4FKNnmDToxmyJpVSHUoAkcuermg9/f/+kp7U7G6zs801qeeHvn7HKziVKSLPSyBgIDHGzaXNq34GjeRYJz01WusqlVOn6Lv2Bgj6S78Oth26UKOFPVtL42KP/t0xBba4aaN2gKKoMPpGli0rh9bceuuDn5082/ww238HNh5aLd2lekMIZLFtVtZQUvh34VE2hIiWJaPKcp1tshc38lQs4/LnZNc+yPLJbHvz986S4b2r3QO+2sOagtHz2biHe6uqI1qafq6b3jFIBnZoXxiMTwV9t6sKqPRCXqCAk3p+GlXNuDHgYaVluUTcf/v5p33B1KsFKkxN3tL4I9riV7DUOHrLKOVSvsmu673N/fyhRSNI0XQ50pWgxf1nAzA5y8j1gxi7dzZ8/P/nzS9nISpcuzbBhw1izZg0DBw7E3d2d2NhYm/1jY2Nxc5OeIjc3N5vPY2NjcXeXfEfSOjY5NBqN3YWW1FAoFC/8eC4aqFzSyMUADy54+lIzJpy4e/EEL3so7+NZ1iNbfvRLdyzqvtrlhBQL3qVGJX8jx69KGWhjEwTyJElyltx34GikqANJhVqxRPrUjwA1yhjZfQZCn8PTCIGi6SgHkBaZ7X9UrChHT1QvAxpNxs9RrbRRyhOUAPefCpQpljOF7kRRwRmTv0+JQlC0QCb6UsYo56G58UAgXyPfFwQZj5LuDr3XQiNFdp2WfpPiBaBlrfQ9M8ndA+3qiqhVIjo97Dgl+XDkVCFCKVGh1K+qpSGPR+a+wza1RVbtkc5z8IJA46rSeXJiDLhyzzK21SiTvt+pQgnLuBEQnP5xIz1k9Tu4ctfSturp7I+ZxlWNPHgKcQlw5Z5AnQrZf5/lxD0gX9shJ7XqTKlSpQgKCrLxW7lz5w5lykgu2aVLlyYgIMDms9KlS8vHWn/26NEj9Ho9xYtbJQTIpZj9MbZ7W9r6cK0lp0y2hV6bvj6VEiqXzNw5rI+74XhrWLrISOi1NUmjY3ISa4fjjDr6mrGJXMrBxHhXAy1JFzMSdm1NUudl60rYZhydQ2btfkum2P6pVLpOD14eAk2rSct3H+Wsg6lNosJMOPqayS35ZETRkkHar2D6o8pycy6Z9JYmSA7rApLHX8HEeBkSZPR6PYmJiRiNRgwGA4mJiRgMBs6cOcOTJ5J9/sGDByxatIjmzZsDktmpZMmSLF26FK1Wy4YNGxAEgZo1awLQsWNHli9fTkREBEFBQWzcuJHOnTsD0KpVK65du8axY8dISEhgwYIFtGnTJkVH39yEeVA+kacARi+Thsjquc+OZHiJWlGONKpYQgrVywzWkUu5JWw5PcUik8O68ndORy6ZC0VC+jP6JiW3RC5ltKJyclg7Y1++K+JVJQ+qPLZKYzcHCzJZjVZKinUYdk4mx8tIRfLUKFXEEvl47AokZMzNxm4EhcBzk0tMWvljrClZxJIYM6XikTmFdQ24qhnoE0Ajq8ilY1dzR5RcdpIhQWbRokU0adKEjRs3snjxYpo0acK2bdu4ceMGQ4cOpWnTpowdO5aWLVsycOBA+bipU6dy4sQJWrVqxbp165g2bRoqkzNAr169qFOnDt27d+ftt9/mzTffpH79+gD4+vry3Xff8dNPP9GmTRsiIyOZNGmSHbvvOMwStV6hIKimbZYzQS1kS5r16/cts8uMlCZISm6suWSrkUn/cdalCnK6RtG5W5brp7fGUlJs6hTlYH9OZTL1vTVJtWWCUsDHKsuvoBZwLey4SczdR6Kc5r1KqYzPipOjUyPLck6WKzhh9XLLSCK8pAiCIFfDTtDmXLhvRhLhWaNWWbJI3woiQxGcjsRoFLli6lOpIrxgvk+LGmXBzUVafhU1MhnykRk1ahSjRo1K9jNrwSUpfn5+LF68ONnPFAoFEyZMYMKECcl+3rRpUzn8+mXCOgfAvgLFGSoEyhoZ95LuKFSOtyVaJ62qWTbzszBr01Lu0chI/91c0k6EZU3FEpIjtk5vmygwJzBrZJTKjA3G1pQtJjkKJ2hz1rRkTrSmVGZeKCvoI1DQRyQkwqJd8mnow7M9UjyaWwk3hAzkpskof++2LA9sZx9/lkpWjpgHL0q+KukppGlvzBozL4+MJSpMjla1BJZskwaz/eehVDqKadqbyxnMgGtNeT9JGxOXAI9CpeSAOU3gE4gxmWYzI0CrVQL1Koocuiid61GoaBf/v5cFZ4kCB1E4H+Q35Sg4EuJG/lYWj3qPMtmVCM8y26iRfLR6uihRCDkHRm5IiqfTi9wx+U2XK54xPwaN2rZUQUJizszIEhItZr9KWTD7KZWCrIaWBufs709MvMDVQGm5emlwd838AGoexJ9FSpWafa38ZNwdqMW0rnQNkn+MPRAEgU6m5HhaXc4kkgsOEXlkyk1Qr2LW/H4AOcMvwIEcKiB5ycqfJK2q10mxLVVgpwZlEdvSBJk7h3XdpVdNK+MUZByEdS2cJ+GQt7fl6clTOeVwTntirXHIiiCjUAjyLO7u45x7+Zu599hiMsuIWcmM+bsw5GCpgiv3LH3IrAbDjPk+E0W4ei9r58oMlwM1mDX0mTUrmbE1lYF3nbx4VpKqYxbuWiiFo7LO2ZuSYAvQrDr4F7bfbNa2XEH2PzuZrXidEsULCrIwcPI6xCVm/8zfrJFRKTOuYSpXPPdVwbbJ6Fsmc9+ndSXs46+Yn4xTkHEg1irCoNIFKD2uFIVfL4T/sCzqdtOBKIpyaYKi+aGAd9YGG7MWw2jM+VmMjX9MBhx9zVS3UkXnlHnJHo6+Zqz7kxMOvxfuuMjLWXEkBdtB/PJdUKgVNN3XiJbnm1NicCZ+7HTy/XLLwD/wNfu+mFvXlgpPguTwm91+GSesCnmmtyJ5Wpi1Mjo9nL3lkvrOdkarE2Whs2IJScuaEXJj8Uhr/7bMamQa2VTCzmKDXjKcgowDsR6Ur9wVqPh1eWovrYlrYcc/+A+eSknsIOOFIpOjcsncU3Mps6HXZqy1UxcDcmYgO29l9quVRY2MbQh29vfnorUgk4XQXkjeeVmhUTjUrHThtsi/h6XlIvlgUHv7nt/DTaBFDWn5wdPsf37srZEBaF3b8twdu569UaQ3H1i0mRk1K0GSEOxcopExT0BcNZLfW2Yo4C3IfTt7S4pafVVwCjIOxCYKI5tfMNahxVmJWDJjWwU7Zx+QzIZem8kNuWTOWRWLzKqgWS0H+yOKcOGulF4gr2fmTH3WVC6JnJ04u5yXpyy13E+fDBBwy6S/Ump0tDYvZWMYtl4vctZ0r5UsLDlU24OWVn4yx69lryBjG7GU8f4ULyAJDJA7BJm4BJEAk89flVKgUmX+NzKHYWt1tmPM/ztOQcaBWA/K2f2CsY5YqpFJm6s11pFL13I4cimzoddmCvkKFPKVli/eyX5Vv8EgyvdDmWKQ1zNrv08Bb4HCpv5cupu9/Ql+BiGRUvCjPRxJ3VwEyptmlVcDpRexIzl/S2SjlTZm5OuOuU6nHKqGfeWeFJ0D9tPGgCQQmQt9XgnUyDldsgNrrWNmzDAKhUVzceeR9DzmJNcCLckKM2tWMmOdGO9VMi85BRkH4u5qeWCu3nP8oGyNdcSSPTQyZYoi1++4Hpj182UFc+h1YV8pe2pmqGHSYoQ9h8dhdmpYOrkZBPGmRGKZzeiblOpW/XmSjf2xMVtk0axkxqxhStQiz1QdhbU25tOBQqajx9KivJ9UFwvg8CWpPEV2YJ3nxV7+MWbM+WSMosChS3Y9darYI8LHPC7r9HA/Z2osytjD0deMtZ/Mq+Tw6xRkHIz5QUvIhkHZGrMTq4ebJIRkFZXKIpTdDMpeocyaiGgpzwhkzYxhbV7K7gy/1irf2uXt83Kx8S3Jxnwyp6wz+trpRZldzsvnbopsOiItF80PI7o47lqCINCxgbSsN8Des467ljUnrRx97amRAWhl5SezL5v6A5YXv5eHlBoiM1g7/Oa0eckejr5mqpSEPKbsHseu5J6Ef47GKcg4mKRRGNlBVKzIXVNZp2qlpVwj9sBsXtLppTDsnCCrEUtmcrJUwXmrjL7208jkTOSSrSBjn3MmLVXgKGy0MQMcp40xY1uuIHteMGaNmVplv3vNTIuaFtP5mn3Z41waGS0SZKojWq00mU5aWN7PclxOR2FaP69ZzSatVAo0ND2Hj8Mk5/JXAacg42CqJxOF4WisHwx7RCyZsXH4DbTfeTNCViOWzNTIQUds29Br+5wzJyKXHOVImh3O2Gdvimw+Ki0XKwDDHaiNMdOylsXJdPtJx8+WI6NFOU9SjbKZT7qYEj55BLo3k5afRsDK3anvbw/MhSIha9oL28ilnC1+ab7HC/na5xl6FcOwnYKMg8mJCJkLVi/KrJQmSIpNCHYOOfzaRCxlwbRUoYTF5yc7NTKiKMqCTNH8kuOxPajkL5UHgOy7z6wdSevbyT8GpKKEnqZoa0dpMacssdxHnznQN8Yad1dBjvZ5+AwOXXTs9U7fsCw3tLNZycz4PpblX9aKDhfOshqxZCa3ZPd9Gg6hz6XlrJqVzNhUwn5F/GScgoyD8S9ssVlmmyBjlRslKxl9k5IbQrDtZVrSqAXZVHYzKPuyFQc+tuT3saeq30UjyN/HtUCpjIOjcUR+EpCiSsyD+r3HEB1n376cuSGy5Zi0XLwADOts19OnypttLS+ZLxY69sVvr4rXqdGwCtQpJ0mzV+/BzlMOuYxMViOWzBTwlnxsIGd9ZGwcfe0kyFg73Ts1Mk7sgkJhCVMMfJI90QpmR19BsJ+UD9IsxmwTz6mkeOaIJbVKMmdkBbN5yWDIvv44wqxkxqz90+mzZ3A+ed1xjqTW9+0VO2tlrH1jPhsk4KLJvhT7/dtYNIlHLjn2xW/z+9hRY5aU4R2i5OUZqx07vllPBqtmYWwTBEuZhftPcy55nK1/jH3uQ+88AlVM75wLARAb//+vlXEKMtmAtXnJ3oNyUvR6UZbyy/tJWUXthZuLQKki0vKNB5bcB9mFwWBJHFW2WNYSR4HtwJFdpQrOWTn62itiyUx2OvxGRIvsOCktq5Uiteyo+QPH9eX0dZH/TNoYv4Lwdif7nTs9qFQC3w5zvFZGFEVZI+PrBWWLp75/VmhbO17ORrv3rG3qB3siiqLsI1O8gOSjkxXM+YqMRuTgiOzGnhFL1pgT4xkMcOam/c6bW3EKMtmAzaDsYEHmZpCUfwNsHVrthdm8FBuPHD2QXUgzJ2k5qxlkIWdKFThSI5Ncen9HIIoiI6aJcv6dZtXicbVz1Q1HRS5NXpJz2hgzPVtYcjudvQn/HrL/NQIfSxXEQdLGZDa6Jz0oFfBBb8v6z2scc+8FhSAn3stqdA/kjlIF5kmnQmGbdDSrvGqJ8ZyCTDZg6/Dr2BfmGSsHv5pZLEaYHJVt/GTsfvpUsZd/jJmccMQ2CzI+eST/KXtiG7lk33Nbs/A/WH9QWvbJA98MDrf7NRxRduHUNZFtpvIAfgVhaEf7nDejKBQC3w23PJtfLhLtnl32RDb4x1gzpIOk+QFYvReCQ+w/ztk6+mb9fDkdgq3Xi1wNNLWluH2jyhpbRS4dvug0LTmxA7YzZcdd53mMyNdWM856Fe1/jUpWkUs3HqSyowOwV+i1mYI+ltT+2VGq4EmYRYtRq5z9Z8l+BaV6R+C4++x6oMj7sy3f04KPoGg+g92v45NHoHgBafmyncouWGtjPs8hbYyZTg0t6v9rgfD3Hvue35GJ8JLD3RXe6SYt6w0wZ70DBBmbjL5Z/+1sNDI5UAX7VrBUEwnso2Gyprwf8ti264xjBMvchFOQyQbyegpyBkp7DcrJMW6WyP0n0nLzGpYU4vbEWiOT3Q6/WS0WmRxm81J4FDwKtc85U+KsVUZfe5uVQBKMzEJzUIjkx2JPEhJF+k0R5fIKo99AziPiCMyDe2QMBGfRjHnymsh2k09PiUIwNJt9Y5IiCALfj7S8jCcvEe0WaRYVa6nmDfYNjU+Nd7sLaNTS8vwt9o82s4lY+j8wLV12gKOvGUEQGGGqG2YwwB+bnIKMEztgHpSjYh2TbXHdAZG/dkrLedzhr8+FLBfwS46KVoLMjZw0LdnBRwayr1SBKIr8tNIymNSv5BhtQHUHRvt8/IcleVflkvDzu47VaNj6yWTtXEm1MRp1zmljzLSsJdC2rrR89xEs3pr1c4qiyOifRXmMaVMHfL2yp6+F8wkMbCctP4+BRf/Z9/zme0+lhIp2eP698wgU9JGWc8K0dMlOoeQpMfoNAZUpt9SfW7IvxURO4BRksglH+mM8ChUZNcNyk879QMC/sGMGLy8Pi8r/2n3IzlIe5tDr/HkhX1779K9GNkUurd4rFQsEaSb4RlPHXMdR0T7/HROZvV5adtHA6q8F3F0d+4K0V19OXLVEWPkXhrdyyDcmOaZa+cp8+5dIfBZfNst2wCqTmSqvJyz8KHsFtvF9Ldf7dZ1ot5psWp0om7IrlMBugqhZK/MoFGLsrEFKC1uNjP3PXzS/QM8W0vKzSKmMxP8rTkEmm7AuhPfXThGj0T4PjSiKvP2jSLgplUOvljCovV1OnSLmyKWIaAiNyp5bKDpOlE0/9tLGgG3kkqMcsWPiRCbNs5x71jjH+Wc4InLpcajI0B8s5/r5HYFqdlaFJ4e9IpdyozbGTP3KgizUPnwG8zZm/lw3H4iM/dXKf2mSQMki2dvXKqUsxTHvP7E4hWeVmw8k3xuw70s/JzP8miNY87jb3/HfzLielt9/zgbHZ17OKZyCTDbRsha4mUJU1x2AD+bY56b6/V9LUq0i+eCPCYJDQy3BNsPvnUdqh17LjLUN217+MWCe3UnLjjIt/bBS5OEzablLY+jY0HG/T1U7mmMAjEaRwd+Lchr1rk3gne5ZP296sC4jkVmNzPErovx8lCwsRdfkNr4dJmB+ZH9YIWbKtyRRK9J/ikhsvLQ+vAv0bpUzAtvEfpbr/rzGPuOcvUoTJKV88ZyJXHoeY/FnzErxy7RoVBVql5eWz96E4/+nodhOQSabKOQr8PdXglwPZ856+Hpx1h7wG/dFJv5uOceSTwW7mVxSw7rm0u1sEmTsHbFkRq2yLVWQVdV+Uu48FJmxWlrWqGHmWMf+Pl4elqSFl+9mvT8zVsOeM9Jy0fyw6GPHC8pmNGpBFppvPJDMCxnFRhszOHdpY8xUKyPQr420HPocZv2T8XN8+qelhlfFEvDreznXz1a1LXlyTt+Aw3aoKWWv0gRJySmHX+vil44wK5kRBOEFrYy9Mdg/aDHDOAWZbKRbM4HFH1vZxJfBzLWZu7G0OpGB34kkmBLEje0B7etnz+CVExoZR0QsmTE7yBqN9o/EGj9XlEMsx/eBssUd/xuZB8aYeKg6RGTHyczdY6euiXy+QDpWEGDFFwL5vbP3BWl+aekNGQ/3P3ZZZNdpaTm3amPMTBlqmeTMWCNmKOJs23GRmWulZRcNrJ4s2DWjd0YRBIEJfW21MlnF3jlkzNialrLP7HLJzqHkqdG3teRXCJI14FGoffs5aCp8vCifnIAxJ3AKMtnM4A4Cs8ZZbtzxc0UWb834jfXtMpGzptTTFUvAT6Ozb+CyzkCZMxoZ+567hlWFcHual3acFNl8VFouml/yz8gORr4uyCaZu4+g4ySRvl8beZyBASw6TuTNb0XZL+HTgdCqdva/HLPi8GutjflisIA6iyUtHEk5P0FO0Pc8BqavSt9v9ThUZMj3ln1njBFs7uecom9rKGYKCth8VPLfyQrm397LAzmVhT0oU8yyfCsbTUvW/muO1MiAlGhvVFdpWW/nUOy9Z0XW7IN/DnnS/D3s5vuZUZyCTA4wrpfA5KGWwWbEdJH1B9J/Axy7LPL9CmlZpYQVXzo+gsSa/N6CLOFnn0ZG+q9UQumi9j23tcPvku2iXQrIaXUiH8yxnGfaaAFP9+z5jTo1Eri4WKB5Dcu2tfuh4iCR3zakL4vsuzNF7pjqWjWojM39mp1Yz77f+1Wk/kgjPb8w8sFsIz+vFlm7T+T4FZHgENt+Hb0ssttkEitdFAbnYm2MmS+HWPKwzFonJVBMDaNRZNBUi//S643h3R4ObmQ6UasE3u9luWcyq3kGiIwW5XIoVUvZ15/E3VXAr6C0nJ2mJUdpmFJi9BsWjd/8zfYpkqnViYydaTnPR2/ikJQf6cEpyOQQX70F7/eSlo1GePNbkd2n0765ouOkwctcsHHyUIE6FbL/5jFrZUIiVfxzwLHXMhpFeZApXcR+oZdmGlWxZME8cgkGT816VNnsdRYtUuOq8Ga7LDYyg1QqKXBgtmDym5K2RcXC2F9FGr8jcv5Wyv1bsUtkuVVOor+/yjltRq1ylorrkTGSz8WGQ9KLfuLvIn0nS/3x6yXi0lakRC8jTd4xMnjqy6ONMVOikMBo08w5LkFy/E2N6aukIo0gafwWf5J9/kvpYeTr0v0DUlj4s8iMP1PxibYTAkdoL8zmpfAoCHvueI2CKFoK+5YoJCVMdTTFCwr0bC4th0RIE5us8us/FnNvrTKJDHFwtGxqOAWZHEIQBH4ZK8h2e60Oun0uzS5TY/xcUa7U2rgqfPymgxuaAoPbWx6+t390XMVbkMJS4xKkZXublUCalW36XsDdVVpfu1/6njMbbfE4VGTKUotvyZz3c+YFIwgCb3UUuLFcsKnyfOo61B0pMn6u8YXcGXceioz52bJt/kSB0kVz7uVYOJ/A/IkCDSpLUXmpfY0Gg5TR+NgVSzXj0kVh4GvZ01Z78Nkgy334x2Z48DT5e/DkNZEvFuas/1Ja5PUUGN5FWk7QShGWGeHeI5Em74gs22HZ1rul/fto7fCbHZFLD55KkwpwvFnJmvesnH5nr89aNFlwiMg3y6TjFQqYMjhcnnDkBE5BJgdRKAQWfiTQzZTmPS4BOn0kppj/Y/MRkYWmbJmebrD8CwFVDs003+5syVcTlyAJYZmZcaUHaydPezv6mqlfWeCfKRb166x1yNFGGeXTP0ViTGGwI7pA7RzQmFmT31tg0ScKDs62RAEZjTBzLVQaJLLxsPS7aXVSCK+57UM6QP+2Of9yHN5F4MQfCh79qyBhj8C9NQKH5gis/FLgx1EC73aXEgzWLg8FvC3HCYKUffhl0MaYKeQryJparU7yhUvK8xjpdzL7L32WQ/5L6eH9XpZn6rd/05/wb/sJkTojLJFY7q6SZrB1Hfv307p4ZHaYl2wdfR1/PTNNqllKo5y5ASevpb5/akz4zRLqP7orVC2pzXoDs4BTkMlhVCqBVV8Jcl2kyBh4bYJIQBIP+qfhIsOn2SZVy8mZsiAI/DEeapSWCu/cfwK9vrRfvRhrHBV6nZROjQQWTLKc/6N5Iit2Zaw/J65aZpDenvDdiNzzgmleU+DCYoGpIwRcNdK24GfQ/XORNz41Mm6WyGlT9fSyxWDOB7mn7WY0ainJW7MaAm+2E/h4gMDcDxVs/F7B2YUKQjYriN8tcPtvgeD1At2a5b4+pMWk/oJc/HPJdrhtFbFnLkFw77G03qgKfJ1D/kvpwb+wQO+W0vKzSGSTZUoYjSJTloh0/lgkIlraVq44nPxDcJhQnd2RS5dsMvpm328nCIKNViazhT33nBFl01QBb/hmmB0al0WcgkwuwNVFYOP3glzc7Wk4tB0v8vCZdKOJoiTEmMPbujXL+aJ3AK4u8Me4ZxTJJ60fuohNZWR74cjQ66QM7STw7TDLwz70h/T5LoE0CL83y7LvN8MECuQydb9GLfDZIIErywTa17ds33xUcgIEKQnd6skCebLJOdneuLoIlC0uUDT/y9l+nzwCE03hywYDNhXtl+2Qyl2AVIIgJ/2X0ot1KPYva1P2PwuPEnn9E5HJS0S59Em3ZnD6T4GqDgxRzs5cMgaDyJ6z2RexlJR+bZB95tbuJ0ORjGBy8LXKHj1tjIBPHnu2MHM4BZlcQh53ge3TBaqUktbvP4F240VCI0UWbIH/jknbC/nCn5Nyj1NfIR8D676xZMedtxHm27nSqiNDr5Pj88FSZWeQwhV7fJG6c6yZpdsllS1I0RVj3nBgI7NImWLS/bZmsiA7Opv5fkTOOJA7sfB+b4uZbPVeuHxHzBUlCDJD3YoCLWpKyzcfwLYTL+5z/pZI3RGi/JlCAT+OEtjwneBwZ9hSRZDNX470kRFFkfdnixw4L60X9LEVorIDNxeBkaaq2HoDzN+csbF65lrbIIbBOejga41TkMlF+HoJ7PpZkMOLr9+XNDMfzrXcbIs+zn2z/IZV4M+JljaN/VXk8EX7CTPm0Ou8nsjVah2JIAjM/cBSAycmHjp+JHLvUcp9iowW+WS+5fPZ7+ec/1J6EQSBPq0FbqwQGNsDfPJIqe3H983pljnJ4y7w6UDp/hFF+GR+7ilBkBmstTIzVts+R0u3S5FnZnNZ/ryw62fJbJgdEza1SqCUqdbRrWAcVo9o6l/wm8nhWamUosxyQps2xioU+4/N6c+YHfTU1sH3tw+FHAu3TopTkMllFM0vsPsXQTbXXAywROyM6gqdG+WOGycpQzoKfNBbWtYboOeXIvefZH1AiEsQefBUWq7g57iaJElRKgVWfS3QuKq0/jQc2k+UNGTJMWWpxfTXu1Xudb5MjryeAnM+UBD2n8CCjxS5ZnB61RnzhiWp3LYT5JoSBJmhcyOLNvXgBThzQ8rXNHqGkaE/WDKU168E5xYKtHGAU29qmP1kYuPhcZj9z//nZpEvF1nGjsUfCzk2lvsVEuhuCjB5Gg7/pDMUe/xvovwuerc71CyXe+5BpyCTCyldVNLMWNseyxWXIjByM9PHCLSrKy0/i5QimeISsibMWKt6s8OsZI2bi8CWHwUqlrC0pcsnIrHxtn26FghzNpiPkb6Hl5HcYq50IuHqIvDlYNvfJDeUIMgMCoXA+D6WNn+5SKTZWFH2ywIY0w0OzRHwK5T9fXNkFewNB0XG/GIZM6aPERjcIWd/v/d62IZip8WuUyLrDkjLBX3gm7dz1/3nFGRyKVVLC+yYLlDAW4p++fur3D94qVQCqycLctrvC7fh7R+zlq/Axj/GL/v77+slsGOGQNH80vrJa9Bvioheb3bEhg9mWwqnfTJAwL9w7v6dnLw8vN3ZNpN1bilBkBkGtbf4/ew4iRwh56qBpZ8K/D5egYsmZ/pWzqoGmj0dfg9ekEp9mBOYTugLE/vn/O/XrIYlo/mp61JeopRI1NoGMUwbLeCdJ+f7YI1TkMnF1K8s8OAfgcf/CtStmLtunJTw9ZKSy3m6Setr9sEPKzJ/vptWg0p2a2TM+BeWHGO9PKT1/47B6J+lyIpdZ93Ye868H0zqnzNtdPL/iVolsPxzgZrlpHsrt5QgyAxuLgLvdLPdVqoIHJ8nMKRjzo5v1hqZW0H28ZG5cFuk66ciiSaz2aD2UpRPbkAQBButTGqh2L+stQh3TapZ8oflJpyCTC7H1UXA1SV33PzppUopKVmZmS8Wimw5mrnBwbrYnKNDr1OjehmBjVMttXAWbYXP/oSpqywhP7+8K+D2kv1WTnI/jasJnF+kYNoYxUtv/nu3hyBrZTo3grMLhVzha2Hv7L53H4l0mCTKGXw7NZQCNXKT/9mb7cDXS1peuz/52l4Pnop891fudPC1xinIOHEIXZta8rGIIgz4VuRaYPrzsVy9J7LwP5FDF6VtggBlszlUMSmtagv89ZnlIZ62CoJDpTLTbepA9+Y51TInTl4OCnhLSRmP/iaw+QcBn1xiovArKPkfQdZNS0/DRV6bIPI0XFpvWAXWTsl9+X7cXARGmEpI6PTw55YX9xk/1+LgO7Y7udas6RRknDiMzwdLETwA0XHwxqciEdEvCjNRsVLSuW+WinSYaMS3i0jVISIjplmq3pYsTK7QdvRtIzBzrG07lAop0/LLPlt24iQ7KJpfoHG13DWzVygEypl8++48Il0V4pMjKlak4yRL5fhK/vDfj7nXv3FMN0GukTRvo2gTir3zlMj6g9JyIV+YksscfK1R5XQDnPz/IggCSz6RbM4XAyDgIfSbLDLnAzhxFY5dETl+VSppn5o/sIsGPh+Uex6iD/oIBD8T+XmNtP5ud8mc5sSJk5eXcsXhyj2pxtXU5dC7pUhF//RH8yVqpXIf5jD54gVg5wyBfHlz79jgX1igW1ORDYfgSTisPwj925ocfH+1jbTKbQ6+1jgFGScOxcNNYNP3UHeESOhz2HUaKgxIfbZTyFeqIdO4qkCjKlCnQu7QxlgzbYxAycIid4MimTrKO6eb48SJkyxSqST8e1ha/nqxyNeLJWfkTg1FOjcSaFWLFP0VDUYY/D3sMzn++3pJSf1yIpQ8o7zXU2DDIWlMnr1epH9bgZ/XWHyFmlbP/RXknYKME4fjX1hg3bfQ9kNLxV4zCgXUKGMluFSVBo/cbqZRKATe6S5y//5zXDTeOd0cJ06cZJHhnQX+OybaFHW891jKxvvbvyJuLtCmjiTUdG6ILKSIIkxZ4SvnWXFzkcxJlUrm7jHMTIuaUhXuy3clTfn6AxYHX6VScvDN7eOxU5Bxki20qCmw4guYulzEr6AktDSuCvUqgudLWpzQiRMn/z+UKipwcYnArSCRrcdh63Ep2ECnlz6PT5RSL/x3THrJVyst0rmRtH3FXil7qVIJ674RaFT15RnTpKrYMHK61K9+UywTzrHds7dCd2ZxCjJOso2+bQT6tsn9D4UTJ05eXcr7CZT3gw/7CETFiuw5Iwk1205IfiRmLt+V/qxZ8olAp1xaRiY1BrSDj/+AiGhkISa3O/ha4xRknDhx4sSJk2Tw8hDo0QJ6tBAwGiVHXrO25vQN2yCF6WNgUPuX48WfFHdXgeFdRKavsmyb8Y7jK4/biwyFX69bt44BAwbQoEED5s+fb/PZli1b6NSpEy1atGDKlCnodDr5s+DgYN5++22aNGnCgAEDuHXrlvyZ0Wjk559/pmXLlrz22musXLnS5rxHjx6lW7duNG3alPHjxxMVFZWZfjpx4sSJEyeZRqEQqFNB4Ku3BE7OV/Bko8DSTwWGdYYfhoa99FXj3+kmoDapNppVl7Q0LwsZEmTy58/PyJEjad26tc32gIAAfvnlF6ZPn87WrVt5+vQpCxculD//7LPPaNCgAfv27aN79+5MmjQJvV4yPK5fv56zZ8+yYcMGFi5cyIoVKzh16hQA4eHhfP7550ycOJE9e/aQJ08epk+fntU+O3HixIkTJ1mioI9UWuHPSdC3ZUxONyfLlCwi8N+PAp8Ngn+n5n4HX2syZFpq2bIlIGlJrNmxYwetW7emSpUqALz99ttMnjyZMWPGEBgYyL1791i4cCEajYZevXqxbNkyLly4QN26ddm2bRsDBw7E19cXX19funXrxtatW6lfvz779++ncuXKNG3aFICRI0fSu3dvPv/8c1xdXZNto1arRavV2nZSpUKj0WSkq2liNFUBM/9/FXnVv4NXvf/g/A6c/X+1+w//X99B27rSH4gYjenNxO7Y/isUaetb7OIjc/fuXerXry+vly1blidPnhAXF8e9e/coUaKEjSBRtmxZ7ty5Q926dbl79y7lypWz+ezIkSMA3Lt3j7Jly8qfFStWDJVKRXBwsM12a5YsWcKCBQtstvXu3Zs+ffrYo6svEBRkx1KpLymv+nfwqvcfnN+Bs/+vdv/B+R04qv+lSpVKcx+7CDLx8fF4eHjI656engDExcURFxdn8xmAh4cH8fHxyR7r4eFBXFycfHyhQoVSPDY5hg4dyoABA2y2OUojExQUhJ+fX7okxv9HXvXv4FXvPzi/A2f/X+3+g/M7yA39t4sg4+bmRmxsrLweEyPZC93d3XF3d7f5DCA2NhY3N7dkj42NjcXd3V0+PrVjk0Oj0dhdaEkNhULxSt681rzq38Gr3n9wfgfO/r/a/Qfnd5CT/bfLVUuXLk1AQIC8fufOHQoXLoy7uzulSpUiKCjIxm/lzp07lClTJsVjS5cuDUgqJevPHj16hF6vp3jxHC6D7MSJEydOnDjJFWRIkNHr9SQmJmI0GjEYDCQmJmIwGOjQoQP79u3j+vXrxMTEsHjxYjp37gxAyZIlKVmyJEuXLkWr1bJhwwYEQaBmzZoAdOzYkeXLlxMREUFQUBAbN26Uj23VqhXXrl3j2LFjJCQksGDBAtq0aZOio68TJ06cOHHi5NUiQ6alRYsW2TjSLl68mK+//prXX3+dDz/8kPHjxxMbG0vr1q0ZNmyYvN/UqVP5+uuvWbZsGf7+/kybNg2VSrp0r169CAoKonv37qjVaoYMGSI7Dvv6+vLdd9/x008/ERoaSv369ZkyZYo9+u3EiRMnTpw4+T9AEEUxfTFWTmwwGo3cv38ff3//V9Yu+qp/B696/8H5HTj7/2r3H5zfQW7o/6v3rTtx4sSJEydO/m9wCjJOnDhx4sSJk5cWpyDjxIkTJ06cOHlpcQoyTpw4ceLEiZOXFqcg48SJEydOnDh5aXEKMk6cOHHixImTlxanIPN/RoIugV03d3Ph4YWcbooTJy9w+9lt/j63ijhtXE43xUkOkaBL4HnC85xuhpP/I+xSa8lJzmI0Gjlx/wSrz6/h38v/8jwhCoBpr//E6Majcrh1TpxAoj6RafumM/Pgr+iNepafWc62EVsRBCGnm+YkG4lOjKbD/I5cfnyF1yq8xidtPqauX52cbpaTlxynIPMSc/tZAKvPr2bthbXcj3jwwucfbfmY5/HP+aj1JOcLw0mOcerBad5dP5abITflbUfvHWPV+dW8Wbt/DrbMSXYz/9ifXH58BYBdN3ex6+Yu2pVvxydtPqZeibo53DonLytOQeYlIyw2jPWXNrDq3GrOBp994XMPjQd1itfm0N3DAEzd8z2R8ZF833mqU5hxkq3EamP5dtd3zDv2B+YE4kqFEoPRAMCX27+iU6WOeLt552ArnWQXzxOeM/vwnBe27761m923dtO2fFs+afMx9UvUy4HWOXmZcfrIvAQk6BLYeHkT/f7qT7nvKzBx8yQbIUYhKGhdrjV/9plPwOe3+G/EFqZ2+lb+/Lejv/Pu+rHoDfqcaL6TV5CDdw7SaFYTfj86TxZiahWrxZH3DtG1yusAPIt5xvd7fsjJZjrJRn4/Mo/I+EgA+tbsw6zuv+LnXVz+fM+tPbSd147ui3ty8v6pHGqlk5cRp0YmF6M36Jm653sWnVhEZDLOcdWKVKVfrb70rtGbwl6FbT57r9l7eLnmZdy/7yOKIivOriQqIYpF/RbionLJri44ecWIjI/ky+1fsez0X/I2V5UrX7T7nHeajEGlVPFDl+/ZfWsP8bp4/jy+gMF1B1G1SNUcbHXu5nnCcxafXEI+d1/alG9DsbzFcrpJGSYiPpLfjvwOgEqh4rN2n1HKtyQDar/J3+dWMWP/DB5EBgGw9/Ze9t7eS5tybfikzcc08K+fk0138hLgFGRyMb8fncfPB36x2VY4T2H61OxNv1p90xz8h9QbjJerF8PXjEBn0LH56hb6LOvH34NW4KHxcGTTnbyCbLu2jQ83TeBx1GN5W5NSjZnTYw5l85eRt/l5+zGx1QS+3fUdRtHIhM2T2DFym9P0mQyiKDJoxWAO3Dkob6tUqBJtyrWmbfm2NC7ZCFe1a5auEaeN4/Ljy5wLPs+T6CcMqjvI5veyB3MOzyUqUQpCGFhnAKV8SwKgUWl4q/4Q3qzdn1XnVzNj/wzZ388s0LQu15pP23xMA/8Gdm2Tk/8fnNWvM4mjK36GxoRSc0ZtohKjEASBvjX70K9WP1qUaY5SoczQuXbf3MPAlYOI18UDUL9Eff55ay0+WfRNyA1VT5MiiiJf75jM7lu7+bLdF3Sq3Mlh18qN/c9ujEYj52+cZ+7F31h/aYO83VPjyTcdp/B2/aHJfjeJ+kQa/NqIu2F3AZjf+w/61+6Xbe22F46+B1aeXcmYde+m+Lmb2o2mpZrSpnxr2pZrQ7kC5VIVCBP1iVx9co3zD89zLvgc54PPcz3khuy3BODnXZwTHxwnj0ueNNuXnv6HxYZRbVoNYrQxqJVqzk84SwmfEsnuqzPoWHVuFdOtBBozrcu1Zvrr0yhXoGya7cpOXvVxwGg0cvnWFUr5l8TLzStH2uAUZDKJo2/eCZsmseDEAgAG1x3E3J4vOsllhOOBx+mzrK8cml21cBU2vv0vBfMUzPQ5M/MdiKLo0Jn3rEOz+XL7V4DkWLqk32K6VXvDIdd61QcwURT558I6Jm2eRERCpLy9bfm2zOo+Ez9vv1SP331zDz2X9gKgoGdBzk44TV7XvI5sst1x5D3wLOYZdX+pT0R8BABv1x/K5cdXOBt8FqNoTPaYEt5+tCnfhrbl29CkVFMePX8oCSwPL3Au+DxXn1xFa9Cmee3hDYfzyxsz0twvPf3/esdkZh78NUPnNQs0M/b/TGDEfXm72Uz5btN3MjyhcxSv4jgQFBnE8cATHA88wYn7J7j25BpzesxmcL1BOdIepyCTSRx5894MuUnDWY0xGA14aDw4P+HsCz4wmeHSo0t0W9yD0NhQAErnK83mYRtTnB2lRUa+g2tPrvH9nh/YeXMX/Wv1Y2a3X+w+EO29tY+eS3vZDPJKhZLF/RbRvVo3u14LXs0BzExkfCTvbRjHpiub5W0+bj782OUH+tXqm25h9c3lA/jv2lYA3mkyhh+7vFzOv468B4atHsE/F/8BoFeNXizutxCA8LgIDgQcYM+tPey9vc/GlJdRlAollQpWpFbxWlQuVJlvd31HnE5KVrh95FaalGqS6vFp9T8kOoTq02sSp4vDReXChYnnMuTjozPoWH1+NdP2TbfR0NQpXod5vX6jYqGK6T6Xo/h/HwcMRgPXnl4zCS0nORF4guDnwS/sN7DOAH7v9VsOtNDpI5Mr+WLbV7Kq98MWH9hFiAGoXrQ6O0ft4I1F3Qh+HszdsLu89kcHNg37lwoFK9jlGkkJCL3DD3t+ZN2ldXL0ytLTy1Ap1fzcdbrdtDN3w+4xdPXbshBTsWBFbphU5m+vHoYoivSo3t0u13rVORt0jqGrhtrMlN+o2pWfu87IsIbvhy7fs/f2PuJ18cw//ieD6g6kSuEq9m7yS8eum7tlIcbbzZufrAQ8X3cfelTvTo/q3RFFketPr7PnluRPcvTesRQ1LoIgUC5/OWoXr0WtYrWoXbwW1YpUw13jbrPPJ/99CsDY9eM49v4R3NRume7Hr4dmyYLR0PpvZdhRWa1UM6juIHpW78m3u77j92NSFNzZ4LM0ndOcT9p8zAfN30eldL7K7EWcNo6zwWdlweXU/VOyf1NyKAQF5X3LUzpf6WxspS1OjUwmcZQUvu/2frotll64xfIW4+z40zYDjT0IjgzmjcXduf3sNgD5PPKxYeh6ahWrmaHzpPYdPIh4wLR901l57m8b+7s1n7f9jI/bfJSpPlgTq42l7bx2XH1yDYBOlTqyYsByPtj4IX+dWQ5IM8+FfRfQs3qPLF/PzP/7TCwpoijy+9F5fLXja3QGHSC9ZL9s8jnDWg3L9Hcwbd90vts9FYDGJRux/SVx/I3TxnHozmEM0Xo61u1ot3sgVhtLg5kN5Sie33vOZWDdgek+9sjdo+y9vZfzDy9QLG8xapuElupFq+PlmroPg8FooMOfnTh5/yQA7zcfx7cdv0lx/9SegSdRT6g+vSYJ+gRcVa5cmnQhy5Oyk/dP8s76sfLYBVCzaA1+7/VbjkW+/b+MA/fCA/lq+1dsvbYNvTHlVB3uanfqlahLQ/+GNCrZiDrFaxP+JDxH++8UZDKJI25eg9FA0znN5Bfyn33m069WX7ucOynPYp7RY0lPLj66BEAelzws6DOfNuXbpDs8O7nv4HHUY2bs/4Wlp5fKLzuQhKXxLT4gr2texm4YJ2+f1f1XhtZ/K9P9EEWRoaveZsPlfwEoV6Ac+9/Zi5erF0ajkfc3fiCHAttbmMnoPfAk6glfbv8KlVLFdx2/I5+Hr13akR1ExEfyzrp32Hptm7ytnl89FvVdiBhlzNJzkKBLoMGvjbgXfg+ABX3+pG+tPnZpt71J1Cey99Ze1l3awPbr24nVxgIwr9fvDKjzpl2u8dnWz5l7RFLRNy/djC3DN2erYHcr5BZN5jQjUZ+IQlCwd8we6vjVTnbf1J6BSZs/Yv7xPwF4r9lYpnb6zi7ti9fF88OeH5l9eI6sgVUr1UxsOYEJLcejUWnscp308rILMrHaWH4+MJM5h+eQqE984fOCngVpVLKhLLhUK1IVtVItf54b+u8UZDKJI368Zaf/4j3TS75WsVrsf2evQ2+M5wnP6bOsH8cDj8vbXFQu1CpWi4b+DWjgX58GJRqQ3zN/ssdbfwcR8RHMPPgrfx5fQII+Qd4nr6sX7zV7jzFNRstREHMPz+WzbV8Aklpy+YBlvG5KkpZRZh78la93TAbAy8WL/e/upVyBcjZttBZmFIKChX0X0KtGz0xdz5qM3ANPo5/SecHr3Hp2C5AixzYP22h3bZsjOP3gDENXDZU1BADjmr3H1+2/Qiko7fIc7Lyxi97LJOGlUJ5CnB1/Ok3tQXahM+g4EHCQ9ZfWs/XaVtlh3hqlQsmqQX/ToWL7LF3rXPB5Wv/eBqNoxEXlwvH3j9k9FDo9/HzgF6bslDQxlQtV5tDYA8kKCCk9Aw+fP6TG9FpoDVrc1e5c/ugiBTwL2LWNZ4LO8s66d7kRckPeVrVwFX7v9Rs1M6hdzgq54UWeGURRZP2lDXyx7UseRT2Stxf0LEjHSh1kwaWUb8lUBenc0H+nIJNJ7P3jRSdGU2tGHUJiQgDYMXIbjUs1zvJ50yJOG8eglUPYfWt3ivuUzV+WBv71aejfkIb+DSiXvxwKhUIOu9t0fxN/HJtPjDZGPsZD48GYxqN5r/l7yYZ5f7n9K2Ydmg1IwtPGtzek6ViYlN0399BrWW/Z92bN4FV0rNTxhf2MRiMfbPyQpaeXAZIws6DPn/Su2StD10vuvOm5B57FPKPzgtdtBlywmMByq31fFEXmHpnL1zumyKpmHzcf5vf5Q35h2/M56L/8TVnjM7bpu3zfeWrWOpAFDEYDR+4dYcOlf9l0ZTPhceEv7OPt5k2FAuU5+UDKQuumdmPzsI2ZzneiN+hp+VsrLj2+DMDX7b9iQsvxme9EFtAZdLT+vY2ssf2kzcd81vbTF/ZL6ff/cOMEFp1cJC23+IApHSY7pJ2J+kR+2jedmQdnyiZspULJhy0+5OPWkxye/DMg9A6HAg6iSlTRv0l/1Cp12gflAi49usRHWz7mmNUkVq1UM7bpu0xsNSFdofdmnILMS4y9f7xvdn3HjP1SWOIbVbuyfMBfaRxhP3QGHesurufgnYOcuH9Szu2REj5uPtQvUR9/X39WnV1FtDZa/sxF5cKIhsP5sMUHqc7AjEYjY9a9w6rzqwFJc7N95LZ027nvhN6l1W+t5IzHn7X9lE/afJzq9T7cNJ4lp5YCkjDzZ5/59KnZO13XS+mcad0DoTGhdFnYlWtPJXNh8bzFeZ7wnOhE6TsbWv8tfu02M9f5hITFhjNm3Rh23Ngpb2vo35DF/RZS3CqtvD2fg/sR96n3SwMS9AkoFUqOvneYyoUrZ/p81qZNAen7NX/PL6wLAkajkZMPTrL+0gY2Xt4kTyqsyeOShy6VO9Ojeg9alW2JgED/JQPYdXcXIAk3O0dtp1KhShlur3XqgCqFK3No7EEbFX52c+nRJVr+1hq9UY9KoeLwewdfcMRO7vd/EPGAWj/XQWfQ4anx5PJHlxxuRr3w8ALvrHuXK0+uytsqFarEr91+oaF/Q7s+X8GRwWy49C/rLq7jwqOL8vYSPiUY2XAEg+oNynKOLkcRFhvGd7unsuTUUpvozvYV2vNDl+8zpf1zCjIvMfb88YIig6jzcz0S9AmolWpOf3iK0vlK2amlGSckOoRTD05JoXb3T3Lh4YU0c0+olWoG1x3MpFYTKJq3aLquozPo6PfXm7I2qHCewuwesxN/H/9Uj4tJjKHNvHZcf3odgC6VO7NiwPI0fwej0cj4TRNYfGoJIAkz83v/kWl/jLTugbDYMF5f2FUeXIvnLc7Wkf9xP/w+PZf2kl+0n7b5hE/bfpKpNjiCk/dPMnTVMJsQy/EtPuTzdp+98GK19yD2496f5PpLTUs1YeuI/zL0EtLqtay5sIY5h397QQOWWdzV7nSo1IFe1XvQtnxbm0y6RqOR23dvM/HARxw0Zd8t6lWU3WN2pplHx5p74YE0/LUR8bp4BEFgz+jduaIatPUEq1axWuwds9tGg5jc7//ehnGyKXdSq4l8+doX2dJWrV7Lzwd+Yfr+GTbOqvk98tO0VBOalm5Ks9JNqViwYoYFm9CYUDZe2cS6i+tstBjJ4aZ2o2/NPoxqPDLXRODpDXoWn1rCd7unyvWuAMrkK8MPXb7PkknUKci8xNjzxxu+ZiRrL6wF7OsUZy8SdAlceHjBJNic4OSDU4TFhgGSMNC/Vj8+bvMRJU1pxzNCrDaW1xe+wZmgM4Bkxto1akeKfjmiKDL47yFy/pIKBSuwd8zudPtTGI1GJmyeyKKTi+X2Z1aYSe0eCI+LoOvCrrKZoKhXUbaN3CoLqP9cWMewNcPl/Wd3n8Vb9YdkuA32xGg0MvvwHKbs+kZW0+fzyMefvefTrkLbFI+x5yCWoEug/q8NCQwPBGBR34XpMgHGJMaw9NRS5h753cben1lcVC68VqEdPar3oEPF9imW9DD337ewL10XvcH5hxcAyel816gd5PPIl+a1RFGk2+Ie7A/YD8DoxqOY9vpPWe6DPUjUJ9J0TnNuhtwE4NuO3/B+c4uzftLf/27YPer8UheD0UBeVy8ufXQp27UTVx5f4Z1179poS6zJ75GfZqWb0rSUJNhUKFghWcEmKiGK/65tZd3F9ewP2J9s9GXNojXoXLkzB24e4GjQsRc+b1a6KaMajaRTpU6ZNiEn6hO5GXKTe+H3cFd74Ovui6+HL77uPni5eKUplB2+e5iPtnwsB5GAlHn7o9aTGNNkdJbNb05B5iXGXj/emaCztP69DQC+7r5cmHgO71yqljQjiiIBoXe48fQG3oa8NKnWJEvfQVhsOO3nd5AdYWsXr81/wzfj6eL5wr4z9v/MN7ukyt55Xb3Y986+F1KWB8fGcyEimvZF86NOpl2SMDNJtuErBAV/9J6X4QixlO6BiPhIui7sKvsXFM5TmG0jt76gtk3q9Pz3wBV2L6lgMBqI08URr40nVhdHvDaOWG0ccbpY4rTxps+kbTtv7mLPrT3ysY1LNmJxv0WpatgcMYjtuLGTPsuk36JwnsKcGX8qRUE1LDaMP47NZ/7xP21mmiA5fuZx9cJ6iBOxWjZtt/68UJ6CvF7ldTpX7pQu4di6/+Fx4bT7oz13wu4Aqd/H1qw+v4aRa0cBktbu5IfpKw9gb86HP2fN/ccMLVOcCl6WNp96cJp2f7yGKIq4qlw5Ou6I/Mwl/f1H/zOGv8+tAtI29zoSnUHH8jMr2HlzJ8fuHUvWQduMtWDTuFQjAkIDWHdxPTtv7LIJXDBTvkB5etXoSc/qPSlXoKz8HWjdtSw8uYi/z62STcdm/LyLM7zhcAbXHZyimU0URUJiQrj8+ApXn1zl8uMrXHl8hVvPbqUYDq1SqPBx98HX3Zd87r6SkGP1d/7hef69vNHmmH61+jKlw2SKeBVJ41tMH05B5iUmNCaU0MehlC9TPtM/niiKtJ/fkRP3TwAw/fVpjGo80p7NdCj2Nq+1m9denk23LteatYNX20RK7Lyxiz5/9ZXLHKwZvPoFlWiUTk+jHcd4HJ/ImyWLMrd+8qpdURQZv2liloSZ5PofGR/JG4u6yTPzQnkKsW3EfzaRVNZ8vu0L5hyeC5idRTdlqdqvzqBj4YlFzDv2B0+iniQ7EKeFIAhMaDmBz9p8kuYs0lGDWN+/+rH9+g4geS1lcGQwcw7PZdnpv+SEa2Y6V+7Ehy0+pH6JenZrT0ok7f/9iPu0m9eeJ9FPgOTvY2vCYsOoO7O+rOFMyWHd0URpddTadpQIrY4ibi6c6NCYPGrLb//pf5/x21GpenWjko3YPmKr7PBv7v+dsLvUm1kfo2jE282byx9dzBUlJwxGA5cfX+bI3SMcunuYY/eOp5rgLTn8vIvTs3pPetXoSbUi1Wy0IEnvgaiEKFadW82fJxbY5LsBqcRC75q9GN5wOCqFUhZarjy+wuXHV+Ss646gVrGaTHt9mt2riTsFmZeYASsGcin4MrN7zqJVuZaZOsemK5sYtFIyJ5QrUI4T7x/LUee+jGLvG/jak2t0mN9RduDtXaM3C/rMR6FQEBB6h1a/tZJnVl++9gWTWk184Ryzrt9jyuUAAATgWIdGNrNLa5IKM4Ig0LN6T4Y1GErjko3TVNkm7f/zhOd0W9SDs8FnASmMceuILalmTTYajYxYO0rO4urj5sPu0TspX7B8qtdOjr239vHJ1k9lM0BmyO+Rn4V9F9C6XKt07e+oQexeeCD1ZzYgUZ+ISqHi2LgjVCxUkVsht5h56FfWnF9rM0tVKVT0rtGLD1t8kK1p65Pr/9UnV+k4v5N8H/eq0YuFff5M9vsZuXYUq8+vAaB7tW4se3NptrXdmu8uB/DL9Xvy+nsV/JlSw3IPxmnjaDirsWzym9F1OiMbjbDp/8h/Rssm8q9e+5KJrSZkax/Si8Fo4NLjyxy5e5jDd4+kKNjk98hP92rd6V2jJ/VL1E/x/k7pGTAajewL2M/8Y3+y69YuMvOqVSlUVChYnqqFq1KhYAUS9YmEx0UQHhdOeFyY6X844XERck6j5Poxuf1XDKwz0CGChlOQeUnZfn0Hff+yVOrtW7MP33eemqE8CYn6ROrNbCAPDGsGr6ZjpQ72bqpDccQNfDzwOG8s6i5rEt5pMobP231Gm9/byc6bXau8zvIBf70gaMTrDdTceoRniRbH5F4lCvNnw2opXk8URSZsnsTCEwtttlcqVIm36w+lX+2+Kc4qrfsfo42h++KenA46DUiDx7YR/6XrparVa+m9rA/7Aw4A0uxvz5jd6Vb93gm9y+fbPmfb9e0226sUroyHxhMPjTvuGnfc1O54aNxxU7vhofHATe2Gu8Ydd7X0uZdLHpqUbpKhWbQjB7Ef9vzID3t/BKS8O4XyFOS/a1ttXghuajcG1x3Ee83GZrpmWFZIqf9J7+PRjUfxU5cfbe5Z6yze3q55OT3+FIXyFMreDgBP4hOpu+0IcQZLFItKEDj0WkMq5rVMAg7dOUSXhV0BycfixAfHKJ63OPfv3yfeLZ5Gs5sgiiL5PPJxadKFHDGPZQa9QW8SbI5w6sEpfNx96Fa1Gy3KNE+XX0t6noE7oXdZeGIhy8+sSFEblN8jP9WKVKVqkapULVyVqkWqUKFAhXQn+JOEnHCbP1GElmVbONRdwSnIvKRce3KNcf++z6kHp+Vt3q55mdJxCkPqDk7Xjznn8Bw+3/YlAC3KNGfzsE25LgQ3LRx1A2+9tpUBKwbJ4YFl8pWR/Q4qFqzI3nd2JztILrj9gI/P22ojFAIca9+Y8l7JO2uCJMzMPPgrsw/PeSFfiLvand41ezGswdsvJNmydvTstayPnNY9n0c+tg7fkqHQ4aiEKDr92Vl2Dq5auArbR21LVaiIToxm+r4Z/Hb0d5tQ47p+dZn2+k/U9auT7utnFkcOYvG6eBrMbGhT08mMt2teRjYeyehGo1J0DM8OUuv/9uvbeXPFQNlJ1FpLkVTDMafHbIbUG5ytbTcz8ex1Ft+RItSKuLnwOF7K7tqsoA8bW9SxGZc++PdDOeqvdbnWrB/yDw8ePGDy8SmyL0ZSh+D/dzLyDMQkxrD6/Bp23NiJt5u3JLiYhJacEGLtgVOQeYnR6/X8umsWs07P5rlJhQzSzHFW95mpht2FxYZRc0YtnidEIQgCh8cepHrR6tnRbLviyBvYOsuxGW/XvOx/dz9l8r9YnExrMFJn+1Eexkkz4N4lCvPPgyfy8vxUtDJmEnQJbLqymUUnF8t+S9bULl6b4Q3epkf1Hrhr3DEajVwPuM74fRM4btrfx82H/0ZsplqRtK+XlKfRT2k7r51c5bd56WasH7ruhagCo9HIqvOrmLzzG55GP5W3F/EqwpQOk+lTo3e2DSiOHsS2X99O37/6y+tFvIowtuk7vFX/rVwx40+r/yvOrOCd9WPldbPA8tX2r/n10CwAmpRqzNbh/+XIS+BudBwNdxxDL4p4qpQca9+I1w+c5X5sPAALGlajZwlLfaSohCga/NqIh88fAvBbz7kUEgrSa50U9VfAswCXJl1IMcrr/5Hc8CLPSXJD/1+9b91O3I6J46S6EXvHHqVvTUvo7qkHp2g6pzlfbv8qRZvlD3t/lH09BtYZ8FIKMY5mSL3BNvknBEFgUb9FyQoxAP88eCwLMa8Vyc+MOpXw1Uj+RuuDnhAQnfxvYY2r2pW+tfqwa/QOjr9/lOENh9u8LM8Fn+Od9WOp+EMlPt7yCRcfXWLsjvdkIcbbzZstwzdlSogByTF4w9ANcsjuobuHGbV2NEajReV/6sFpWs9ry5h178pCjIvKhYktJ3B2/Gn61er7fzWYdqzUkW87fkO78u2Y02M2lyZd4L1m7+UKISY9DKw7kG86TJHX3//3A6btm86cI5KDt0apYXb3WTn2m31/JQC9aS77bgV/inu48WMti0/XlxduEq2z+CJ5uXrxa7eZ8vpnWz9n2rEZ8vqHLT54pYQYJ7kDp0YmE/xz/zHvnrqKXhQZXqY40+pU4kDAQT7cOF42gQCU8PZjetfpNr4vt0Ju0WBWIwxGAx4aD85NOGO3MLjsxtGSuCiK/Lj3J5afWcGk1hNTLC5pMIo03HGMOzFS9MqO1vWon9+bmdfv8a3J8bevfxHmNch4ddyYxBj+ubiORScWyWaf5PB2zcvm4ZvsUuPl9IMzvL6wqxyNM6bxaN5vPo6vd0xmjcmZ0kzXKq/zbadvKZWJHD72IDfMxnKS9PRfFEU+3/aFXAjSGntVgM8MlyKiaLlbMofmd1FztlNTOVLpzSPn2fFIiqB5t7w/39a0dT4fsWbkC/di4TyFuTjpPG5qt2xofe7B+QzkfP9fvW/dDjQq4INGIdmNF98J5nJENC3LtuD4+0f5rO2nsingQWQQff/qx4AVA2VV7Bfbv5Rt5h80f/+lFWKyA0EQ+LTtJ1z75EqqFbI3P3wqCzFNC/hQP783AMPL+uFj0sr88+Axd9KhlUmKp4snQ+u/xeH3DrF3zB4G1HkTV5WrzT5erl5sHLbRboXq6pWoy9I3l6BUKAGYd+wPqk+vafPiqFyoMpuHbWLFwOU5JsS86iQajCy+E8y2sGiMqcwHBUHgu47f2mhuQXIo/7DFBw5uZcp8YxLyASZUKm0Tbv1DzYq4KqXXwx+3H3DteYzNsT92+fGF4IYJLce/ckKMk9yBU5DJBMXdXZlQScrQagQ+On9DShalduWTNh9z/P1jtCzTQt5/y9X/qPdLAz7cOEGuXVPUqyjvNRub3OmdZABRFJlpFTY6vrKltEMetYp3ykuRLEYRfr5274Xj04sgCNQrUZd5vX7nxqfX+aHz91QuVJmSeUuy4a111C5eK/OdSIYOFdszp/ssed1cIsLbzZsZXadz5L1DtCzbIqXDnTgYvdHIiBOX+ej8Tb68+5R3T19DZ2UCTIpCoeD3Xr/Rrnw7aV1QMLv7r+mOSLE3h0PC2fdEyl1TwsOVt8oUt/nc39ONDyqWBMAginx87oZNtFg+D19mdJ0urxfLWzTHnJWdOHEKMplkTPkS+LtKs/2ToZGsuf9Y/qxs/jJsGraRhX0XyLOWGG2MnK8EpMq27hr37G30/yG7HodyJVKaLdb29aJFQdusmSPKlcBbI800/3nwhLvRcS+cI6P4uvvwbtN3ODbuCJv6/ktdP8fUxBlYdyBfvSZFtikVSkY2GsH5CecY2WhErq2Y/SogiiLjz17nv4eWopL/PHjCwKMXidO/mMbejFqp5u9BK5jdfRZbR2zJdJXsrCKKIt9csmhjPq1SBhfli6+CcRVLUtJD0rAcfRbBepPzvJluVd9gUquJlPctx7xe82xqUOU0eqORxQFBLLsTnKn8LU5eLpyCTCbRKBRMKmFRrU6+dJsorVW1XUGgT83enBl/mrfrD7U5tlaxmi+omZ1kHFEU+dlaG1Op1Ash7F5qFe+Ul4pQGpLs/zIwsdUETrx/jMuTLjKj63SHVxF2kjbfXA5gxT0pA7VaENCY7rndj0PpeegckVbjQFJcVC68VX8ITUo1yZa2JsfWh884Gy5FWlbK60mvEsmbt12VSlvH34u3iLJy/BUEgc/bfsY/vdbSvHQzxzY6A8TqDQw8epGJ527w4dnrLLv7MKeb5MTBOAWZLNAorzudi0rCTEiClh+v3n1hHx83b37tPpPdo3fRpFRjKhasyO+9fnslncLszZFnEZwJkwbkil4edCiafELCkWX9yGuy/6+9/5h7MVnXymQXIQmJLAiGVQ+1GIzOmWVOM+dGILNuBAJS5ujf61dhbvmieKokf6aToZF02X+GJ6ZcLLkNvdHId1cs2pgvq5VFqUg5f9VrRQvQyfRcPU3Q8tPVOynumxsIS9TS7cBZdj22pPr/8eodYnTJ1ypy8v+B822aRb6rUQ43k1p2QUAQ1yKjk92vgX99to/cxqkPT+Sa0u4vO78k0cYoUkgo6KVRM8bkK2MQxSz5ymQnoQla3jhwlmV3H/LD1Ts2LyAn2c/Kew/5+pKlds702hXp7leIOl5ubG5RhwIukr/LtecxdNx32i5mTHuzOvAxt6Ikp/cG+b1pXyTtZILf16ogO/7+eTvlMS6nuR8TT8d9p2Vtk5mQBC2/3XoxqaKT/x+cgkwW8fNw40OT469BFGXHXyeO5WzYcw4+lbLwlvRwo5tf6lkxR5UrIWtl1tx/TGAu18qEJ2rpfvAsN6MskVazbgSyOvBRDrbq1WXrwxDeP3NNXv+sahneLusnr1f3ycO21nXxc5f8RO7HxtNp/2kuR+Sel36CwcCPVhqVr6uVTVc28RIeboy3GuMmnct9Y9yliCg67DtFgEl4LOLmwtLG1VGZ+jf35n1CEnKnliwtRFFk7s1Aam09wkfnbti4MDiRcAoydmBsBX9KeUpOcceeRb7gFOfE/lhHKr1fsSSqNEx1eTVqRltpZX7Jxb4yz7U6eh46x1VTyKt1WOwHZ65xMjQyh1r2anIkJJzhxy9jtuyNKldCjlq0pkweD7a3rkdFUzmMkAQtXQ6c4diziOxsboosDAjikcnk1b5IfhoW8En3sWMr+FPaNMYdD42Us2bnBg48DaPL/jM8TZAi+8p7ebCzdT26Fi/EkNLFAMlvZnoypv/cjiiKfH/lDl9dvM392HgWBgTRaOdxtlo5mjtxCjJ2IalT3FdJnOJyOxGJOn67eZ8twU9TzYeRW7gWGc22R88AaebVr2TRdB03ulwJvExCwerAx9yPiXdYGzNLlE5Pz0PnuGiayRd21bCvbX3eNoXHao0ig45eICg297X9/5GLEVG8eeQCiabQ6j7+RZhas3yKmoyi7q5sbVWPevmkGlnROj29Dp1jh+l+zSmitDpmXg8EJN+eL6uVzdDx0hhnKYD61cVbuUIzsO7+Y/oePk+MKVqsfr68bG9Vj+KmaKtJVUrL/ktL7z5MV4bv3IIoinx7OeCFAIXH8YkMOnqRwUcv8jg+IYdal7twCjJ2ol2RAnQ0OcU9SdC+FNK/KIqsu/+YhjuO8eXFWww5dom2e05yOCQ87YNzkF9NzpYgzRSTCx1NjrwaNaPKSVoZfS7UysTo9PQ5dI5z4VL5igIuGv5tWYcyeTz4oVYFObQ8NFFH/yMXbFLHvywkGozciop9KQTmgOhYeh86J78kXyuSnzn1Kqfoi2XGx0XNhhZ1aFNYKjWRYDAy6OjFHDULzrl5nwiT4NHHvwiVvTNe4qFtkfx0KVYQkLRNP+TwGDf3ZiAjT15BZ1KVdS5WgH9b1MHHRS3vU9DVhbEVLFGL315+OfzMRFFk8qXbNmPdZ1XL0M7Kp+m/hyE02nGcpXeCX4rnyZE4BRk78n3NCriYTBzzbz/gRpJsmLmJezFx9Dp0npEnr/AsUStvvxARzRsHztLv8Plc2f670XFsCJLU2vlc1AwuXTyNI2wZU76EbKpZFfiIB7lEsxGnN9D/yAVOmaKwfDVq/m1RmwpengCoFQoWN6pOGU8p99C15zGMPnnlpRnADEaRFXcfUnvbERruOEa97Uf5/eZ9nueCWX1yPIxLoMfBc4QmSu1rmN+bxY2qo05ntKGHSsnKJjXlgosGUeSdU1f57Wb2O50+jU9knsnZVa0Q+KRKmUyfa2rN8lbBDQ+4kgOOv0ZR5PMLN/nqosXxemiZ4ixtVAM3k/bFmnfK+1PQVXLE3hIcwqlcbpoVRZEvL95ijtW9MqN2RSZWLs3qpjVZ0LAq+U3CWpROz/iz1+l64Ay3o14ebZO9cQoydsTf0433Tdkw9aLIx7nQ8VdnNPLr9Xs02Xmc/U/D5O0diuanqrenvL7rcShNdx3ngzPXclUo6eybgbKvwuhyJfBIZuBKDW+NmlHlJCdNfZKswDlFgsHAgKMXOGrypcirVrGhRe0XZs0+Lmr+blpTdlre/uhZrp9hiqLIjkfPaLbrOOPOXOOx6V66FxPPFxdvUfW/w0w4e53ruUhoDk/U0uvQOYJNRUir5PVkVdOauGfwXtMoFcxvUJURVk7BX168xbeXbmfruDDj2l3iDJJpbGiZ4vh7Zr6MgJ+HGxMqSYVbjSJ8fP5mtvYl0WBk5InLzLv1QN72WdUyzKhdMcUwck+1ykZ4m5zN339GEEWRzy7c4ner/v1Sp5LsWC4IAj1LFOFEh8b0tzKpH3sWSfNdJ/j52l20hpQzTP+/4hRk7Mz7FUvib7LPHg6J4N+gpzncIgunQiNpufsk31wOIMF0sxd1c2FFkxr83bQW+9s25Lf6VSjqJtWKMorw192H1Nt+lB+v5HwuhodxCawyqefzqFUMt3pBZIQx5fxlu/nfgY9y1N8k0WBk8NGLcgRWHrWK9S1qU93HK9n9y3l5sLhRdZQm80ZujmQ6Zcqp8uaRC9ywmi2W97JUR47VG1hyJ5gmO4/T7cBZtj4MydF8OTE6PX0PX5CjxUp5urGueW3yatRpHJk8CkHgx1oV+LiKpWr7zBuBvH38MkdDwh3e13sxcXJCOA+VUhZCssK7FfxlzeDJsOdsDcserUyUVkefw+fYYBpTlYLArLqVmVi5dJrRVwNLFaVcHqnNJ0Ijc9xnKTlEUeST8zeZf1sSYgRgVt3KL5SPAPB10fBb/SpsaFFbzr6caDQy9codWu85KefXelWwqyAzcuRIGjduTLNmzWjWrBnjxo2TP1u6dClt27aldevWzJo1y0Yivnr1Kv369aNJkyaMHDmSx48t6f4TEhL48ssvad68OZ07d2bHjh32bLLdcVMp+d6qUuyXF2/luADwXKtj4tnrdNx3Wp75KgRJo3G8Q2M6mezeSoVA/5JFOd2xCV9WKyu/7GP1BqZdu0vd7UdZeicYfSo1ZRzJ3Jv3ZXv48DLFM/1y8XGx+MrojDmnldEajAw9fpE9ppo3niol/zSrRW3fvKke16pwPn6oaXEuz22RTLejYhl89CId9p3muFW76ubLy3+t6nKiQ2OOd2jEsDLFbTRqh0LCGXT0IrW3HWH2jUDCrUye2UGiwcjgYxflPCSFXTWsb16bQibBPrMIgsDHVcowrVZFzK/bTcFPef3AWar+d4iPz93g+LMIh5gJv79yB73pvO+U96eAa9ZrO7koFfxY23L/zQoKc7iJ8HF8Ap33n+FwiKS1dFMqWNGkBoNMUUlpoVIo+Kp6OXl98qWAHBvHksNoCmtfEBAESELMnHpV0uxfy0L5ONK+Ee9V8JcnN9eex9B+7yk+PX8zx9892YUg2lHHNnLkSLp160anTp1sth85coQffviB+fPn4+rqyrvvvkv//v3p1q0bWq2W7t27M2LECDp27MjChQs5f/48CxcuBGDWrFkEBATwww8/cO/ePcaNG8eSJUsoWbKkvZqdKVIrXS6KIv2OXGC3Kbvk+xVL8rXVQ5RdiKLIpuAQPj1/Qw5NBKjhk4eZdSpT0zf5Wb+Z0AQtM67dZfGdYHkwBGlGPbl6OdoV8uXBgwfZUr49NEFLja2HiTcYcVMquNC5WZYG5fBELTW3HiFGb0CtEDjbsYkc6ZBeslK+Xm80MuzEZbYES2GU7koF/zSvTaMMhMROPHudxXeCAcjvomZv2wb4ZbAPWcX6OwhJ1PHT1TusuPcIg9X9UjaPO19WK0uXYgVfmDlHaXWsCnzMwoAguYK5GVelgl4lCjOyXAmqWpnZ9EYj4VodoYk6whK1PEvQEpaoJTRRR2iiltBEaT1Ob0AhCKgEAZVCQCUoUAiksK4gMCZO9lHKq1axtVXdNJ1iM3oPrH/whA/PXJMdiK0p4uZC1+KF6OZXiHr58qbpVJwWlyOiabH7BCD5k53t1FSO2rMHQ45dlO/fsnncGVvBnz7+RXBVZswElxqRWh3L7gTzx+0H8hjmq1GzqllN6uXzztC5RFGk477T8m88s04lhiSj7cgoWRkHQBJiJliVUhCA3+pXSXc0ppmLEVG8f/oal6z8loq7u7KwYTXq5/fOcLvSS1b7bw+yRZD57LPPKF26NMOHDwdgy5YtbNmyhT///JPjx4/z008/sXHjRkDSwLRp04a1a9dSrFgx2rdvz08//UTNmjUBmDx5MkWKFGHUqFHJtkGr1aLV2s7kVCoVGo19q8wajUaCgoLw8/NL9se7GxNH010n0BpF1ILAodcaUC6PRzJncgxBsfF8dP4mu59Y/GA8lEo+qVKaEWWLXt1qswAAWslJREFUp5l3xZo70XF8dyWALQ9t1bGN83szpqAn7SuWc/gNPPXKHWaaPPhHli3O91YaCXucc2jpYkyvXTH1A5KgMxgIfBBEGf8SGeq/QRQZfeqqbHZ0VSj4u2kNmicpeJnm9Y1G+h65wCHTLLVyXk+2tqxjk3fG0RiNRq4H3mdjrIH5AUGyLwZAQVcNH1cuzYCSRdK834yiyP6n4SwICJI1VNZUzuuJzmgkLFFHhFaHIw0ybkoF65vVStfgn9Y4kBzROj27HoeyKTiEvU/C5NBua4q6udC1eEG6FS9EHV+vdCWuS0rfwxfYa/KD+65GOUabtJD24mFcAo12niDOYBHKCrpoGF62OEPLFMcnkxpTgIDoOP4MeMDqwMc291QJd1fWNKuZ6bH0VGgknQ6cldrqquF0h8YZ9rNLSmbuAflYUWT8uRty7S4FkhDT2+QknlH0RiN/3A7ip2t3iTd9b/k0ava3rU9Rd8cU9cxK/9NDes5pd0Hm7l0pJK98+fJ8+OGHlCtXjv79+zN69GhatGgBwPXr1xk7dix79+5l5cqVXLhwgenTLSXh+/bty9ixY6lRowatW7fmwIEDeHpKjqgrVqzg0qVLTJs2Ldk2zJ8/nwULFths6927N336ZH+Rxt+Dw1j0WHrJNPBy47fyRTM1IGWEp1o928KiWfgonAQr+3tzb3c+KlGAIi6ZH1wuRsfza1AYl2Jtcxd0yZeHscXzUUDjmBdotN5A50v3iTUYUQmwqZo/hbPQDzORegOvXwwkziiiFmBjKufViyL34rVcj03kelwi12MTuBWvJdEo4qtSUsRFRWGNiiIaNUVcrP+ryGM1UBpFkcn3QmS/ArUAM8sVpVHezFVCf6438Na1YB6YomtaeHswo2zhLM/m04PWKLIu5DmLHocTqbe8bDwUAkOK+PBmIW85wiUjBCVoWRvynE2h0cRmwXFRCaRcizp5XBQC08sUpol39kw6YgxGDkXGsjs8muPP49AlMxoX1qho6+NJcVc1IpJmQQTZ6d2IaNqOtB3pmVn2JFI+/t9q/mhSqamUWa7EJDArOJRz0bZjgqtC4I38Xgwo7E2xdD6roihyOjqelU8iOfLcVjsnAC29PfjYv0CWx5mJtx+zP1LygRpTzJfhRXOmEKtBFPkuMITNodJYoAC+LV2IDvkyHhqflOAEHV/de8rFGOl3qenpyvyKxeRMx/bkTnwiDxP1NHfQM1Oq1IvJJ5Ni1zfPuHHjKF26NAqFgjVr1jBu3DjWrVtHXFwcHh6WTnp4eBAfLzlYxsfH23xm/jwuLo64uDh5Pbljk2Po0KEMGDDAZltOaGQAvi5WnJ27ThAcl8DJqHguq9x5vXhBu7ZDbzRyJjyKPY9D2f0kTM4Ga6aQq4Yfa1agS7ECWRai/IHXq4j89/AZ31wJ4J4podx/YdHsi4xlXIWSvFO+RIajO9Ji5vVA+YXW178oDcpnLJlXSvgDI+KMzLp5H50I62P1TCtfFp3RyM2oWC5GRHMxMoqLEdFcjYwhIQWberjeQLjewNXY5KO78qiU+Hm4UdzdlUSDkYOyECOwrHF1XktHvZvUWFOwEB32neG5Ts/ByFiWR+v5KoMJzzJCgsHAinuPmH3zvpwpFqT+vF2mOB9WKkl+l8w/b/5A0wrwvV7P2vtPWHwnmBtRsXiolOR3UZNPoyG/q8ZqWU1+Fw35NGryuWgo4Kohn4saN5OJwyiK6EURvVHEIEp/5nX5M1HEYBTJ76rJkCbBHrPRKsAYJF+2HY9D2Rj0lANPw9GZ5phPtHpWPI3M1LkBvqhennIlk69wnVX8jEaqeroS4uHFvNvB/PcwBCOQYBRZE/Kcf0Ke07V4Qd4t70+tFEzZCQYDG4KeMv920Avjl4dSyZulijCirB+lPTMn7Cfle9/8NN11EoMosvzpc8bVqpIlM3Vm7gGDKDLuzDVZiFEKAvPrV0mz1Ep68QfWlfKn1Z5TBMclcCEmgZUOGBcexyfw4b4zPI5PZFqtCsk6JmcHdhVkqlatKi8PGTKEzZs3c/nyZdzd3YmNtUQtxMbG4uYm2fLd3NxsPjN/7u7ujru7u7xu1shYH5scGo3G7kJLaigUihRvXk+Ngqk1yzPk2CUAvrh0m7ZFC2RZlRmaoGXf0zB2PXrGvqdhRGpfdOgSgGFl/fiiahm8sqDiTY43ShSmU/GCLLz9gJ+u3CHKYCTOYOTHa3dZHviIydXL0sOvsF20T3F6A38ESF78CgE+qFTSrurLsRVLsvBOMLF66eV8wSS0JKfuT0opDzc8MfLMAE8TElM0d0TrDVx7HsM1q0FaJQgsblydDsWyLthWyJuHxY2q0+fweQyiyOyb96mY1zPDNva0iNMbWHY3mDk3AnmSYGu+7elXiC+qlctSaG9SvDQahpcrwfByJdAZjenO4ZIUBXYe6JK7RirjQHrxcXWhf6li9C9VjEitjm0Pn/Fv0BMOPg238VHLCLV9vehbsiiKdGhjRKMBoh+BV/EMP7v18vvQoGA+7sXEMe/WA1bee0i8wYgR2BgcwsbgEJoW8OG9iv60KZwfhSAQkpDIkjvBLA4ItsllBZJvx4iyfgwuXSzTTv0pUSFvHgaXLsaSO8HE6A38ciOQnzJoVk6O9N4DBqPIe2eusfa+lA9LJQgsbFSNrsXtI8SYyefqwqJG1ei87wx607jQuIAPr5kSt2aVKK2Ovkcu8tA0mVke+IhBZYpn+jnNCg59vs0/aqlSpQgICJBNS3fu3KFMGSmuv3Tp0qxbt04+JiEhgeDgYEqXLo2Xlxf58uUjICBA9pGxPvZloEuxgrQqlI/9T8N4GJfAzOv3+CKDUrFRFLkcGc2uR6HsfhLK2bDnKb40a/t60bZwfrr6FaJyXs8U9so6aoWCUeVK0FChZ3W0jsV3HmIQRR7GJTDixBX+vB3E1JoVqJsv9Qic1IjS6ph1I5Awk9mkW/FClLGzn1E+Fw3Dy/ox60YgWqMoZ9VNSmlPN2r4eFHDx4uaPl7U8MlDHpVSdnLTi/AwPoGg2ASC4uJN/xN4GCetB8clyBFXKkHgz4bV6GwHIcaMOZLpo/M3ACmSyc/dlSYZ9LtJjhidnsV3gvnt5v0XXjjti+RnkI8rHSpXcKifVE4MjjmJt0bNm6WK8mapooQnajkUEkGCwYCAFNItIEVDWa8rBBAQUAjSNo1CQf18eVPMr2KNKIro1g3GeG09ynqjUXf+NVPtLuXpzrTaFfm4SmkWBwSzIOCBnFTwyLMIjjyLoIKXB9W987Ap+CnaJOHndfPlZUz5ErxerGCG/PgyykeVS7P2/mM5/H9UuRKUzmMfjU9qHH8WwZcXb8njjEoQWNyoOl3srKk3Uy+fN5Orl+OLi7cAGHPqKgfbNchwYENSEg1GBh27KE/OirmoWNWkRo49p3YTZKKjo7l69Sq1a9dGEATWrl1LVFQUVatWRa/X88MPP9C+fXvc3NxYuXIlffv2BaBOnTokJiayadMmOnbsyOLFi6lUqRLFiklhZ506dWLx4sVy1NLBgwdZvHixvZrtcARB4KfaFWiy8zg6o8icm4HsfxKGYBpsFKZ9FJjWTYORIIDC9P9qZLRN1JE1XmoVrQrl47Wi+WlTOB8FXbMWKppRvFVKfqhZmrfL+vHlhVuyo+bpsOe8tvcUffyL8GW1shRLh6NZosHIqbBIDj0N58DTcM5HPMd6nPswmUJ99uDd8v6svf9YTtZWNo87NXy8qO6dh5q+0v/kZoVGK62NRqmglKc7pVJQfxtFkacJiQTHJVDI1YUSDoguGl7OjxtRMSy+E4zWKPL6gbOU9nSjWUFfmhfypWkB3wyp0KN0ehbeDuL3W/cJTxJe26VYQSZWLkXVvJ7cv5/92WpfJXxdNHYzOaSE8fYOjNfWA2A4/QfKOm+jKFw90+fL56JhUpXSjK3gz5r7j/nt5n05Ku1mVKxNVXelIPB68YKMKV8iw5FImaWQmwvvlvdn2rW76E2lC5Y0znx/0yIgOpYpl26z1SpgQq0QWNqoOh3tOKFJjjHlS3A8NIKtD58RodUx7MRl/mtVN9NCh1EUGXv6qhwK76tRM6d8kWx/91hjN2ffiIgIxo0bx/3791GpVJQvX54PPviAihUlld2SJUtYsWIFRqORbt26MW7cOFl9efXqVb799luCgoKoXLky33zzDUWKSDbdhIQEvvvuOw4ePIiXlxfvvfceHTp0sEeTs0RGQ86+SVI3IytU9PLgtSIFaFckH/Xze+eYFJzcd7DncShfXLzFLauByk2pYFzFkrxXoaSN/4xZ03TwaTgHn4ZzIjRC9rRPSne/Qixq5LiBJjxRS2BsPGXzeKQ7RDU3hB0mRWc00ufQeQ6mUC+rcl5PmpsEm8b5vZM1Oz7X6ph/+wHzbj3guVUeCgHo5leICZVKyWHJufE7yE7+H/ovGvRo/6iH+Oy6vE1R8Q00/dakeWx6+28URbY/esacG4Fy+LOXWsXg0sUYWdYvyxqCzBCt01N321FZy7irTf1MaZBT+w7CErVMu3qXJUlSWFTO68m02hVpnIF0C1khUquj5e4TPDAFaoyt4M83NcqncVTyfHXxFnNN5RPclAr+bV6bAjGR/z/h168SGR3AYvUGBh6R0tAbEclIQk83pYLmBX1pVyQ/7Yrkz/ZcISmR0negMxpZduchP1y9IxeqAymk9JMqZdAZjRwMCedwSITN50mp4OVBi4K+tCgk9d2RqubMkFtfYnF6A3/efsDux6GcCX8um7SSohQEavp40bygD80K+VI+jwdL7gTzZ0CQTUFKhQA9/QozvnIpufaTmdz6HWQX/w/9159djH7LOy9s14w6iaJIjVSPzUz/z4Y9JzA2nvZF8uOZjakCkmNRQBCTzknm2MYFvNnSsm6G/YOS+w4SDAbm33rALzcCbZ6lwq4aPqtalv4li6bL5GdPzoU/p+O+0/J48HfTmnTIoL/MH7ce8NmFm4A0LqxoUpPXCufL8WfAKchkEnsMYJYwSkmwMQs45m2iKGJESi2eG30E0voOIrU6pl29y8KAoHQ5KxZ1c6FFoXy0KORLs4I+FHFzTN4De/EyvMRi9QZOhEZw+GkEh0PCuRARle4cLEpBoI9/EcZXKpmib9LL8B04kpe9/6I2lsTZVSFGyqauqNYX42VJE6Oo8Dqa/v+kevzL3n+d0UjjHcdls9eqpjVpn8GXu/V3gCCw/sETvr0cINfqAmkMf6+CP+9WKJnlYI+sMP/WAz41CSLeGhUH2zVM98R4Y9BThh2/JI8fv9SpxFtliueKeyBnxeFXHGtnvf9HvDVqvq9VgaFlivPVxVvsNGU6NpNXraKZSePSopAvZTzdHZ5n51XDQ6WkTeH8tCkshXhHanUcfRbBoafhHAoJt/FVMKMSBN4sVZQPKpakpJ1CXp3kTgzHZ1uEmIpvoO46n8TAwxD9COPNLRgfnUdRtFYOt9JxqBUKvqxelrdMkaWTL92mTeF8mdL+Hn0WwdeXbnMhwpJZVyHAwFLF+KRKGQpnsdSFPRhZzo9joRFsCQ4hUqvn7eOX2dqqLpo08j0dDQln9MnLshAzsXKpHAu1Tg6nIOPE4ZTz8mBVs1rsfxLGzsfPKOzqQotCvlT39sp29eqrjrdGTediBeWIqafxiRwOkYSaG1Gx1PTx4r0K/rnGfOnEcYgxT9Ef/VlaEZSo2n6LoHZF1XQS+u0fAqA/8B2aN9fnYCsdz+vFClI3X17OhD3nZlQsqwIfJ1vjSBTNeYiQcxEZRZGgmDi+uv2Yg5G2lejbFs7H5BrlHRo9mlEEQWBO3cpcjogmMDaes+HPmXL5NlNTyZR+/XkMA49elCPM3ixZlE+r5K7IYacg4yTbaFU4H60K58vpZjixopCbC738i9DL3zEJ05zkXvQHvwetFD6rrDMMRX7J+VNZeyj6IzMg+iHGW1sxPjyLolidnGyqQxEEgW+ql6PT/jMATDh7nS8v3pKTJ5r/0uvXWCWvJ9/UKJ9rxzovjZrFjarTYd8ptEaRebce0LiAT7LpIB7GJdD70DnZ6b9N4XzMrFsp12nOXz6jZi5CFX7Doec3xMdzdexYLo8YgS4y0qHXcuLEyauDMfQWhjNSYV40nqhafi5/JqhdUTX7SF7XH5ya3c3LdhoW8KGjyTdGL4pE6fTE6g0kGIzojOkTYgq7ujC3XhUOtGuYa4UYMzV9vfiuhkUL8+6pq9yPsc2Y/1yro8/h83L27po+eVjSqHqu9Nd0amQygRgfiX7bBxS4vBpj3q0oyrZxyHXu//Ybj1asAEDh4kKVuXMdch0nTpy8Wuj3fgWiVIlK1WQ8gqdtnhpl7bfQH5kOUcEYb23D+PAMimJ1c6KpmcIYegvdym4gKNAM3org7Z/mMT/Vrki0Ts+j+ASUgvDin8K8jM12lSBQQSkysW41PO2chdiRDCtbnKPPItgU/JQonZ63j19iW+t6uCgVUsK7oxe5bkp4V9LDjdXNauV4lFlK5M5W5XIMV/7BeHk1APrNI1GOOYvg5m3fa8TG8uD33+X1RytWUPytt8hb9+UZTJw4cZL7MD44hvH6RmnFszDKRu+/sI+gckHV7GP0W98DTL4yAzZmXyOzgKiLR7f2TcQIqYCxbtenaPr8neZxxd1d2dwq4+OrOWrH3jXmkiKKol1NOoIgMKteZS5FRnEvJp7zEVF8fekW39eswLunrnDkmZTwLp+Lmn+a18rRhHdpkft0RC8ByjrDEEq2lFaiHqIzOcbZk+ClS9GF2yY1uz5xIqIho/V8nThx4kRCFEV0uz6T11WtvkTQJB9ar6w1BPL6AabMv8GnsqWNWUW/YyJiyBV53XhtA8YHx3OwRVlHf+RnEn8oiG7jSMSE53Y7r5daxZJG1XExmYv+vB1Er0Pn2BD0FJBymK1qWsvupWHsjVOQyQSCQoHqjfkY1ZI3uvHSKgxXN9jt/MbERO5bmZFcS5QAIPrCBR4uX2636zhx4uTVwnh9E2LwCQCEApVQ1hyS4r6CSoOq2cfyuv7Adw5vX1YxXF6L4eyiF7brdn3Ky5oyzfjkEvq9X4I2GsOFv0icVxdj4GG7nb+6jxff17Rk+T3wVJpAKwRY3Kh6lurlZRdOQSaTCHn9iGrwhbyu+28sYvRju5z70apVJD6WzlWgc2eqzpsnfxYwZQra8OTTz2cWURQxBOzGGBaQ9s5OnDiREaOCERIicroZ6UI06NDvsYxZqrZTEZSpexcoaw6GvNJEyhiwC2PQCYe2MSsYwwLQWWUoVr3+O0KBSgCIwScwXvs3p5qWaURRRLf1fRCtSrc8D0K79DV0e75A1Cdfgy+jvFWmOD2S1PP6uXalDCcHzCmcgkwWiC/dFaFSN9NKOLpNo7Ms9Rv1egJ//VVeLzVxIj5NmlC4d28AdBER3PnOfjMjURTRbxqFbsXraOfVw/j4gt3O7cRJUsSYEAzXNyHqE3O6KVnGcHkNutlVKPRPCwyHf0I06NM+KAcxnF2EGC5NVgT/ZijKd0zzGEGlQdX8E3k9t2plRF0Cun8GyuHkiupvoqw9FFW77+V99HZ88WcXxosrEYMks5jgWxbBv6npExHDkRloFzbH+Czr0bOCIDCzbmVq+ORBAD6vWoYhuSjhXVo4BZmsIAioOs8GD0mSNQbsxHB2YZZO+XT9euIDAwHI17o1eWtJWTXLffMNSk/JlBW8ZAlRFy5k6Tpm9Ls/w3DhL9OKyUnOjjbY/2ciT5wgZv/+l1Zlnd2IcWEkLmiCbk1fdH/3QDQmXyD0ZcAYfArdxpFg1CEYdRj2T0G7uDXG0Ns53bRkEROi0B+whFGrX/sh3Y6jypqDELxLAmC8sydX+pvod32C+OQCAEL+Cqg7z0YQBBTlOqAo1RIAMeIuhjN/5lgbM4oYH4lut5U/U6eZaIbsRNV2Kiik6CjxyQW08xuiP/VHlsehPGoVu9rU59YbLZhQuXSWzpXdOAWZLCK450f9xnx5Xb/z40ybaESjkXszZ8rrJcePl5ddixSh9Eem3A6iyI2PPsryi0B/5GcMx2babBMj7qLbNMr5ck6DZ9u3c7ZTJ56MH8+dKVNyujm5HlEU0W0eA8+DADDe3Yvh5G853KrMIUY9RLu6LxhstUriw1No/6iP/uS8XCek6Y/9AnHPAFBU7Z2hUGpBqUZpo5X51u7tywqGa/9iOP2HtKJyRd1rBYKLNOkTBAHVaz8i1W6XkgCK8fYzBcbeusXFAQMIX7TI7mOmfv83EBsCgKJSd5Rl2yEolKiaTkAz4jBC/oqmHRPQb/sA3co3EKOfZOmaaoWCfC6arDY923EKMnZAWb4DyjrDpRVdHLp/h2VKzfxs61Zib0hqQu+GDfFp0sTm8xKjR+NRXnLKen7qFI9Xrcp0m/XnlqDfY0mCpWr5BbhKJeWN1zdiOOHMWZMSiSEhXH3vPXn9/uzZPF63ziHXMtzeSeLvddCueAMxJsQh18gODGcXYryx2Wabfs8XGEOu5VCLMoeoi0e7uo9cn0go0ZSw9svA15SyXR+PfvuH6FZ0QTQJbTmNGPUQw7FZ0opCjap1xgVvZY0BCD6lADDe3Yfx/lF7NjHTGMPvots0Wl5XdZiBonA1m30URWqiqN5fWokPR394ml2urYuM5Hzv3oRu30743LmEHzhgl/OC5OArC2dqd9Ttf7L5XFGkJppRx1HWH2M5JmAXib/XwXB9k93a8bLgFGTshOq1HxF8JHWcGHwSw9EZGTpeFEXu/vyzvF5qwoQXVL8KjYYKP1lu6NuTJ2cq46/h+ib0W961tL31FFQtv0Dd3eLtr9/9aa527MspRFHk2nvvoQu1LYB5bexYu5n7AESjAd3+b9Ct7IYYchVjwE60y9ojxjy12zWyC2PIdfQ7LJliBf9m0oIhEd2GoS+N34Ioiug2jUR8dBYAwdsfVZ+/0Rauj3rkCZT1LC9U4919JP5eF8PFlTmu3dTv/xb0UtZWZf3RKHwzbjaQtDKfWs6ZC7Qyol6Lbt1gSJRM4YqqvVHWGZbsvurWk0HlCoDh5G8YI+5l7dqiyNV33iH+/n15270ff7TLby0ajTYOvqrmnyJ4l3hhP0HthrrTTNQDNoFnYWljfJhkut00GjExJstteVlwCjJ2QnDxRN1jMQjSV6o/8B3GR+fTfXz4/v1Em16EeapXJ1/btsnul69VKwp27QqA9tkz7v74Y4baabh3AN26QfJDomw4DqUpHbmyQieUTSZIOxr1aP8ZgBgbmtKpXkmClywhdOdOANQFCuDZoQMAxoQELgwYQGJI1rUmYmwoupXdMBz8HrAMjOKz62iXvma36Lik6CIj7f7SFXUJ6NYPtrxI641CM3CzJZrkyUX0B3OnA2lSDIenYbzyj7Si8UTdfz2Cu1RVXNB4oO78K+pB/0EeU8HBxOfo/h2Gbk0/xNhnOdJm49OrFh84l7w2jrsZRVn9TXmyZrx3INMhwI/XrOHWF1+gDQvLdFvA5Lz7SKqPJPiWQd3ltxT9fgTvEigbmrSoBi36vV9n6dr3587l2bZtNtuenz5N2L59WTovgPGSlYNvvnLJJiy0RlmuPS5jTqOo2FXeZji/FO0f9TEGncxye14GnIKMHVH4NUTZdKK0YtSj+/dtRF1Cuo69O8OiwUlOG2NN+alTUbhJ1YmDFiwg+urVdF3D+OgculW9wSDNgBU1BkiaJKtrqVpPQShhMmlFPZTMZLnM3p9TxN6+za3PLea4ynPnUmjKFPLWrw9A4sOHXBo8GKM28xoG48MzJP7ZCOOd3dIGQSEJmqbEZGLoTUmYiXqY+Y4kIT4oiPP9+nGgZEnOdumCPsZ+Mzn9ni8Qn14GQChQWbrf1G6oeyyRHRYNR2ZgfHDMbtfMDAmPHvFg3jyClyxJ9n43XN+Mfp/55Seg7rEERaGqL+ynLNMWl3fOWEwZgPHGJknlf2OLo5qfIvo9n1tm9s0+QnDPfA0gQalC1SJrWplHq1ZxZdQo7s+dy4U338So02WqLYYb/2E4MVtaUWpQ916J4OqV6jGqphPBJHgar6zFGHw6U9eOOHaMgMmT5fVib70lL9/94YcsTQYkB18rk3/HXxBUafusCB4FUPddg6rrPFBLyevEiLtoF7dGf/rlcXDOLE5Bxs6oWnyBULgmIM2g9Xu/SvOYiOPHiTwmDeQe5ctT8PXXU93fzc+PUhMkzYloMHBj0qQ0Hx5j6C20K7qCNhoARflOqLv+gZCkAJigVKHp9Re4S/kDjAE7MRyZnmYfcguiXotu32QSl7TDcHun3c5r1Om4PGIExnhJs1B8+HDyt2uHoNFQbdkyXIoWBaRIpvT8Hi+0WxTRn16AdnFr2SEWj4KoB29D3eYbNG/tliNHxLDbaJe0y7IPhmgwcP/33znesCGhO3YAEHH0KJeGDsWoz3ooseHWDgwnTb5WShfUvf5CUEsCuKJITVStvjQ1xCgJzInRWb5mhtoXH8+T9es517Mnh6tW5eann3L9ww+5Z2XiBTA+uYxuw1B5XdVmCsqKKT+jgpsPmh5LUPdZBW4mwSE2BN3q3ug2jsi2qEDD3f0Yb0u/K17FUTZ4J/UD0oGiWn8E37IAGAMPYQw8lO5jn58/z/UPPrCsnzzJnakZL0gpRj5At3GEvK5qPw1FkZppHie45pV8AU3odn2S4ec0MSSES2+/LWdYLzVhAhVmzEBTrhwAz8+cIWzv3gyd0xobB9/KPVCWbZfuYwVBQFV7KJrRpxCKSZMrRAP6re9juL459YORxqDAOXM437cv4YfS/7vmBpyCjJ0RVBrJxKSU6lIYTszGcO9AqsdYD5wlP/jgBeEiOfzHjsWtlOR8F3nsGE/Wr09xX/F5MNrlnSFOMhMJJZpIMxhl8gXOBK9iqHsuRfb03z8lzT6kl8SnTx3mMyBGPUK77DUMh35EvH8Y3d/d0Z/8Pe0D08Hdn36STX/u5cpR/ptv5M9cChWixsqVKFyk3/zhsmUEL16c/nZr49BtHC7VtTFpywS/hriMOo6yVEsAFD4l0by1y+KHFXFXEmYi76dw1tSJuniRU23bcuuzzzDExtp8FrZ7N9c//DBrM8voJ0leNj+9oMFQNpmA4NdQ2j/iHvqdH+FoRFHk+ZkzXB8/nkMVKnB52DDpxWOlhbn388/EmVIgiLHP0K7qCTrpO1JU64uy6aR0XUtZuTsu755DUb6zvM1wYTmJ8+piuHfQfp1KBtFoRG81s1e3mSILkVkhqVbGcOA7SMd9khgSwsWBAzEm2kZ6Bf76K8927Ur39UWDDu26wWBKQqio1A1lvVHpPl5ZZxhCPknoEB8cxXjzvwxc28CV4cPRPpEig3yaNaP0p58iKBT4jhwp73cnk1oZ4+OLqTr4phdFvjJo3t5nMaUhols/BOPDs6ked3faNG5/+SWhO3dytls3AmfPznH/rvTiFGQcgKJgZVRtLC863b/DU5yFRV24QNiePYBUisCc+C4tlK6uto6/X36JPvrFGa0YF4Z2xevyLF8oVB1N//VpDmrKMm0ssxfRiG7dkCyF9umeP+d8374cqlCBs126oHtu31mp8f4REuc3QrR2UBaN6LePR7d9AqIx8zWqIo4f594vvwAgqFRUW7AApbu7zT55a9Wi0qxZ8vrNjz8m/MiRtNsdFiAltbq4Ut6mbPiepIHxKmazr+BdQhJmTDNiMTKQxCXtMuS4aIiN5dYXX3CqdWuizlt8uIq//TY1/v4bQSOpsR8tX87dnzI3kIpGI7qNwy3hvuU7J/uyERRKycHcpAo3nFuC4ebWTF0zLRIeP+ber79yvEEDTrVtS/DixeijouTPXUuUwKeplGzMmJDAzY8+wqhLRLumHzx/ILW3aF1Ji5mBwn2CZyHU/dehemM+aPJIG58HoVveGUPAbvt1MAnGK2sRH5+T2lCoOopq/dM4Iv0oqva1EgaOoHmSuh+GUafj0ltvkfhQMofmbdCAclYpC66OHk1CcHC6rq3fN9lSYsHbP+O/h1It5WExn2/3Z4iG9Jm37vz4o6yp0BQqRLWFC1GopMzIHq1b41mlCgBRZ88Sujtjv61oNKLb9oGtg6/JnJwZBKVK0lRV6ytt0MejXdUzxYnPo7//5u4PP1g2GI3c/uorLg8daldTs6NwCjIOQtnwPRQlm0srUcHoto9Pdj/zCxKg5Pvvo1Cnvwx8gddeI7/J2TTx8WPuTrc1AYmJMWhXdkd8dh0Awae05GiZzkrdyuafoCjdRlqJfYpu/eBMhZXH3rnD6XbtZCfZiKNHOdejR6YirpIiiiL6E3PRLusAsaaIHq/iKOtatAGGk7+hW90nU178uufPuTJypDxjL/PZZ3jVrJnsvkX79cN/7FipXXo9l4YMIf7BgxTPbbi+Ge2fjS0F7tQeqHutQN1hesrasrzFJWEmn6k2yvMHaJe0wxh2J82+hO7ezbGGDbk/d66sGveoWJF6O3ZQ6ZdfKNipE1X/+EPe/+6PP/Lwr7/SPO8L/ToxB+MdSTjHswjqN+an+LJR+JZB1cFy3+o2j7FbmLkhIYEn//7LuV69OFylCgGTJxN765bl2u7uFOnfnzpbttD0wgVqrlolmwhDd+3iybc9ER8clfuh6bcmU1oNQRBQ1RqCZsxpy5hg1KNb2x/j44tpHp/w6BH3fv6Z8MPpc641PjqPbus4eV312vfp0vKmF0krY0nUlufC3FRn7rc++0w2nbsUKUKNZcvwHzeOAp06AaALD+fSsGFp+ssYbu/EcNSkvVaopXwx6RzLrFFUfF32AxTDbidbmykpoXv2cM80vgpKJdUXL8alkCWlv6BQUOpjS12qjPrK2Dr4lk/TwTc9CIKAuut8hBKNpQ0xT9D+3eOFSXXY/v1cG2e5X/K1aSMvP924kVNt2xIbkLvL1zgFGQchKBSouy0AF8kBzXhxJYYktT5ibt4kZIvkAKgpVIiiAwZk+DoVfvxRNmk8+P13Ym7eBEy+Imv7IT40Vaz1LIx60H8IeQpnoA9KycRkisIwBh7KsINf2IEDnGrTxuYFAtKs5Vz37lkSZkRtLLoNb6HfMRGMkoClKNUSl1HHUXeZg6rrH6CQZkzGW1vRLm2HGPUoQ9e4+fHHJARJ2izvRo0o+X7qA0zZyZPJ17o1ALqwMC4OGPCC6UY06NHt/hzdmj6QKGkFhPwV0Iw8irJqrzTbJHgVlYQZc0KsqGC0S19LMatsYkgIl4YN43zv3nJfFC4ulPniCxoeOoR3w4byvoV79KC8VQmM6x9+mCHVv/HReZt6PuruCxE88qd6jLL2UIv5JTYE3ZZ3s6TSTnzyhJuffCKZjoYOlTSeVqYj78aNqTx3Li1u3qTqvHn4NmsmFYLNk4cKVrPS24sPS5Y+pQuafmtf0JBlFIVPSdSDt6Oo+Ia0QRuDdmW3VM2DYfv2caJZMwK+/Zazr7/OvV9+SfW7MYbelnzhTPeVolI3lGWSj4DMCoqqfRDyVwBAE3IW8d7+ZPd7uGIFQQsWACBoNFT/6y9cChdGEASq/P47rn6S1iEtfxnRFHhgRtVuKori9TLVdkEQUL9mifbUH/guVb+l+KAgaTJjouxXX72Q4wukunh5qkk5bKLOn5cnbmkhxkfaViTvlD4H3/QgqF3R9F2LYMpzJIZclbK3m7RQ0VeucHHwYESTT5zfiBHUWreOmqtWofKS3l2xN25wqnVrQpJEaeUmnIKMAxG8/VF3tGhcdFtsC0sGzpwp25f9x45F6eqa4Wu4lywpv1xFvZ6bH3+M0aBH9+8wy6zY1RvNwC2Zyx/hUUBy/hWUABgO/5QuJ1pRFHnw55+c79kTvUlY8ahUiZqrVqHOL73Yos6f5+wbb6CLyHimTWPYHbQLW2C8vEbepmwyAfXA/xA8JEdlVe23UA/cDC5S9Vbx8XkSFzbH+PRKuq7x5N9/ebx6tXTuPHmo+scfCEplqscoVCqqLV6MW2npu46+fJmrY8fKLx8x+gm65Z0sM0tAUaU3mhFHUBSomM7eg5CnsCTMFJTU2UQ/lISZZzflfUSjkeC//uJYvXo8tfKh8mnWjIZHj1J64kQUmhcHTP+xYykxRkq0JRoMXHrrLZ5bmaFSQtTGSqHWRmmQVDYej7JMmzSOMs8cf7c4mN/cguH8sjSPS0piSAg3P/uMIzVr8uCPP9BbmS9d/fwo/dFHNDl/nnrbtlFs4EBUefK8cI6CXbvi26gmANoYgQcnVajfmJ/pl2ZSzJMDobhJeIx5LCU7TJJtVjQYuPPDD5zr2ROdVZhywDffcG3s2GQj48TnQWiXd5JNekKJxqi7p99XK6P9sNbK6DeNeEEr+PzsWa5bZSevNGMG3vUs36Pa25vqS5YgmMwzKfnLiPpEtOuHyD5+ivKdrfw/MoeieD0UVU1m/LhQ9EeSz/tl1Gq5PHQoOlOh3vwdOuD/XvLXFgSB0lZamTvpzCuj3/+NxQxbuYfdBU/BI7+Ua8bNF5Cyauu3vk98cDDn+/TBYHJJKNCpExV+lKJYC3TsSP19+/CoKI1J+qgoLr75JgFTp8ra3NyEU5BxMIoaA1DIhSXDJNW5KBIfGMiTf6ScFGofH4oPHZrySdKg5AcfyDOb8AMHeDKlJ8arpnwXKjc0b254Idtlhvrg3wRVW8ssXbdhKGJkyiYTo07H9fHjufnRR/JNn799e+rv3EmBjh2pu2ULmgLSSyv64kXOvvFGhip6G25uszXJaDxR91mFut2L1XyVpVujGXZAruBLVDDaRa3S9E9IePiQ6x9+KK9XnD4dN3//dLVP7e1Nzb//lmtjPf33X+7N+An98Tkk/tHAEumhUKHq8LMUzePy4ks1LQTPgmiG7EAoZPptYx5LwkzIdWJv3eJMly5cHzdOfqGrfXyo8ttv1Nm8GY+yZVM9d/mpUynUrRsAxrg4LvTpIzvApoR+x0TEMEkrJBSpjar15Az0pZAkzFidK72+P9qwMG5//bUkwPz+O8YEKeWBws2NIn37UmfTJppevEiZzz7D3eQgnxJiWAClq19DUEovoIcXNMSrqqe7H+lBULuh6b/O4usUegPt6t5yIU1taCjnevaUfJRML8I81S1teLRypSTgWGkzxdhQtMu7JPGF24CgsfXlsieKKr0QippKHUQ/RrusvfybJT59ysVBgxBNAlfx4cMpNnjwC+fIW7duqv4yoi4e3ereiPdN/mZexVF3W5Ahv5iUULX5BpSSIG84MSfZMe32V1/x/IyUq8a1RAlpMpOKma5A587ybxV94YIcDZgS9nLwTQtFvrJo+q2V+5t4fDHnu7Qm8ZGkoc5bty7VFi60mah5lC1L/T175HEA4N706Zzv188ubgH2xCnIOBhBEFB3mWspLHl7B/qt47j360z5Je83ejQq00svMyjd3W1U4gFLDmDQAQoV6j6rUJhtpFlA2fgDFBW6SCvx4WjXDUI0vDgr1IaFca5bNx4uWSJvK/nBB9T8+29ZVelZqRJ1tmxBU7AgANGXLnG2a9c0E2SJRiO6/d+iW9VDzuYp5CuPZsQRlJW7p3icomAlXEYctgy62mh0K7uhP7MgxetcfecdWZNUqEcPivTtm2rbkuJZsSJV/7Tkb7gz9UeezvvY4seTRzIPqRq+m6VBWfAoIAkzppB/Yp8SPKkVx5s0kf0SAIr07Uvj06cpOmBAuq4nKBRU+eMPvBtL94722TPO9+yJNklGYzPGa/9iOGf6zdUeknCWQfW4suLrKGsNkVa0MaYcRinP/nQREQR89x1HatQgcNYsjHFxAChcXfEfO5Zmly5Rdf58fFu0SJePiBgfiW5VT9zcovCrJ6naRYOR6xMm2D16Q/DIj3rgJlkLJd4/gu7f4UQcO8aJ5s0t6e4VCsp+/TUNDhyg2pIlshk54vBhTrVrR9y9e4gJUWhXvoEYKmnjBN8yaAZtyZT/SIb6oFCienMDOm/J8ZeoYHTLOmAIucOlIUPkl6R3o0ZU+P77FM9T4p13KNBRqsRt7S8jJkajXdEVY4BJS6NyQ9N7BYK7r13ar/AphbKBKcO5PgHdvsk2nz/duJEHJp8xQaOhxl9/ofb2TvWcgiBQ+hNL0sHUtDKSg2+SDL5ZcPBNC4V/U9Rv/InRANe3qokNlHzR3EqVoubq1S8EMACoPD2ptmQJ5b79FkzPUNju3Zxs2ZLoy5cd1taMIogvS3xVLsNoNHL//n38/f1RpGOQNNzaju5v6WWrjYVTi90QDSJKT0+aXb6M2scnS+0xPL3K+S5tiLgtzUb96uspN2MByur2i1YQ4yPQzm8o2/QVDcbysOK78ncQc/06F/r1k9N2K1xcqDx7dopCQOytW5x5/XW0T6WXu2eVKtTZtAlN/hd9KsT4CHQbhlryYiDZ/9Vv/JlmIiz5HNo46Rw3LLVIlI3Ho2r7nc2L7v5vv8mJ71yKFaPR0aPJDmCp3QNi1CP0x2dzb86f3D9iyqKsEanZT4tn466ou8xG8Cz0wjkzixgfgXZ5F8KOnufaFjWiURJW3EqWpNLMmeRr1SpT59VFRnK6Qwe5BljeunWps3mzPOgZjUaCrh6n0NaekBAJgOqNP1HVenH2na5+JEajnVcPMTJQOlfb76REZkna9GDePB7Mm2cTeaRwcaHYkCH492mGKvIkYvhdq6NMwpssxAk2/0BADL2J+ERyvhV9KnF2qUj8PakdVX77LVkftoyOAy8c//CMlOBQG8fDc0ruHdWAURqSNQULUm3RInybNZP3jzx9mgv9+8slMtT5fKkyoDB5FFKEEnmKonl7HwqfkhluS2YwGo0E3ThLoX3DwSRIBRz15fFpSah0KVqUBgcO4GKatKSELiKCE82byz5c/mNH4+9/zOLjp8mDZsC/KPyb2rX9YnwEibMqm8K5BTQjj6EoWovYgABOtmolm10q/vILfm+/new5kt4DoihKL/qL0r1UY+VKCnbu/MJxhgvL5RQFQr7yaMacsZtvTEqIosiVHk15sl9Koqpyg3r/LMazaY80jw0/eJBLb78tmzoVbm5Unj2bQj17ZukZsAdOQSaTZGYA059fhn7r+9zdp+fhOckEUmJgZyrMXZnGkaljuLQa3ZZ3iAuJ59xyDaJRQFCrqLFiJQXat8/SuZNifHgW7eJWcr6TyEZTKNDyXcL3HebyiBEYTKF6moIFqbFypY1NPDlib9+WhBlTbgbPypUlYcZkegKpgJpuTV9Es6lBUKBq842UhySD2gzRaES/53Obqt+KSt1R91iMoHYj+upVTrZqJavE62zahG+LFsl/F8ncA8awOxiO/ozh4gowaBFFuLFNTehtSWXrVtKPBgcOpzmzywwRB/dwrldvjDrpkS5URaT8+AG4tPkk2Vot6SU+KIjTr71G4mPJv6tAx45UX74chUqFQa8j5s+WuIRIOSoUVXpL2pgsaJmM94+iXdIWEEGhlvyHitRAHx3Ng/nzuT9njo3/i6BWU7RLE/wau6AOPwzx6TdTJotbPjQjjxBx/h7nekgDvDpfPpqcOfPChCOrggxAwpl/uDZ6GGEBluN9mjSh2qJFuBR+0Tk/PjCQ8337Emty7BeUIuXb6ShYyxvN0L0oClbKVDsyg7n/JXw16Je15/Ghe9zeI0XcKVw01N2+g7y1a6frXM/PnOF0hw6y42mVN7T4ljKCmy+agZszVLE7I+iPz0G/U8oNpCjZAkXvDZxu146Ya1JB08J9+lB1fsqRd8ndA8+2b+dCf2kSmadaNRocOmRzvBgfSeKcarJvjHrQfw5xyk7Kne+/5+40qWimQilSracWr3IF0Aw/iMIndbMrSGPBpcGDbVI3+I0ahWbYMEqWLZtjgozTtJSNqGoNQei5hcdXTA+6UqSI23p0Oz9Jdy4Da0R9Irqt76Pb8Bbo4nD3ESneTFK7ijo9F/r1I3DOHLuqxRXF6qBqbwmXzXvsa+72LMqFN/vLQkye6tVpsG9fmkIMgEe5ctT97z9cihQBIObaNc68/jrx145guLgS3fYJaBe2sAgxbvlQD9yCqunETL0sBYUC9Ws/oOo8W66LZbz+L9ql7dGHPuDKiBGyEOM/dmyKQkxSjI8vov1nENq51TCcW2xJbKdyoeIng/CsKKnf4wODuDx8eKZTs6dE9NWrXBg8TBZi8pc3UK5NIlxeTOLsKui2vIsxIjBT53bz86PW2rUoTc6xz7Zvl/yfRBHjkWmyEEPeEqi7zMmy/4LCvwnKpuaaXzoS1rzFvV9mcKRGDe58950sxAhKBYUbFqDu4DhK++9A/XBT1oUYjSeavqtR+JQiX+vWFOouaVF1YWHctvLlsBdRFy9yZvhUGyHGr56eGj8OS1aIAUnLVnfHDrwrS1oO0SBwc4eGh9G9ETLgMG5PhDxFiK/2EwH7LWkDynbOg1f59Ed65a1bl7KfWZyDb+5Uk2jIj+atXQ4TYkCq/yXXkAo8yI3Rb8pCjEeFClT65ZcM39P5O3TAq1YtQHL4f7bVNj+Sfv8Uhzr4JsfDv/6ShRgEgYqDy+FVVJQiBVd2R4yPTPMcbn5+1N2+naIDB8rbgubP5+Ho0XapM5dZnBqZTJLZmVjA1KlyPoIiNfSUbSXNPgS/hmh6rUDIWzxd5xEj76NdO0AumgagrDkYRdtpXB37AU83bpS3Fx0wgEq//CLb17OKKIro/n0b/blV3N6jJuSGxUEsfzkD5TsqUJWqj7JUCxQlWyAUr4+gSv7aokGPGHqD2NM7OTfuZ7ThkkrazddI9Z5aNB6WfYUitdH0XYXgnT6n27Qw3N6J7p8BoJUEsLvHvXl4UjLNeZTwpe7MYSi8CiC4eoObj+1/V29EQcGjU+spELAcMSBJtIWLF8p6I1E1fA/BsxDxgYGcbN1ajn7wadqU6suWocmX+do3ZuIDAznVoYOs1fJt2YKqoyojnl8g9w0AhQplzUEom07KVARb2MGDnO/ZU54xlx7xBsU814FoAEGBZugeu/hjgZQ+QLugKTHXLnNts5qE51bPmAIKVdTj18CAW94kw5eLF4oy7VCW74hQorGUj8dmiDNFjyWzDUBwL4DgYvFXS3j0iGP160tCuiBQb9cuGwE9s+OAKIo8XLZMijI0ZbtVebpQoU20pIVQuaIZsgOFX8Nkj9Xv/Ajd0Tnc2a/iyRWLg3uRvn2pPHu23Z71tDD3v7CLC6dat5bvwaI19ZRpqUcoWAXNkJ1phuADGCPuoV3agavLHxF+VxpT8tauTt2dezOUXyszGK6uR/fPAJ5cVXJ7t3Qtpbsr9ffuw7NS5dTbncI98GznTi6YzOqeVavS4OBBhJhHGO7uR795tOQbo3bHZexFh/rGAITu3cuFPn1kv8zy339Pibf6oV3UEjFMSo2hKNUS9YDN6TJvme/fG5MmIZomZW5lStP4xEmH/1bJ4RRkMkmmTEtRURyuVg398+cIKhUNl4xHdf4HOVwVt3yoey5Ns76G4fZOqfaLefapdEHV+VdUtaXIJ9Fo5O5PP9lkZvVu2JAay5fbmGyyQlxgIJff7EHUNYsfgn8jHX71DbwweVG5ofBriKJkc4Ti9RGjghEfncf4+Dzik0tyZeT4SIHL6zUkRpv8O3yMVO+lReMhoKzzNqoOPyOoMx6inhrGJ5dIXNGNiMtPuLpJeoAFpUit/lo88qfxaKg95NT1Mu4FUDUcK83ykjhbhh86xLlevWSNj5u/PzVXrcKzcuoDZWokhoRwun174u9JGqu8detSe+NGVJ6eiHFh6I/PxnDyd7nGlqmDKGsMQNnsYxT5yqTrOqIo/q+9O4+Lqvr/B/66l2FHkUXZkU3ccEOUXDHULBTLKMSlpMytDHvYx49mmVpZ1i81rVTyq/IpDCWzXLPPxwVzSVly3wFlUTbZl2GGmXt/f1y4MGwCMlxG3s/Hg4dyGZhzzpy59z3nnvM+4LNv4OG2r3Hz6+p8SJ7PKWHTh4Oe34fQr9o7qZXk7NuJq/Pfg1pZ1aF4dOvFwdlXBWOLGsGHpTtYz4lgPV8A6zyi1ecZpGzejDvLhaXGnfr1w9CTJ8WMri05D6hLS3Fz8WJk7KlOHdDZ2xv9du6ELGEVuCtRwkFjKxjMjgFr3UPj91WnvhA+0QPgwSKjYiaSvo8Wf95l+HAMiIyEgWXrTIptDMdxuH/3Lh4tXIjCOGETxi6+g9F33D2wpcJkX8Z2AAxe/6PRSbpczm0ofwwAih+gohy4uNsUigLhouvy3nvoUWOTRm3geR6Z//bGjR3J4NRCf+v5vBLd+puCdfIV+pXzcLAOQ+qsBGuoD3CKEsT5P4ui68JKvt6vmMLaUXNBg2zsp5CNatqWFy1VfOUK4gICxBFz5wULxMUhXJ6QxqJqabveoFmQNZItmVeUgM9LAp+XCD73LgpiL+D6ptNQFqnQ+9+vwnF5/QsotI0CmRZqyQns3jffiLum2s+Ygb7ffw8uPQ7KX2aIadABBnqjl0I2ZgUYVjNnCc+poYr5DOq/1qLqUyRj4SqsTKpn07TM337D9QULxOWoRk5OGBgVhU5edXftbaqKggLcW7cOqeHh4gWZNTFB33Vr0LWPIbj7p8DdOyVO1mwueSGDq/uMoSgUJsiaONth8O+/wsit+Rd7nudRkZcHRWam8JWRAUVmJpRZWeL/FZmZUGRlgq+ozljsNroCDt7NzJVg7gTZ8MXQGzSr0SWvBbGxuDxzJpSVw7B6ZmbwCg+vdzLg41QUFiJ+4kSUXBOWoZv27AmfP/6ocwHj5flQnf8W6vPfi6u9AACMHtj+IZCNWlbnYgkAfFkeuOQT4JL+B3XiMaBYSDGfFqeH+2eFT10My8PjNXc4rTsPvVYMINJ27MDtJUuqMxB35dDrhQqYWPIAKwPrPAKsZ4DwVU/ZWxOnUuHCmDFiO/dcuxbO8+cLP2vmeaAgLg43wsJQevOmeMxpzhx4fvYZWENDIZHlrhfBVSaYYyxcYTD7FBgz4TaSKjYcqiPVSRllk7dC5h2KrP37cW3ePPG9buLujoHR0TB1b1qg2lIcxyFuzhwUVeYpMnJ0hO/Jk5CxRVBGjAeKK4MZO28YvH6k3pVUXOZVIYipyn9j3Qvyfl8gPug1cfRvYHQ0uj73nHbqoFQi8bPPkLJpk3jMrr8KHv71ZDFnZWDsvME6DwfbfQRYp2HgjS2Rcv8enM0qgIdx4NJjwaXHgc+6irxkXvyAZGLNwXuGUvygxzg9A4NZ/9XqBF95Whpix48XR8q6BQaif0SExjJrLu08lBETAHXlyOCzK8H2fgl87t3KgCURXO5d8HlJ4utZk7IUeJSoB8e35sBg4jdaq0tjKJBpoeaewMofPMCFMWOgzMkBGAbDY2NhWrljKl+Wh4rf3wJ3pzpzIuviB/2g/4iZePnSHFT8GgouuXpnVbbnJOi/9H+NLrMsvHgRl6dPFydq6pmawuuHH5p94eQUCqT+8APurVsnLksGAJmtLbx374Z5rbT9fEEK1PdOVe6Qe6p6R+daGAs3MPbeYO0GgbEbCNZuEMpzihE/aZK4gsHE3R32M2eCKy8Hp1BALZeDUyiE78vLoa48XvU9p1BAVVQERXa2GGw1leWoYRj4w5dgFIXgywuERGVV/8rzwZfnA/IC8WdK1gTGw9+BrH9Ig9sK1Fb+4AEuTZ8urmoAw8Djo4/gsnhxk+/Fq+Vy/BMUJC6xNnJ0xJA//4SRQ8NzEnh5AdQXvofq/LfiCiPh+VmwXsGQjfo3oCiuDFz+B/5BnLg0VOPv8EDSWQtkxMvFY/avvw7PTz+Fvrl5k8rfYBnVatz56COkbtkiHrPydkSvF1SQuQ+HnmcAWPdxWl9aXFtBbCziKi+kep06YXhsLIzs7Jp8HpCnpODuJ59oJCbUMzNDn02bYPuy5ooRvrwQyh1jxTxJjL0PDEL/BHf7ECp+DUXVhxjZc2shG/6e+HuFCQm4NG2aGCTrW1ig17p16DZxolZuNamKi5GyZQuSK5dWs0ZGGHL0qLiFB/foDpQ7x4spBxiHoTB47ZDGKkMuPU7IRFy5CSRjO0B4jGlXjdWD+paWeOavv2Dk2LRb701Vdu8ers6ejaJ//hGPWY8Zhr7vvwBkXACXeg4oyWjkLwCw9ABXkgNWWTc7MM8Dl/cYoDhT6Bt93vBEt8AAsA5DwXqMb/I5oyUqCgoQ98ILYtBsPnQoBu/fDz3juttsqK/tRcXemXWON5XauBv0B86AwYQvHv9gLaBApoWaE8gUXLiAy6+9Jp5gbKZMQf8aeVYA4XaQ+twGqI5/LMw5AIRtBYL+A0ZmCGX0DPETsbBq51PojWjaha88IwOXZ8yofrMyDDw+/ljYafsxv89zHDL37UPiJ5+gvMa+QayhIZzmzwcbFAQ3L69G24DnefD598DdPwU+8yoYcycw9oPA2g5s8IIkT00VgplG9ipqDfrW1jC0tYWhrS1M3N3htnRpk4fkn2TFirqsDNcXLkTWvn3iMZugIPT99tt68zloPK9KhSuvvYacP/4Q6mBlhSFHj4qB8ePw5YVQx26B6u9NTZ8cKzMG6zIarMd4sB7PAV3ccHX2bGTvr17Kbmhnh97r14s5QZpLVVKCq7Nna6R27/7uu+ixatVjMyq3hRuLFuHBf4SMwzZBQei/fftj+4CqqAj3NmwQEvXV2PnZrG9f9I+IaPA14wvTofg/P/E9zzgMFTaBrNyKQ2/kEuiPq7tdiDw1FZdCQsTJqgAg69IFtkFBsAsJgbmPzxNNxubVauSdOoWHu3cj++BBcPLqYLbv1q2wDwnReDyXfRPKiOeqR1ucnoHBzENgDM3A3T8N5c9TxHlcjKMvDGbsF88JPM/j8vTpYj839/WF96+/PlHOrZoyoqNx8/33xSXWjL4+eqxeDef588V0DFXnLj71HLjKL/7RrSb8dQZMt75gHYci72EnXFkWDkBYlfnMmTOtuvdVFU6lQvGVK8g/exb5Z8+i4O+/xYnxJu7uGPLf/zY6J091+v9BdbyR28MmXcFYuYO16gHG0gOMlQcYqx7gu7giNeMRLb/WRU29iD348UchoVblhCgjZ2f4HDoEY+f6l8NyKWeEbeqrhvAYVviqPIHB1AYGr/5UvflcE6nlcuHCWeMTod3Uqei9cWODWyPk/fUX7nz8MYovXao+yDCwmzoV7h9+CEMHB63mD5CnpSEhMBDyx2SU1cCy0DM2BmtsDEMbGxja2VX/a2ur8X+Dbt3qTdHfVE+69Jbnedxfvx6Jn1ZfkDoNHIiBu3Y1OLLCcxyuv/MOMqKEeRR6ZmYYfPAgzCtXSDTr+RXFQkBzbiMgr5uMkOnapzJwGQ/WeWSd+Uk8xyFt+3bcWbkSfGUyOgCwfeUV9Fy7tt58QA0pT0/HxZAQ8fYNI5Oh17p1cJw1q9n10hZlXh7ODRki5tHw/u03WPj51T8/QqXCw59+QuKaNWLOF0AIOt2XL4fDrFniPJuGcFnXoNzhL+6bVEVv8FuQNbI6rKKwEFdnzxb2mKrFxN0ddlOnwi44GMYuLk2tOkpu3kTG7t3IiI4WR3dr6v7ee/BsYB4Ll3kVyv9MEINmpvtIyJ55Vxhdqpwfx7qMhv60X+tkuK6dX0bWpQuc3noLTnPnPjY3TUNUJSW4tWSJ+B4ChHbpt317gxvC1sSX5oBLO18d2DxMgNrAHDLnZ6Dn5AvGcShY+8FiXXieR9z48WKG4P4RERrZcluKq6hA0cWL1YHLhQtiUFaTvrU1hv7vf4/Pas3zwhY0if8VPmxaeghBi5WHELg08KGzNVIQPCkKZFrocS8eV1GBO8uXixumAcIeN/0jIh67UoUvyUbFvlBwySc0jjPdR8LglZ/AdLJrUZl5nse9r7/W2JzNfMgQDIiM1NjJteTmTdxdtarOpmeWzz4Lz9WrxRTcbdGB1eXlyIuJAa9SgTUyAmtoKAQqhobC90ZG0Kv8lzUyAiOTtUr68qZorfpnHz6Ma/PmPTYHD8/zuLtiBVK++w6AkG3U+5dfmrxEvCG8ogTquHCo7xwBY2YD1mM89NzHN2kFHcdxSDx/HiXr1iH3ePVtT30rK/T88kvYBgU99vUovHhRuCVSeR9f1rkz+v/0E6yesF7a8CAyEjcqdzg3cXeH75kzSMvM1OgDj44fx52PPtKYB8MYGKD7ggVwWby4Wbff1MknURE5WVwQwPZ9VdiriW18hIrnOOSePImM3buRfeiQxshJlS7DhsFu2jTYvPhivWVSPnqEzL178TAqqvo2aA2yLl1gExQEZswY9Jw4sdH3AJdxSdihvuYtzUpsj+ehHxzV4M7ihfHxiJ80SZz/AwgjwnbTp6P7woXNmgdUdOkSrs6ejbKk6n2h7KZNQ6+vvqp3762mUKvVSE1JQXcXlwbb4NHx47gYFARA2HNu2NmzzR6VUZeXoyghQQhczp1DQWysmM26PvpWVrAcNQpuH3wAs549m/VczUGBjA5r7MVT5ubiSmgo8k+fFo85zZ0LzzVrmrw0jefUUP/1BVQxawDwQgbasZ/U2UuoJbL278e1BQvEN4GRoyMG/vwz9Lt2RfIXX+BBZKTGbsFmffvC89NPxV2dq7SHDiyl1qx/8fXruDx9upgVmTEwQJ9vvoH99OniY2pOFgfLov/OnbB58cUnet4nJSZEc3ZGVnQ0bn/wgcYcKusJE9B7/foGR5iyDhwQJqlWXmiNXVwwKDoapp6ebVH8ZuM5DvEBASg4fx4A4PbBB2CDg9G9e3eU3b6NOytW1BkJsZkyBT1WrmzWCEhN6uv7UHF0CViXUUIm62ZODlUVFyP7wAFk7NmDvNOnay1HF4KCrgEBsAsJgcWIEcg9fhwPo6KQe+yYONm2CiOTwfq552AXEiIk29TXb/J7gHsQL0zqrTHCxPaeUnn7vPE6ldy+jfsbNiBz717NMjEMuk2ahO5hYY3mreJ5HqlbtuDuypXi6LiemRl6r18Pu+DgRp/7cZpyHuB5HnETJqAwVshU3G/Hjjpzo2o/vjwlBYUJCSiMj0dhQgKKL1/WuD1Zm4GNDSxGjIDF8OGwGDECpr16tcmHuvZwHaBApoUaevGKr1/HpWnTxLkdjL4+eq9bV++GaU16nvx7gFrV6iszii5fxqXp06F4INyDZ6tSzteI8A0dHODx4Yewmzq13jkK7aEDS6m166/MzcWVWbOQf+aMeKz7woXosXo1HuzahZthYeLx3hs3tovbLrXbQJGdjVtLlmjMnZF17owen3wCh1mzxBMrz/O4v3FjdWAGYYRgQGRkq+TW0abi69dxYfRo8Gq1MDIQHg7ExODBjz9qfAAw9/GB55o16OLrK2FpNZWnpyPjl1+QsXu3mBlYA8PUCXQAoPOgQbCbNg22QUEar09z3wNc2gUoIwMBRRHYATOgPzm8WR/OytPTkbp1K9IjIsQRzCpdhg2DS1gYrCdM0BjtUD56hOtvv41HNXbW7jxoEPpt3w4Tt+bnU6pTpya2Qe7Jk/inMsGiac+eGHbunHherSgoQNHFi0LQUhm4VDSwr1kVQwcHIXCp/DJxd2+z0eia2sN1gAKZFqrvxcvavx/X334b6lIht4hBt24Y8NNP7epEVpMiKwuXZ8wQ791WkXXuDJfFi+E8b169M9yrtIcOLCVt1J+rqMDtZcuQvn27eKzz4MFCSvDKi6THxx/DdfHihv5Em2qoDbIOHMCtJUvEfbQA4dZqn02bYOTggJuLF+NhZKT4M7vgYPT59ts2S+T2pO589JF4i682I0dH9Fi9GjYvvyzJhaUpeJ5H0aVLyNi9G5l794rzfmoytLcX5tNMnQqzXvVnDG7Je4CX54MvSAFjO6DF7VNRUID0iAikbtmi0ccAIUjovnAh7IKDkX/+PK7NnavxmO7vvguPFSueaH5cTU1tA57nEff88yi8cAGAsOxeVVKCooQElN6589jnMXZ1hcWwYegyfDgsRo6Ecffu7aJ/tYfrAAUyLVTzxWMAJK9dW53+GULEPyAystHlsO2BurwcN8PCkBEdDUZfH06zZ8N1yZImfSpuDx1YStqsf9r27bi9dGmdoX3nd96B52eftYsTGNB4G1QUFODOhx/i4a7qvcRYY2OY9uiB4itXxGPuH34I13+1bMsJqahKSnBu6FBxh2dAWJbtungxnOfPb/QDQHvDVVQg9/hxZOzejZIbN8TRF8tRox67WkzqcwCnUCAjOhop331XZ4RJ39paCNAqL3EGXbui79atsB47tnXL0Iw2yI2JwT9NmOgr69IF5j4+MB88GOY+Pug8eHCbJDhsCan7AECBTItVvXgOlpa48fbbGntp2AUHC6uBdORkxvM8iq9cgUG3bjCya/pE4vbQgaWk7frnnT6NK7Nmidsa2IWEoO/mzVpZutlSTWmD3BMncGPRInHlSRXW0BB9t2xpdK5Ae5bz55+4PHMmeLUaDq+/Dvfly1u8kkZXtZdzAM9xePTnn7i/aRMK/v67zs8tn30WXlu3aixqaC3NaQOe5xE/caKY/wkQph906tevOmjx8YGJm5vOBPbtoQ9QINNCHMch8cwZPPr3v1F6qzKvAMuix+rV6L5woc50wifRHjqwlNqi/vL795G8bh2MHBzg+v77kuxj0pimtoGqpASJn3wirOLjeRh07YoBP//cpI1F27PS+/fxMCcH7oMH03ugndS/IDYWKZs2IfvwYTB6evD4+GPhnKyl8jW3DZQ5OUjbtg36Fhbo7OODTv36NZgCQxe0hz5AgUwLPTpxApdDQ8EVCTPwZZ07o9+OHbAep/1dTNuL9tCBpdTR6w80vw0K//kH+WfPwjYoCEb29m1QQu3q6H2gPddfkZkJMIxWRmFqas9t0BbaQ/2ffC1vB5QRHY3rCxZU7wPj6YkBP/8MUw8PiUtGSPtm7u0Nc29vqYtBOgBDW1upi0DaSMcLH1uB+dCh0Oss7BdiPWEChh47RkEMIYQQIgEKZFrAxMUF/XbuhMXs2egfGQlZ586P/yVCCCGEtDqdCGTy8/OxaNEijBw5Ei+//DJiK7MjSsly9GhYLVzYLjazI4QQQjoqnQhkvvzyS1hZWeHYsWNYtGgRPvjgAxQW1t0ynRBCCCEdS7uf7FtWVoaYmBjs378fRkZG8PPzg7u7O06dOoXJkyfXebxSqYRSqdQ4JpPJYNBKWRyrcJVZVqv+7Yg6eht09PoD1AZU/45df4DaQNv1b8pKqHYfyKSmpsLExAQ2NZbQeXh4IDk5ud7H79y5E9tq7DgNAK+++iqCn3BjsIak1Ury1RF19Dbo6PUHqA2o/h27/gC1gbbq7+rq+tjHtPtARi6Xw9TUVOOYqalpg7eW3njjDcyYMUPjmLZGZNLS0uDk5NQhcwcA1AYdvf4AtQHVv2PXH6A2aA/1b/eBjLGxMUorN2GsUlpaCpPK3ZprMzAwaPWgpTEsy3bIzltTR2+Djl5/gNqA6t+x6w9QG0hZ/3bf6s7OzigrK0N2drZ4LCkpCW6tsP06IYQQQnRbuw9kTExM4Ofnh/DwcJSXl+P06dNITEyEn5+f1EUjhBBCiMTafSADAMuWLUNOTg7Gjh2LDRs24PPPP4e5ubnUxSKEEEKIxNr9HBkAsLCwwKZNm6QuBiGEEELaGZ0YkSGEEEIIqQ8FMoQQQgjRWRTIEEIIIURnUSBDCCGEEJ1FgQwhhBBCdBbD8zwvdSEIIYQQQlqCRmQIIYQQorMokCGEEEKIzqJAhhBCCCE6iwIZQgghhOgsCmQIIYQQorMokCGEEEKIzqJAhhBCCCE6iwIZQgghhOgsCmQIIYQQorMokCGEEEKIzqJApgXy8/OxaNEijBw5Ei+//DJiY2OlLlKbmzt3LoYPH45Ro0Zh1KhRCAsLk7pIWrV3717MmDEDvr6+CA8P1/jZwYMHERAQAD8/P6xevRoVFRUSlVJ7Gqp/fHw8hgwZIvaDUaNG4eLFixKWVDuUSiVWr16NiRMnws/PD6Ghobhy5Yr484iICIwbNw7+/v7YuHEjnsadXxprg4MHD8LX11ejH2RmZkpc4ta3Zs0aTJgwAX5+fpg6dSr++usv8WcdoQ8ADbeBpH2AJ822dOlSfvXq1bxcLudjYmJ4f39/vqCgQOpitak5c+bwhw8flroYbebkyZN8TEwMv2zZMn7r1q3i8bt37/Jjxozhr127xhcXF/MLFizgN2/eLGFJtaOh+sfFxfEvvviidAVrI2VlZfwPP/zAZ2Rk8Gq1mj969Cjv7+/Pl5aW8qdPn+YDAgL4tLQ0Picnhw8ODuZ/++03qYvc6hprgwMHDvALFiyQuohad+/ePV6hUPA8z/PXrl3jR48ezefn53eYPsDzDbeBlH2ARmSaqaysDDExMZg3bx6MjIzg5+cHd3d3nDp1SuqiES0aM2YM/Pz80KlTJ43jR48ehb+/P/r27QszMzO8+eabOHz4sESl1J6G6t9RGBsbY86cObC1tQXLspgwYQL09fWRkpKCI0eOYMqUKXB0dIS1tTVmzpyJI0eOSF3kVtdYG3QULi4uMDAwAAAwDAOVSoWcnJwO0weAhttAShTINFNqaipMTExgY2MjHvPw8EBycrKEpZLG+vXrMW7cOLz99tu4e/eu1MWRRHJyMnr06CF+7+HhgczMTJSVlUlYqraVlZWF8ePHY8qUKdi2bRvUarXURdK61NRUFBUVwcnJCffu3avTB5KSkiQsXduo2QYAcPXqVYwdOxavvvoq9u7dK3HptGft2rUYMWIEXn/9dfj4+MDDw6PD9YH62gCQrg/I2uyZnhJyuRympqYax0xNTVFYWChRiaQRFhYGNzc3sCyLPXv2ICwsDHv37q3TNk+72v3BzMwMgDByZ2JiIlWx2oyLiwuioqLg7OyM+/fvY9myZTA2NsbMmTOlLprWlJeXY8WKFQgNDYWZmRnKyso0+oCpqSnkcrmEJdS+2m3g7e2NPXv2wNbWFjdu3MC//vUvWFhYYOzYsVIXtdUtW7YMS5YsQUJCApKSksAwTIfrA/W1gZR9gEZkmsnY2BilpaUax0pLSzvERasmLy8vmJiYwMjICLNmzYKJiQmuXr0qdbHaXO3+UFJSAgAdpj9YW1vDxcUFLMvCzc0Ns2fPxokTJ6QultaoVCosW7YMTk5OmDNnDgDhta7ZB0pLS2FsbCxVEbWuvjZwcHCAvb09WJaFl5cXQkJCcPLkSYlLqj16enoYOnQoYmNjcebMmQ7XB4C6bSBlH6BAppmcnZ1RVlaG7Oxs8VhSUhLc3NwkLJX0WLZjdiU3NzckJiaK3yclJcHW1rbDBDK1Pc39gOM4rFixAgzDYNWqVWAYBgDg6upapw+4u7tLVUytaqgNamMY5qldtVOTWq1Genp6h+oDtVW1QW1t2Qee3rOOlpiYmMDPzw/h4eEoLy/H6dOnkZiYCD8/P6mL1maKi4tx/vx5KJVKVFRUYNeuXSgqKoKXl5fURdMalUoFhUIBjuOgVquhUCigVqvx/PPP48SJE7h58yZKSkqwY8cOTJw4UeritrqG6h8fHy8usUxNTcX27dsxevRoiUurHZ9//jlyc3Oxdu1ayGTVd+UDAgKwb98+pKenIzc3F7t27UJAQICEJdWehtrg3LlzyM/PBwDcunULe/bseer6QUlJCY4ePYqysjKoVCocO3YM8fHxGDRoUIfpA421gZR9gOE7QtjcyvLz87Fy5UokJCTAxsYGS5cuha+vr9TFajP5+fkICwtDSkoKZDIZPD098d5776FXr15SF01rwsPDsW3bNo1jK1euRGBgIA4ePIjNmzejtLQU/v7+WL58uTir/2nRUP0LCwuxa9cuFBcXw9LSEgEBAXjrrbc0LnJPg4yMDAQGBsLQ0FBj1GnTpk0YNGgQdu7cicjISHAch5deeglhYWENjlboqsbaICYmBkeOHIFcLke3bt0QHByMkJAQCUvb+kpKSvD+++/j9u3b4HkeTk5OePPNN+Hv7w8AHaIPNNYGGzZskKwPUCBDCCGEEJ1Ft5YIIYQQorMokCGEEEKIzqJAhhBCCCE6iwIZQgghhOgsCmQIIYQQorMokCGEEEKIzqJAhhBCCCE6iwIZQgghhOgsCmQIIe1KfHw8fHx84OPjg4cPH0pdHEJIO0eBDCFEMqtWrYKPjw/mzp0rHjMzM4OXlxe8vLyeuq0eCCGt7+naEIUQovN69eqFiIgIqYtBCNERtNcSIUQSgYGByMjIqHN869atmD9/PgDgwIEDsLe3x6pVq3Do0CHY2dlh3rx52LJlC0pKSjB58mS88847+P7773HgwAGYmZnhjTfewCuvvCL+vZycHGzevBl///03CgoKYGNjg8DAQISGhj51m1sS0hHRu5gQIomePXtCLpejoKAApqamcHV1BQDcunWrwd959OgR1q5dC2tra5SWliIqKgrnz59HdnY2zMzMkJWVha+++gqDBw+Gq6srCgoKEBoaiqysLPE5kpOTsXXrVjx48AArV65sq+oSQrSE5sgQQiTx9ddfY+TIkQCEoCYiIgIRERHo1atXg79TUVGB7777Dvv27YONjQ0AIC0tDVFRUfjll19gaGgIjuOQkJAAAIiOjkZWVhasrKzw+++/IyoqCl9++SUA4NChQ0hLS9NyLQkh2kYjMoQQndG5c2cMHDgQAGBra4usrCy4u7vD3t4eAGBhYYHMzEzk5eUBAK5fvw4AyM3Nxfjx4zX+Fs/zuHbtGpycnNquAoSQVkeBDCFEZ5iamor/19PTq3OMYRgAQpBS+/eqbl3VZGRkpI1iEkLaEAUyhBDJVAUS5eXlWvn7ffr0wdmzZ6Gnp4fPP/9cHLkpLS3FyZMn8eyzz2rleQkhbYcCGUKIZFxcXAAAN27cwNSpU2FsbIw5c+a02t8PDg7G/v37kZ2djaCgILi6uqK0tBRZWVlQqVSYNGlSqz0XIUQaNNmXECKZyZMnw9/fH2ZmZkhKSsK1a9fAcVyr/X0LCwvs3LkTgYGBMDc3R1JSEhQKBQYNGoTFixe32vMQQqRDeWQIIYQQorNoRIYQQgghOosCGUIIIYToLApkCCGEEKKzKJAhhBBCiM6iQIYQQgghOosCGUIIIYToLApkCCGEEKKzKJAhhBBCiM6iQIYQQgghOosCGUIIIYToLApkCCGEEKKz/j+tVPkXn6NoRwAAAABJRU5ErkJggg==",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "complete_ts = AustralianTourismDataset().load()\n",
     "ts = complete_ts[[\"NSW\", \"VIC\", \"QLD\", \"SA\", \"WA\", \"TAS\", \"NT\"]]\n",
@@ -337,128 +133,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 391\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "f63c65898e8a46359badf46e5e392a79",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/110 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-01, min: 1.26E+00\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "b4c803be617149dfa126e172663d2886",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/110 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.07E-01, min: 1.04E+00\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 1.04E-01\n",
-      "Epoch[391/391]: 100%|██████████| 391/391 [00:03<00:00, 98.15it/s]\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 96.875% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency D corresponds to 87.5% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - D\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "# model supports only datetime indexing\n",
     "ts = TimeSeries.from_times_and_values(\n",
@@ -490,7 +167,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -499,20 +176,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "train, test = ice_heater.split_after(split_point=0.8)\n",
     "\n",
@@ -528,943 +194,23 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.forecaster.add_lagged_regressor) - n_lags = 'auto', number of lags for regressor is set to Autoregression number of lags (40)\n",
-      "INFO - (NP.forecaster.add_lagged_regressor) - n_lags = 'auto', number of lags for regressor is set to Autoregression number of lags (40)\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 413\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "86bf965d96784d6c8d67494053139ce4",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/110 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 6.02E-02, min: 1.88E-01\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "40f6c7fc9a824898a8a1c7ef7378dde4",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/110 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 5.09E-03, min: 1.88E-01\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 3.13E-02\n",
-      "Epoch[413/413]: 100%|██████████| 413/413 [00:05<00:00, 81.31it/s]\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "model = NeuralProphetDarts(n_lags=len(test), n_forecasts=len(test))\n",
-    "model.fit(train, ice_covariates)\n",
-    "preds_cov = model.predict(len(test))\n",
-    "\n",
-    "preds_cov.plot()\n",
-    "train.plot()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 413\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "1e5d52c456314ce385d0d19f73f4203d",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/110 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.32E-02, min: 3.40E-02\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "238f395a39f64f70a2a48a9ca58293f8",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/110 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.32E-02, min: 3.33E-01\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.59E-02\n",
-      "Epoch[413/413]: 100%|██████████| 413/413 [00:04<00:00, 89.92it/s]\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
+   "source": [
+    "model = NeuralProphetDarts(n_lags=len(test), n_forecasts=len(test))\n",
+    "model.fit(train, ice_covariates)\n",
+    "preds_cov = model.predict(len(test))\n",
+    "\n",
+    "preds_cov.plot()\n",
+    "train.plot()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "model = NeuralProphetDarts(n_lags=len(test), n_forecasts=len(test))\n",
     "model.fit(train)\n",
@@ -1476,18 +222,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 14,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "12.828873303132173\n",
-      "12.877951456793664\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "print(mape(test, preds_cov))\n",
     "print(mape(test, preds_no_cov))\n",
@@ -1496,260 +233,56 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.forecaster.add_lagged_regressor) - n_lags = 'auto', number of lags for regressor is set to Autoregression number of lags (40)\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.\n",
-      "INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16\n",
-      "INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 627\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "fa455f96c51b4ba6af70833fdf223ed4",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/106 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 7.69E-03, min: 1.69E-02\n"
-     ]
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "d1fb1758fe784c05a0735623cdafc24c",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/106 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
+   "outputs": [],
+   "source": [
+    "ice_heater = IceCreamHeaterDataset().load()\n",
+    "ice_train, ice_test = ice_heater[\"ice cream\"].split_after(0.8)\n",
+    "heater_cov, _ = ice_heater[\"heater\"].split_after(0.8)\n",
+    "horizon = len(ice_test)\n",
+    "\n",
+    "model = NeuralProphetDarts(n_lags=horizon, n_forecasts=horizon)\n",
+    "model.fit(ice_train, heater_cov)\n",
+    "preds_cov = model.predict(horizon)\n",
+    "\n",
+    "model = NeuralProphetDarts(n_lags=horizon, n_forecasts=horizon)\n",
+    "model.fit(ice_train)\n",
+    "preds_no_cov = model.predict(horizon)\n",
+    "\n",
+    "print(\"MAPE with lagged regressor: \", mape(ice_test, preds_cov))\n",
+    "print(\"MAPE without lagged regressor: \", mape(ice_test, preds_no_cov))\n",
+    "\n",
+    "# for some reason results vary a lot here for model with lagged regressors\n",
+    "\n",
+    "ice_train.plot(label=\"train\")\n",
+    "preds_cov.plot(label=\"cov\")\n",
+    "preds_no_cov.plot(label=\"no cov\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Future Regressors"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Following the example from their website"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 7.69E-03, min: 1.69E-02\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 1.81E-02\n",
-      "Epoch[627/627]: 100%|██████████| 627/627 [00:04<00:00, 152.45it/s]\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
-      "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
-      "WARNING - (py.warnings._showwarnmsg) - /Users/blazejnowicki/miniconda3/envs/prophet/lib/python3.9/site-packages/neuralprophet/forecaster.py:3089: PerformanceWarning: DataFrame is highly fragmented.  This is usually the result of calling `frame.insert` many times, which has poor performance.  Consider joining all columns at once using pd.concat(axis=1) instead. To get a de-fragmented frame, use `newframe = frame.copy()`\n",
-      "  df_forecast[name] = yhat\n",
-      "\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
       "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
       "INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.\n",
@@ -1762,7 +295,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "16160f58de794b048e7d14108e7553f7",
+       "model_id": "f19524ee3e9146f19ad620f792f60592",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1777,13 +310,13 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 2.68E-01, min: 3.98E-01\n"
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 9.36E-03, min: 1.22E-01\n"
      ]
     },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "af53cc0bb3664f85818ef12cb4e022c5",
+       "model_id": "ef7414cdcd5f4336862c0a79f860e6c7",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1798,9 +331,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 1.00E-01, min: 5.91E-01\n",
-      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 1.39E-01\n",
-      "Epoch[627/627]: 100%|██████████| 627/627 [00:03<00:00, 162.89it/s]\n",
+      "INFO - (NP.utils_torch.lr_range_test) - lr-range-test results: steep: 9.36E-03, min: 1.00E-01\n",
+      "INFO - (NP.forecaster._init_train_loader) - lr-range-test selected learning rate: 2.06E-02\n",
+      "Epoch[627/627]: 100%|██████████| 627/627 [00:04<00:00, 137.84it/s, SmoothL1Loss=0.0015, MAE=1.45, RMSE=2.07, Loss=0.0012, RegLoss=0]  \n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 91.139% of the data.\n",
       "INFO - (NP.df_utils._infer_frequency) - Defined frequency is equal to major frequency - MS\n",
       "INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to 90.0% of the data.\n",
@@ -1878,17 +411,9 @@
       "\n"
      ]
     },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "MAPE with lagged regressor:  57.026634304236026\n",
-      "MAPE without lagged regressor:  12.331045940870737\n"
-     ]
-    },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhwAAAG8CAYAAACVCnwPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8o6BhiAAAACXBIWXMAAA9hAAAPYQGoP6dpAACkMUlEQVR4nO2dd2AUZf7Gn63JpjdIKCEBQpEi0ixYsCBIsdcTT8Re7uSKp56e5fRnO8vJnVc89UBP9Ozl7Ao2bAiKIp1AQggkgfS6dX5/bObdd2Znd2d2Z7bx/fzDZLOZmXc35H32+TaTIAgCCIIgCIIgDMSc6BsgCIIgCCL9IcFBEARBEIThkOAgCIIgCMJwSHAQBEEQBGE4JDgIgiAIgjAcEhwEQRAEQRgOCQ6CIAiCIAyHBAdBEARBEIaT9oLD5/Nh165d8Pl8ib4V3aG1pSa0ttSE1paapOvaUnFdaS84CIIgCIJIPCQ4CIIgCIIwHBIcBEEQBEEYDgkOgiAIgiAMhwQHQRAEQRCGQ4KDIAiCIAjDIcFBEARBEIThkOAgCIIgCMJwSHAQBEEQBGE4JDgIgiAIgjAcEhwEQRAEQRgOCQ6CIAiCIAyHBAdBEARBEIZDgiNFqKysxKOPPpro2yAIgiCIqLAm+gbSmeOPPx6HHXaYLkLh22+/RXZ2duw3RRAEQRxUNK9uQeO7TbA4LBh0Zhnyxucm5D5IcCQQQRDg9XphtUZ+GwYMGBCHOyIIgiDSjfb17aj5Zy0AIG9ibsIEB4VUDOKSSy7Bp59+iqVLl8JkMsFkMmH58uUwmUx49913MXXqVGRkZGD16tWorq7G6aefjtLSUuTk5GD69On46KOPJOeTh1RMJhOefPJJXH311cjJycGoUaPw5ptvxnmVBEEQRLLj7fWyY0uWJWH3kbIOx7Rp09DQ0KDquV6vFxaLPi9yWVkZ1q5dG/F5S5cuxbZt2zBhwgTcddddAICNGzcCAG6++WY89NBDGDFiBAoLC1FXV4d58+bhnnvuQUZGBp555hmceuqp2Lp1K4YNGxbyGnfffTduuOEG/PWvf8Xf/vY3LFy4ELW1tSgqKtJlrQRBEETq4+3mBIeDBIdmGhoaUF9fn+jbCEl+fj7sdjuysrJQVlYGANiyZQsA4K677sLJJ5/MnltUVIRJkyaxr++++2689tprePPNN/GLX/wi5DUWLVqE0047DRUVFbj33nvxl7/8BWvWrMEpp5xi0KoIgiCIVMPb62PHFkfiAhspKzjETVwNejscsTJt2jTJ111dXbjzzjvx9ttvY9++ffB4POjt7cXu3bvDnmfixInsODs7G3l5eWhqaor5/giCIIj0QRJSyU7ctp+ygkNNWAMAfD4famtrUVFRAbM5OVJW5NUmN9xwAz788EM89NBDqKqqgsPhwDnnnAOXyxX2PDabTfK1yWSCz+cL8WyCIAjiYMTbw4dUyOFIS+x2O7xeb8TnffHFF7jkkktw5plnAvA7HjU1NQbfHUEQBHEwIHE4EpjDkRwf+dOUyspKfPPNN6ipqcGBAwdCug+jRo3Cq6++ivXr1+OHH37AhRdeSE4FQRAEoQsShyOBVSokOAzkhhtugMViwbhx4zBgwICQORmPPPIICgsLMWPGDJx66qmYM2cOpkyZEue7JQiCINIRaUiFqlTSktGjR+Orr76SPHbJJZcEPa+yshKrVq2SPHbddddJvpaHWARBYPkpIm1tbTHdL0EQBJF+iCEVc4YZJospYfdBDgdBEARBpDFiWWwiwykACQ6CIAiCSGvEkEoiwykACQ6CIAiCSGsCgiOxWz4JDoIgCIJIY8QcDks2ORwEQRAEQRiAz+2D4BYAUEiFIAiCIAiDSJamXwAJDoIgCIJIW7w93OA2qlIhCIIgCMIIJA4HCQ6CIAiCIIwgWQa3ASQ4CIIgCCJtSZY5KgAJDoIgCIJIWyhp9CDC5/PhT3/6E6qqqpCRkYFhw4bhnnvuAQBs2LABJ554IhwOB4qLi3HllVeiq6sLAPDBBx8gMzMzaD7KkiVLcOKJJ8Z7GQRBEEQKkiyD2wASHIbz+9//Hvfffz9uu+02bNq0Cc899xxKS0vR3d2NOXPmoLCwEN9++y1eeuklfPTRR/jFL34BADjppJNQUFCAV155hZ3L6/XihRdewMKFCxO1HIIgCCKFkDgcCW78lbLTYqdd4UNDi4onCoDXOwQWCwCTL+LTI1FWBKx9Qp1O6+zsxNKlS/HYY49h0aJFAICRI0fimGOOwRNPPIG+vj4888wzyM7OBgA89thjOPXUU/HAAw+gtLQUF1xwAZ577jlcdtllAICVK1eira0NZ599dszrIAiCINIfSVlsgh2OlBUcDS1A/X61z07MMjdv3gyn04mTTjpJ8XuTJk1iYgMAjj76aPh8PmzduhWlpaVYuHAhjjzySOzduxeDBw/GihUrMH/+fBQUFMRxFQRBEESq4u3xsGMSHFFSVqTyiQLg9XpgsVgBUxyvC8DhcMR0renTp2PkyJH473//i2uuuQavvfYali9fHtM5CYIgiIMHcTQ9kPgqlZQVHGrDGj6fD7W19aioqIDZHN+UlVGjRsHhcGDlypW4/PLLJd875JBDsHz5cnR3dzOX44svvoDZbMaYMWPY8xYuXIgVK1Zg6NChMJvNmD9/flzXQBAEQaQu0sZf1IcjbcnMzMRNN92EG2+8Ec888wyqq6vx9ddf46mnnsLChQuRmZmJRYsW4aeffsLHH3+MX/7yl/j5z3+O0tJSdo6FCxfiu+++wz333INzzjkHGRkZCVwRQRAEkUokU5VKyjocqcJtt90Gq9WK22+/HXv37sWgQYNw9dVXIysrC++//z6WLFmC6dOnIysrC2effTYeeeQRyc9XVVXh8MMPx5o1a/Doo48mZhEEQRBESpJMjb9IcBiM2WzGrbfeiltvvTXoexMnTsSqVasinuObb74x4tYIgiCINIcafxEEQRAEYTjJ5HCQ4CAIgiCINCWZqlRIcBAEQRBEmuJLoqRREhwEQRAEEWd2PFyNVZM+xb43Gwy9Dp/DYc6ksliCIAiCOGjweXzY/mA1+vb0ofrPOw29lqff4bBkWWAy69D9MgY0C46tW7fi0ksvxcyZM3H66afj9ddfB+BvsPXwww/j+OOPx+zZs7FixQq975UgCIIgUh5nowuCWwAA9OzqhSAIhl3L1+9wWByJ9xc0l8XefvvtmDVrFp588kls27YNV155JSZNmoS1a9di3bp1ePXVV9HV1YWrrroKo0aNwuGHH27EfRMEQRBESuLc18eOPZ0eeNo9sBXYDLmWGFIxJzh/A4hCcOzbtw9z5syB2WzG2LFjUVlZiZqaGrzzzju46KKLUFRUhKKiIpxxxhl4++23FQWHy+WCy+WS3ojVCrvdHv1KQuDz+ST/phO0ttSE1paa0NpSk2RcW8/eXsnX3TXdyDs0T9M51K7L0x0IqRj1GqgdG6JZcJx//vl49913cdlll2HLli1obGzExIkTsXPnTowaNYo9r6qqCqtXr1Y8x7Jly/DEE09IHjv33HNx3nnnab0d1dTV1Rl27kRDa0tNaG2pCa0tNUmmtbVsbJV8XbtuN/Lyc6M6V7h1CYLAHA6vxYPa2tqorhGJ4cOHq3qeZsExY8YM3HHHHfj3v/8NwN+6u6SkBL29vZJR69nZ2ejp6VE8x+LFi7Fw4ULpjRjocNTV1aG8vDzuw9uMhtaWmtDaUhNaW2qSjGvrc7oA7Gdf5/TloqKiQtM51KzL5/Jhi3c7AMCR79B8Db3RJDja29vxq1/9CrfddhtOOOEE7Ny5E7/85S9RVVUFh8OB7u5u9tzu7m5kZWUpnsdutxsiLsJhNpuT5pdNb2htqQmtLTWhtaUmybQ2Z4NT8nXf7r6o7y3curx9gZJYa7Y14evXdPU9e/YgMzMTs2bNgsViwahRo3DooYdi3bp1GDFiBHbs2MGeW11djREjRuh+wwRBEASRyjj3SQVHb11viGfGBt/W3JwEVSqa7qCiogJ9fX345JNPIAgCdu7cifXr16Oqqgpz587Ff/7zH7S2tqKurg6vv/465s+fb9R9pwTHH388rr/+etx4440oKipCWVkZ7rzzTslzdu/ejdNPPx05OTnIy8vDeeedh8bGxrDn3bNnD372s5+hpKQE48ePx+GHHy4Z8PaPf/wDI0eOhN1ux5gxY/Cf//yHfe/CCy/E+eefLzmf2+1GSUkJnnnmmdgXTRAEQYSlj6tSAYCe3QYJjiQa3AZoDKnk5OTg/vvvx1//+lfcfvvtyMvLw4UXXogjjjgC06dPR11dHc4880zYbDYsWrSISmIBPP300/jNb36Db775Bl999RUuueQSHH300Tj55JPh8/mY2Pj000/h8Xhw3XXX4fzzz8cnn3yieL6uri7MnDkTQ4YMweuvvw6v14vGxkaWffzaa69hyZIlePTRRzFr1iy89dZbWLx4MYYOHYoTTjgBCxcuxLnnnouuri7k5OQAAN5//3309PTgzDPPjNfLQhAEcVAiCAL6ZCGV3t3+Xhwmk76NuSSD27JTTHAAwFFHHYWjjjoq6HGz2Yzf/va3+O1vf6vLjUVi9YlfwdXkjPg8AYDX68VOSy30eCvtAzNwzKrg9Yfi0EMPxR133AEAGDVqFB577DGsXLkSJ598MlauXIkNGzZg165dKC8vBwA888wzGD9+PL799ltMnz496HzPPfcc9u/fj2+//RYFBQWora3FzJkzWWzuoYcewiWXXIJrr70WAPCb3/wGX3/9NR566CGccMIJmDNnDrKzs/Haa6/h5z//OTvnaaedhtzc6LKkCYIgCHV4Oj3wdnslj3m7vXC3uGEv1je3MaUdjmTC1eRE377IgkPEA4+BdxOaQw89VPL1oEGD0NTUBADYvHkzysvLmdgAgHHjxqGgoACbN29WFBzr16/H5MmTUVRUpFhTvXnzZlx55ZWSx44++mgsXboUgL8a6LzzzsOKFSvw85//HN3d3XjjjTfw3//+N+a1EgRBEOHp26u8b/XU9uovOHq4SbEkOKLHPjBD1fNEh8NisejmcGjBZpN2jzOZTDE1X3E4HFH/rMjChQsxc+ZMNDU14cMPP4TD4cApp5wS83kJgiCI8PD5G+YMM3xO/37QW9eLgin5ul5LElJJ8Gh6IIUFh9qwhs/nQ21tLSoqKhJeEiTnkEMOQV1dHaulBoBNmzahra0N48aNU/yZQw89FE8++SRaWlpQUFCgeM4vvvgCixYtYo998cUXkvPNmDED5eXleOGFF/Duu+/i3HPPDRJGBEEQhP7wFSr5k/PQ+nUbAH8eh95IQipJIDiSawc+yJg1axYmTpyIhQsX4rvvvsOaNWtw8cUXY+bMmZg2bZriz/zsZz9DWVkZzjjjDHzxxRfYvXs3XnnlFXz11VcAgN/97ndYvnw5/vGPf2D79u145JFH8Oqrr+KGG26QnOfCCy/EP//5T3z44YdBTdgIgiCSAW+vF/s/PiD5pJ7q8A5H4RGF7Lin1mDBkWplsYS+mEwmvPHGGygsLMRxxx2HWbNmYcSIEXjhhRdC/ozdbscHH3yAgQMHYsGCBZg7dy7+9Kc/wWLxq9czzjgDS5cuxUMPPYTx48fj8ccfx7Jly3D88cdLzrNw4UJs2rQJQ4YMwdFHH23kMgmCIKJi/dU/4ttz1mH9VT8m+lZ0g8/hKDyigB0b0YuDT05NBocjZUMqqYBSaevrr78u+XrYsGF44403NJ23oqICL7/8cshw0TXXXINrrrkm7DkOOeQQQ0ciEwRBxErbt20AgNY1bQm9Dz3pawg4HPkT82B2mOHr9cXB4Ui84CCHgyAIgkhKvL2+/n/TKaTidzhMFhMySjOQVe4vBOit69X9QyDlcBAEQRCECsQKDm+PN20cWTGHI2OgHSaLCZmDMwEAvj4fPB36tm9ItioVEhwEQRBE0iH4BCY4IPg35FTH5/bBtd8FAMgY5Bca1txAZoOnS2/BkVx9OEhwEARBEEkHExv9pENYxdno9DeHApA5yN/TyZITEAKeTn3XSDkcBEEQBBEBfrQ6kB6Cg++OnSk6HDkBh8Oru8PBh1QSv90n/g4IgiAIQoY8hMKHB1KV7upudpw52O9w8ILD02Wgw0E5HARBEAQRjFcuONLA4Whe3cKOC6YVAACsfEhFb4eDQioEQRAEEZ50y+EQBAHNn/kFhznTzASHJce4pFFXsxuAvwTXnJH47T7xd0AQBEEQMuQCI9Xbm/fs6kFfvb8ktvCIQlj6BQDvcMjH1seCq9WFrq1dAIDc8TkwmfQYXxobJDgIgiCIpCPI4UhxwSG6GwBQclwRO7ZmG+NwtHzRyipiio8pCv/kOEGCgyAIgkg6gpJG4xBScbW6dA9riDR/HhAcxccGBICkLFbHpFE+X6SIBAdBEARBKCMvi/X1Glul0rmlC6smfIpVEz5F396+yD+gAcEnMAFgzbMib1Ie+55RZbFMcJiBoqMKwz85TpDgIAiCIJKOeIdUmt5r8rcX7/Sg8d0mXc/dubkLrgP+DqNFMwphtga2XmmnUX3W6DzgQtdmf/5G/qQ82PJsupw3VkhwEARBEElHUNKowSEVZ78gAMCSLfWi+fNmdlx8XLHke0aUxbZ8wYVvkiScApDgIAiCIJKQeJfFijNOAKBra3eYZ2qne0cPOy6cmi/5ntWAsthQ+SKJhgQHQRAEkXQENf4yOKTi4hyOzm36Ohy8eOL7bvi/5spidQqpiILDZDWh8IjkyN8ASHAQBEEQSYhPPkvF4NbmfEjF1eSCq8UV5tna8LkC9y5vwGW2mmHO9D+mh8Ph7fGie4ffock7NE/ioCQaEhwEQRBE0hHv1uZ8SAXQN6zCV9yY7cHbrigK9Ega9XQHREvGQHvM59MTEhwEQRBE0hHPPhyCT4CrWS449Aur+FwCO7ZkKgkOf1hFD4eDf90smYmfn8JDgoMgCIJIOoJCKgYKDne7G4JHkDymq+DgcjiUHA4xr0OPPhz862RWEDeJJLnuhiAIgkgZXC798hzkeONYpSIPpwD+RmB64XOGFwGiw+FzCZJ8j2jwksNBEARBpBMXXHABCgsL8eqrrxpyfnlnUT0Hm8nhK1RE9MzhYA6HyV85IkdSGtsdm8vBO0NmR3Jt8cl1NwRBEETS09TUhBdeeAE9PT1YtmyZIdfwOuUhFeOqVJwKgsPZ6IS7za3L+UXBYc4wK05tlbY3j01Y8a8TORwEQRBEStPa2sqOe3p6wjwzeuKZNMqHVCzZgU26S6d+HLzgUMKiY7dRviJGKUE1kSTX3RAEQRBJT3t7Ozvu69N30JlIPMtieYeDH3SmVx6HN4LgsGbrNzGWF2pmBzkcBEEQRAoTD8Ehr1LxGdhplHc4io4OtALv2qZPHoeYCGoJJTh0bG/OCzNyOAiCIIiUJjEOhw+CIIR4dmzwSaP5hwZGx7ubdc7hUCiJBaTtzj2dsSaNcg4HCQ6CIAgilUmEwyF4BUkDLT3hQypZwx3sONbNX4QJjhACwKrjPBW+nJiSRgmCIIiUJi6CwxlcleIzKI9DDKlYcyzIGJDBHtdDcAiCENHh0DOk4qPGXwRBEES6EJeQikIZrFETY10HnAAA+wA7zA4zTBZ/6apbD8HhFoB+YyZk0qhEcMTocPCNvyhplCAIgkhl4uNwBG+8RlSq+Fw+uNv8wsJekgGTyQRrro6zTfi25hnKAkDPslgflcUSBEEQ6YJccBiRzClPGgWMERz80LaMEv90VWtu//RWHRwOrzP0aHoRaeOvWKtU+KRRcjgIgiCIFIYXHD6fDx6PPsmVIoIgBDX+AozpNspXqNgH+AWHWDWih+DgHY7QZbGcwxFjC3dJ4y9qbU4QBEGkMrzgAPQPqygljALG5HDwFSr2fofD1u9w+Hp98LljEzn8MDZ1ORx6lsWSw0EQBEGkMOkkOPimX/KQCqCDAFARUrHoOkuFcjgIgiCINMFowREqdGJEDodTIaTCC45YBYBEcIQsi9UzaZSqVAiCIIg0wXiHQ3mTNyRpdH9wSEWsUgFiL43l1xKqL4bJbGJD42Ivi6U+HARBEESa0NHRIflad8HBfUo32QLj3A0JqfBVKgoOR8ytxlU4HEDA5Yh9lkr/9UyhQziJIrnuhiAIgkhqnE4nnE6n5DHdQyqc4LAV2gKPG12lopTDEaPgUFMWCwTyOGItixX7cJgzzTCZTBGeHV9IcBAEQRCqkYdTAGNDKnaJ4DDA4eAFR5GxDkeoslggMKLe0+WNqa+JKNbCXStRJN8dEQRBEElLNILD4/Hgtddew48//qjqGryTIXE4DAipuNv8E2GteVbW0lxXwaGiLBYIlMYKHiFklY6q6/WJg+KSK2EUIMFBEARBaCAawfHvf/8bZ511Fo488ki0tLREvAa/4doL7ezYEMHR7hcUtoKAsJH0xeiMsUqlT1tIBYgtcVRMGk22pl8ACQ6CIAhCA9EIju+//x4A0Nvbi23btkW8Bl9pYSsyLqQiCAJzOGwFgQ2fr1KJn8PBj6iP/prkcBAEQRBpQTSCg/++2+2OeA3eFbBzgsOnc9Kot9sLwePPl7DlcQ6HniEVlQ6HHt1GBUFgoizZmn4BJDgIgiAIDUQjOPiqFjWCQ+JwFBjncLjbA/dilTgcOnYadWkriwWin6ficwlAf75psjX9AkhwEARBEBqIh+DgXQFJSEXnHA5xLD0gy+Ewqg+HyhyOaEMqviRu+gWQ4CAIgiA0EGtIRc1kWUkfjgLjBIeHcziMEhxeSVlsaNfBqkPSKF/dY6EcDoIgCCKViYvD4eRHrFtg7q+40D2kwjsc+QZVqahNGuUTVTt0cDioSoUgCIJIZeKSw9Er3aTFfAT9BQfvcAREhsnCzTaJU9Ko2HQMkLZb1wLvDJHDQRAEQaQ0calScUonnlqy+gVHj75VKu4QIRUg4HLEqyzWXswJjpbIr5ESyTyaHiDBQRAEQWgg3lUqcXM48q2S77FhavFyOIoDgodvtx71tahKhSAIgkhl4l2lYsk0BxyOOOVwAIHEUU+nJ6bZJrzDEW6+icThiDqkQg4HQRAEkSbEu/GXOdPC2nQLbgE+t35hFWkfDmXBAcHfICxa1JbF2gpsbEeOVnD4JLkv5HAQBEEQKYwoOKzWQAjCyJCKJdMsaWKlp8shzlEBFHI4cnWabeKUhodCYbKY2GRcV3OUORz860ZVKgRBEEQqIwqOgQMHsseMdThkgkPHxFFPuBwOnXpxqO00CgTCKlE7HDJnKNkgwUEQBEGowuVyMfFQWlrKHtfd4XBKyzvFHA4A8PbElsTJIyaNWrItMNuk26FugkNl0igA2PoFh7fbK3Er1EI5HARBEERawOdvqBUcgiBoTxoVwyYmwGQzwcLPGYkhvCFHzOGQJ4wC+jscJpsJJrMp7HMllSpRhFUkfTioSoUgCIJIVXjBUVhYCJvNv0GGExxut1tS5aGqtXm/w2FxWGAymSSTXPl25LEiVqnwTb9E9BpRLzoc4SpURGKtVPH10iwVgiAIIg3gBUd+fj4yMzMBhBccvLsBqM3h8G+cYgjCmhcQBO4o237L8fZ6WQWJPGEU0N/hiBROAWIXHNRplCAIgkgLOjo62LGxgqPfFej/lG7jBEe0c0bkSCpUlEIqOToJDqcoOCILAGlIJQqHg2apEARBEOlANA6H/HvqymL7N+l+wWHN19/h4LuMWhVDKvoIDi8THOHzNwC5wxFFDgdNiyUIgiDSgbiFVMRNun/TlORwdOiTwxFujgqgY0hFXEuEklhAKjjcUYVU0rBK5emnn8b8+fNx3HHH4cILL0R3dzcAYPny5Zg1axZOPPFELF26NKZ2sARBEERyEQ/BIQgC2zgtCjkcuoVUwvTgAOSCI/ZOo2r6YvCCwxlVSCW5Z6kEv8oRePHFF/HVV1/hqaeeQmlpKXbs2AGbzYbVq1fjpZdewvLly5GZmYnrrrsOFRUVOOOMMwy4bYIgCCLehBMcgiDAZAoOG2gNqQhuAejfN8VN02ZASMUTZo4KIKtS6YrumoJXgODxf/BW53AE7sMdVUgluR0OTYLD6/Xi3//+N5588kmUlZUBAEaNGgUAeOedd3DmmWdi6NChAICLLroI//vf/xQFh8vlgsslVW9WqxV2uz3oubHi8/kk/6YTtLbUhNaWmtDagLa2Nnack5PDBIfP54PL5WJlsjy9vb2Sr10uV9jreHoDm7s5wwSfzwcz14fD3ebW9B6EWpuzNbAHWfOtQd83Z3PX7NR2TRFeAIhrCQefS+I84Az7fKV18SEVk4rr6YXZrE7caBIcTU1N6Ovrw0cffYTnnnsOOTk5+PnPf44zzzwTu3btwpw5c9hzq6qqUF1drXieZcuW4YknnpA8du655+K8887TcjuaqKurM+zciYbWlprQ2lKTg3lt9fX17Linp0cSNt+2bRtycnKCfmb37t2Sr9vb21FbWxvyGp7mgOBwep2ora2Fl2v21dXUFfbnQyFf24Hdzey4xdUCd630QzA/dK17f3dU1/R2BO7b6XOqOocp0wShT0B3Y4+q5/Pr6mkLiLs9jXtgaomcqKoHw4cPV/U8zYKjq6sLu3fvxptvvom6ujpcc801qKysRE9PD7Kzs9lzs7Ozg5StyOLFi7Fw4ULpjRjocNTV1aG8vFy1CksVaG2pCa0tNaG1+V1ukbFjx6KgoIB9PXDgQAwYMCDoZzZv3iz52mazoaKiIuQ1es292I6dAICcohxUVFRA8AnYZqoGBMDqtob9eTmh1taDPhyAX3QMGTUEhRUFkp8TBAHbbNUQ3ILma4o4G53YBv8H7+yCbFXn2DVgN/rq+oAOIezzldZVL+xDLwCYgYqRFYohrkSiSXBkZGQAAK644gpkZmZi1KhRmD17Nr744gtkZWWx5FEA6O7uhsPhUDyP3W43RFyEw2w2p90fCRFaW2pCa0tNDua18X04CgsLJX/jXS6X4s/Kcza8Xm/YawjOgGtiybT4n2v2J4562j1wd3iiev3la/NwfTgyCu2K57TmWuFuccPTFd01BW7plgyLqnPYi+zoq+uDq8UNEyK3Q+fX5et/7SwOCyyW5Esa1fQKVlRUwGazSVSTeDx8+HDs2LGDPV5dXY2RI0fqdJsEQRBEouGTRvPy8lgOBxC6UkVrlQofyuCbV4nNv4yoUrEqJI0CgeZf0V5TshYVnUYBIKOk/8O4T3qPqq4nVvckYcIooFFwOBwOnHTSSXjqqafgcrmwa9cufPjhhzj66KMxb948vPrqq9izZw+am5uxYsUKzJs3z6j7JgiCIOKMKDhycnJgsVhUCQ6tVSqS9tzcJm3VW3DwnUYVGn8BgeoYd4cnqjYPPiefNKpuu7UVcd1GD2grjRUbfyXjaHogirLYm266CXfddRdmzZqFgoICXH311Zg8eTIA4JxzzsGiRYvg8/lwxhln4PTTT9f9hgmCIIjEIAqO/Px8ADDE4eBLUPn24mLpqs/pg7fPG3MnTdE9MGeaQ55L7HAquAV4e7ywZmvbMiUOh4qyWACwl3DdRlu0ORzeJHc4NAuO3NxcPPjgg4rfW7x4MRYvXhzzTREEQRDJRzwEh7eb6yXBlabKm3/FKjg8baFH04vw33O3eWISHGpFgL2IExwaHQ6x8VcyNv0CqLU5QRAEoQK3242enh4A2gSH5pCKRHAENni+86cezb/Ec4QKp/i/F1tLdW9UDgcXUmlRLzgEn8AETrI6HMl5VwRBEERSIZ8UCxgUUunmQiqcw6HnxFif28eETViHo0DqcGi+jkt70qjU4VAvcnxJPpoeIMFBEARBqEDe1hyIh8PBhVT48EZ7bAPc+GFsfKhGDi9ytFaMANFVqUhzONQ7HN4kH00PkOAgCIJIG3p7ew0bmhmt4NDucAQ2TknSaJ4+01vlP8+HauRYJQ5HnARHlFUqyT6aHiDBQRAEkRasWrUKJSUlmDFjhiEzNOIlOLxcSCVc0mgseLhW6eEEB5/fEU3eSMwOh4aJsT7e4aAcDoIgCMIonn32WfT09ODrr7/Ghg0bdD9/vEIqEoeDz+GQhFTi43DY9HQ4VCaN2gpsMFn9DTVd+7WEVDiHg6pUCIIgCKM4cOAAO+YTPPVCL4fD4wkvFkJWqUgcDh1zOHJDb868yPFEITi8Tu1hDpPZxFwOZ5MGh6NXe5OxeJOcd0UQBEFooqWlhR13dnbqfv645XB0KVep8E5EzCGVTuXmYnLETqNAlFUqUTgcAJBR6p9b5jrgguBTl5MjcTgopEIQBEEYRXNzYNx6MgkOvapUJJu/noJDbUglCldFUharQQRkDPA7HIJXUN1tVNJGnUIqBEEQhFHwDkdXV5fu549fH47+jdMkzUWw5nHhjVhzOFQnjcaYw9EXncNhH8Alju53hnlmAGmVSnJu7cl5VwRBEIRqBEFIm5CKWKViybJIRrNLemJ06pnDEVpwmO1mWLL8oidejb8AIGNgBjt2NqrL4+D7cFBZLEEQhAIulws//fSTYf0jDgY6OzslyZjJJDjkj3u93rDvtehwWHOkm6bZYWbVG/rmcITfnMVk1ZgdDk2CI+BwONU6HD2c4MgiwUEQBBHE2WefjYkTJ2LJkiWJvpWUhXc3AGNCKno5HEB4l0PM4bDIBqWZTKbAuPg4lcUCgbBKNCKHdzgs0TocKktj+fuz5mueyxoXSHAQBJEw6urq8NZbbwEA/vrXvyb4blIXPmEUMN7hyMvLA2CU4PBvnHyFiojoNujqcEQUHP7ve3u8EgGhBq8zulLVDD6Ho0mdw8GHfGx5oefDJBISHARBJIx333030beQFsgdDiMFR3Z2NqxW/yYcTUgFCC04fC4ffC5/uMWiKDgCbkMsIThJ6a1KhwPQHlaJtizWPoBzOFT24uDny5DDQRAEIeOdd95J9C2kBXKHw8iQihhOAfR3OPg8BGt28KYpJo4KXkHyXK14OpXntSjBuwWaBYcrIIo0lcWWas/h4F0fXiQlEyQ4CIJICE6nEytXrmRfWyzJmeiWCsTT4eAFR0ZG4JO4FsERqtuoJ8QcFRFJt9EY8jjEkIolywKTxRT2uZJ5KhqvKZlvosHhsBXYYLL570u1w8GJIRs5HARBEAFWr14t+STu9XojlkwSyhgtOLxeL3uveMFhsVhgs/k/TesRUvF2KTf9EuEFRyzNv0TBEa6tObsmH1Jp1+pwRNf4y2QyIaO/vbnaPhy8GOJfp2SCBAdBEAlBKX+ju7s7AXeS+hgdUlFq+iUihlX0CKlIB7eFDqkAsY2oF3M4IoVTgBhzOMSyWDNgtmrbbu0DxfbmbgjeyPkqnn4xZM2xaL5WvEjOuyIIIu1Ryt8gwREdRjscSiWxIuEEhyAI2nI4IoVU8qN3G/h7CjgcKgRHDPNURIcjmkZcYi8Of3vzyGEV8fXgX6NkgwQHQRBxp66uDps3bw56nARHdBhdFhut4HC5lDdKdQ5H8CYtcTiiDKl4e7xAv/GgTnDwLdW1iRxxWqzZHj5PRAmtvTjEkEqy5m8AJDgIgkgA1dXVio/39PTE+U7SA6XGX3p2bo1WcCi5G0A4h4PL4VAId1h1EBxaKlQAWUhFa9KoKDiiGBcv7cURXnB4+7wsfJOsFSoACQ6CIBJAU1OT4uPkcESHXHB4PJ6Qm300RCs4QuV1hHY4lEfTi/AlqmqnqAZdQ0PTL0BWpRJlHw5zhvaQip13OCI0//KkQMIoQIKDIIgEsH//fnY8bNgwdkyCIzrkIRVA37CKWsEhd1VicjgUBEdWpYMdd26Kbn3Spl+RhQAfUtEqOEQXRs115EjnqYR3OPh8FhvlcBAEQQTgHY7hw4ezYxIc2vH5fEEOB6BvpYoaweHz+YKEhFbBEalKJeeQHDaYrG1tW+QbV7qGRoeD79qppfeH1+ljDoc1ilbjGVy30UilsXyoh3dkkg0SHARBxB3e4aisrGTHJDi009nZCZ8veMaHUQ6HOEdFJDs7mx3LRY7WkIq3K3yVitlqRv5k//V76/rQ16A9bKRVcFiyLKwJl5bKGE8H5zpEEebgczgijajnr0VVKgRBEBwkOPRDKZwCxC+kwn/NPw+IwuHoCl+lAgAFUwvYcdu6NsXnhIMXHEqJqXJMJhNLxNQSUokkbO55RkDFuT688olycm9GKV+lEsHh4Ae3UZUKQRBEAD6kwguOdKtS+e677/Dwww/jwIEDhl1DKZwCxC+kUlBQoPg8ILTgCNXaXJrDobxxFkwLXL9tbbvic8LBCwGbCocDCORFaOnDIRkXL3M4PB4Bf1wuYHcjcM9/lAWHNd/Kymk15XAohG/6nAKcLv2qlqKFBAdBEHFHdDiys7NRXFzMHk8nh8PtdmPevHm44YYbcMsttxh2Hd7h4IepJYPDwYdU7PZAiCDaKhVAB4eDd1FUCw7/8zydHlVdPwFp63X5dRpbAXf/t6v3QrGE2WQysamxkapUpDkcwYLj+ZVA5iwBA0714fmPEic8SHAQBBF3RMExcOBASQ5AOgmO+vp6NDY2AgA2bdpk2HV4h4Ov+NFTcPCO1IABAyTf4wVHW1ub5Hu8w5GTk8OO1fXhUBYcmWUZcJT7hVX79x3weYLzV8IhDXWoqx6R9OLoUBdWkUxvlTkceznDq6MbaA3xVtnFeSoHXBB8oYWCJ8JoevF6B9oBR0bQt+MGCQ6CIOKK1+tln8oHDBiQtoJj37597JifRaI3vODgw1N6hlQaGhoAAA6HA7m5uZLvqc3hUCM4IlWpiBRMKwDg7xratVnbOrUmjQKAVdKLQ11YJVxIZa8swrZrHxQRB7jBFz5hVeJwKAqOgFgZVBz07bhBgoMgiLjS3NzMLOSDRXAYMS5ehA+pVFRUGHJNcS2DBg2CySRt0602pKLO4fBvnGaHWTI2/qedAmZc48Pv/u53M/g8jlaNeRxak0aB6NqbuztDC459srSbmhCCw1bENTo7EE5wcDkcCiGVvVxe8WASHARBHCzw9nw6h1QS4XAYITicTidaW1sBAGVlZUHfV5s0qsXhkLsbS18W8NVG4KH/ArsbBWkeh8Z+HNEkjYqhDQDo2q7ud1RyHVkiJ+84AKEdDv664Qa48a6LUs8P3lEpI8FBEMTBAl8SK3c40qlKRS449JxtwhPK4dArpCKGUwC/wyFHbUiFD8VEyuGQ9+Co4zrht3QAeYfmwWT1OyCdG7UJK28USaPFRxex48a3ldvyy9EWUlH+3bAXcYLjQGjBwVwXs388vZx9/b8iAwsBm1X7IDm9IMFBEERc4QVHOjsc/Ebt8XhCNsGKFaMdDi2CQ540qjWkIlapyCtUmloDxz19gCXDzHIV3BqHuImhDpPFBLND3RZYdFQhcxv2rzzgnzgbAUkzrtzwgqOmAYrYS7hk1TCzY8TXwJZvg8ksFRQ+n8AEx+CSSHdtLCQ4CIKIK/KKB4cjMB8jnQQH73AAxoVVjK5S4dehFFLRK2lU8Arw9fpzNOQ9OPa3BY57+z/oi5s4H7pQgzhLxZprCcpHCYXJYkLpXH91jrfHi/0fR+6r4g5TpSLP4QgZUuEdjuZwIRX/66mUMHqgHfD066NE5m8AJDgIgogz8pCK2WxmooMEh3bE0tv8/HxJPkWqhVSkFSoBh0MQBDS1BZ7X02+a8IJDS7hKFChqE0ZFyk4tZccN/2uMfB0NIZWaBuVeHPbiyIJDEAQ240WprTl/rURWqAAkOAiCiDPykAoQmMeRzoLDiEqVjo4O7Nq1CwAwZswYiYsQL4cjLy+POQVqq1SUOo16uaZffA5HZw/g4vRJb7+GEQWH4BbYkDQ1iIJDbf6GSPGxxUw4NL2/Hz5X+GtKBAcnbtweQRIiAvxrkj8GAPZirkqlOXTei9iMTLkkNnBMIRWCIA4qlJpIpZvg8Hq9knUCxjgc33//PTueMmUKLBYLc4vilcNhNpuZe6GfwxHYOOUbcY9McADS7qHh8Hl8LGyjtkJFxGw3Y+Ac/++rp8OD5s+VZ9iwexKFTY5FUuLbEOLHlMIqahwOaQ+OCCWxJYlLGAVIcBAEEWfkIRUgfoKjvb0dixcvxq233mpY1QjgF1XyCa5GCI5169ax46lTpwIIbOx6hVQiORxAIKwSLmmUTw5WEhzSOSoBh4PP3wACIRW+E6naPI5oKlR4SucNZMctXypYEhxiDoe8THUfJwDM3A6sJDhsBTa2S4cSHJG6jO5Lkh4cAAkOgiDijPjJPycnh30aFzejnp4eQ4XA448/juXLl+Pee+/FN998Y9h15OEUwBjB8d1337FjueDQ2+Ewm80sBCZHFByxJI3yIRW+tFMp/ADIHA6VgiOaLqM8OaMDa+hrDD/fJBC6kVbc8I7DoSMDx0qCw2QxwV7oFyyhQioRHQ6u5weFVAiCOKgQHQ5+JkdWVhYAfwKcUeWjAPDDDz+w4z179hh2nXgJDtHhsNvtGD9+PIDAxq53DseAAQNgsSjPHhEFR29vr0RMRBtSkTgcskaiYkjFFoXgcEu6jKqbo8KTMTAQ4nCGERw+j485NnKHg8+pmDE+cFzTEKIXR39YRY3DQTkcBEEQ/Xg8HlbGyQuOePXi2LFjBzs2svtnPARHV1cXtm7dCgCYOHEim8YqbuwulwsuV/ix5pHw+XysCkYpf0MkVKWKlj4coUbTBzsc/o2ZT8QUS10j4e2KzeGwFdpgsvWPjG8K04hL0mVUVhLbHBAWR08M5FSEKo219QsOb7cX3r7gXBXe4VCsUul3VEwmYGBByFuOCyQ4CIKIG/wcFd6ej5fg2L59OzuOt+DQu0pl/fr17LWcMmUKe5x3EmLN42hpaWHiIFT+BhC6vbk2h0N5NP3+Nuknf3lZLBC/kIrJZELGwMgj49WWxI4fDuT6zb0wvTjCV6pEHtzm/7e0ELAmsMsoQIKDIIg4opQwCsRHcDQ3N7OZIICxA9Xi4XAoJYwC0LU0NlKFikiobqOacji6lEMqoapU+D4ank51VSr8ADSrxj4cImJYxXXAxcpR5XjCDG7jcziGlADD+1/W3Y2AV+F8/DwVt8I8FU+YwW1er4CG/iZjiQ6nACQ4CIKII4kUHLy7ARjrcPAbtVHXU0oYBfR1ONRUqAD6hFT4iae8GOCbfgF80qj2KhW+YVfOmOwwzwxNRqnf4YAPcO4PkVcRpsuo6DjYrEBxPjCsv5+Y2xNckQNIu406Feap8IPb5A5HUysgFkuR4CAI4qBCPilWJB4D3OIpOOLpcFitVkyYMIE9zguORDgcoUIqkQRHz+5eduwYmsmOQ5XF8iERr4ocDmejE00f+AVvRmkGSo6PrkaUCQ6EDqu4QzT9AgJlqoOK/SEavvvnPoUeHZHmqfBD3Xg3RH6+RJfEAiQ4CIKQ0dTUhOOOOw5nnnlmyCFb0ZJIh4NPGAXiIzj4OTF6Xq+npwebN28GAEyYMAGZmYENWs+QSqwOhyg4rFYrMjICG7VSp9HeOk5wDAu8bqFCKlqrVOpf2MtCIEMvHAKzNbrtT8zhAEILjlA5HC63wASU6DiUBQbRKguOCBNj+Xvg7w1IrqZfAAkOgiBkLFu2DJ9//jlef/11fPTRR7qeW6mtORAoiwVSP6QiCALbqEeMGAFzf3cnPXNGtmzZwhqLTZ48WfI9PUMqah2OUEmjolvlcDhgs3Gf1BWEbG+tX3DYCqyw9ZeSCoIQ5HAo9eFwRxAcgiBgz4q97OuhC4eEfX44pKWxIbp/hhAcDdzQNtFxGFRsUvy+iKS9uUIOh1gtY8m2BLkpyVQSC5DgIAhCRnV1NTvmBYIeKLU1BxKTw2FU0mhraysrRx08eDDy8vIA6CtweOeBH0kP6BtSicbh4JNGxd+fkpISWK2cQJAJDp/Hh769fiXhqAiIz7auwKRTkUCnUfUOR+/6XvTs9Iuf4mOKkD08K+zzwyFxOEL04pDmcAQEg5IAiBhS4dubH1AIqfQ7HLwQClwvkISa6MFtAAkOgiBk1NXVsWO9XYBEhVQEQYibw8Fv0oMGDWICwCjBIRcCRlWpaA2peDweNDf7d9CBAweGdTj66vtYuMNRHjp/A1B2OLwRZqn0fB8I1ww5f3DY50ZCTQ5HqCqVRs7BKCsy9f8beIzv0SEiERwyh8Pn8rGkUXk4BZCFVEhwEASRCPr6+nDcccdh4sSJqK+vl3xv9+7d7FjvTTlRDkdzc3NQ2+14CQ4jHI5woQ4jqlRycnIkQkaOkuAQxQbgf6/DCY5eLmE0i3M4lCaosuFtGmapeFoDgiS7Knp3A1ApOEKEVLoCy2T9N3jnQTmkElpweFsC61IUHBRSIQgi0axcuRKff/45fvrpJzz//POS7/EOh95hB9Hh4OeoAMYLDrm7AcRfcHR3d8PrVdcvQss15M4DLzhiXaMobMK5G4Cy4JBXJFksFjbGXi44JBUqXMKoksMhhlRMZhMTHZEEh5cTHPwGHg0ZA7gcjhDdRt2dXIkv58T0ck939OuD0sLAY0ohFUuWBWaHf6uWh1Q8BwLrVgqp7Ok3FC0WYECB4q3GFRIcBHEQwm9E/CfR9vZ2icgwKqTCuxuA8WWxoQSHEYPiQgkOID4j40NVjGjF5XKxny8tLQ37XKWkUaUSaNHlCOtwhKlQAaSbtpjHoUlwlMQmOCwOC3Mt1OVwcIKDe7ooODIzTCjs14hKDgfAzVORORye5tAOhyAI2NmfJ1tRClgsVKVCEEQC4Bsy8ZsgH04B9BUc8pg+TyIcDrfbLekToRfxEBz8NeRiQC/BESrfRomcnBxWjSMmjSr9fCjB0VMboiS2LfhaPdxsP9E98PTncAheAT63L+hnxJCKyWYK6vwZDWJYRY3gsIYSHJzuEcMq+5qhKIJFweFudkPwBb7vaeYcjlKp4DjQDnT2a/eRsaWt6AYJDoI4CAklOPhwCqCv4JDH9HmMLovle3CMGjWKHRtRqRIqaRTQ7/UUHY7i4mI2tE1EL8ERqkmbEiaTiQmrWB0OadJoYHO19e/bbg/g8fQPcGOCwwNnoxOrDv0UH0/6FH17pROHvW1+wWEvsrOwTixklAYGqikNjhPLYs0OM8y2wDar5HAAgcTRXifQofDrL85TEbyCRMzwDoddFlKp5lKzRkZfBawrJDgI4iCE/2TPb4JGOhzhPjEb7XDU1NSw40mTJrFjI/I45OEO3uHQ43p8nw+l3IpECA7+urEIDvsAO6zZVng8AlxuQRJSKeduQQyrsPwIAdjz33o4G5xwNrrQ+G7g2oIgwNPi36T5rp2xICmNVWhvLooCeVtzcdItIBUcWhJH+fbmEodDFlKRCI7BiQ+nACQ4COKgRK3DoacDEG4DM1pwiNcuKipCSUkgXd8IwSGKgdzcXGRnZ+suONrb29n7p9SMKzMzk3X15HtiaEVLSAWIXnB4nT70NfgFcNYwB1xuAZMvE5A1W8A7XwfOX8FFjgLtzQOVKp2bAhU5rtbA+T2dHqB/X441YVQkUi8OMafEmisVOJEcDiBULw6+vXlAcHjD5HBUB/qcYQSFVAiCSBS8wxGvkEoiHQ4+WVVvASBHFByiGND7emp6Y8g3/2jQ6nCIiaNOpxN9fX1hczj41uZ9e3qB/g/+jmEOfLcN+GkX4PUGchAy7cBArpqD9eLgmn91bgr8Hrs5wcFXdtgH6CQ4wpTG+tw+TnBYJN/jE16zApEjSbfRiO3NuXkqEodDtradewNuCoVUCIJIGLzDEa+QSqgeHABgt9thsfj/OOtdpdLb28tE1cCBAw0VHN3d3exaSoJDD8dIniOiRCIEhzyUo/R+i91GeYdDXhLbLU2/AABUDZFu0D0Kzb+6tgeEqkRwNAd2+QzdHA6+vblUcHRu6mICKmuEtOdHpKRRQDmkYskOCBdvT8DVEHM4bEU2mO3S7ZwcDoIgkoJEJI2GmqMC+JMORZdDb4dDfl0jBYeSGNA7aVSLwxFL6W+0IRVAKjgKCgpYYqtSSEVeEstvyrOmAZfOA5640YQsLmKgNDFWcAfW6W4LnN/NCQ7dQioSh0Oaw9G2ro0dF04tkHxPXUgl+P2yZHGCo7e/IkcQWB8OpaZfYg7HwEIgN4tyOAiCSBBKIRWfzxckOHp7exUne0ZDpA0sHoJjwIABus4akaPUkEtvgaPF4fD5fFF3G43V4VDquaIkOCQlsRVSwTFnuglP3WzGkeNNkg1aqb05T8iQSow9OETC5XC0rQu4SgVT8yXfCyU4IjocvODo9gsOb5cXQn8Sqjyc0tMnsNBMspTEAiQ4COKgRO5wCIKApqYmxSmeem3K4UIqQKA0Vm/BId844+1wJCKHI9T0Vi2Ir5vJZEJxceRBHLzgaGpqYtflxUokh8NR7gi5KUcKqfC4QoRU9Esa5buNygTHWv+6zXYTcifmSb4X0uGIMMDN4gh2OCRj6WU9OHZy4ZRkyd8ASHAQxEEJ73AIgoDu7u6g/A0RvTblRDkc6SY4tDgcQPSVKnxlDz/pNRS8yOH7nigJDo/Hw0I9THCY/IKjT6H9NwBkZQTCAoGkUWlSpggfUpEIDp0cDnuxHab+zp28w+Fuc6N7h//3N3dCHiwZ0i2WTxrN5G6lIAfI6P9aUXAo5HA4G7nclHA9OMjhIAgikfAOB+B3MeThFBG9BYd8joqIKDicTqduM0f46wLpITjCtTUX0aMXh/i6qQmnANKOp2vWrGHHSiEVIFCp0rvHLzgyyzJgyTCHdAEcEXI4eNytbiZoXM1cSKVYnz4cJrMJtv5z8VUjbd8HXuvCaflBPyeuLcMOmM0BAWUymTCoP49DMaQicTj8nVQlDkeYktiRQ5IjfwMAYu/xShBEyiEXHB0dHRLBMXDgQPYJV69NWTxfqA1MPk+Fz7XQ47qA8WWxkZJG9axSyczMlKyFJ1bB0d3dzZwmtYLj0EMPZccffvghO1ZyOAB/WMVms+H4745Db10fcyV4wcG7AFpCKoJHgLfbC2uOFa79+jscAGDLtcLV5JJ0/mxb28aO86eGFhwOhdsoKwZqGoDmdsDlFmC3BYSCUg4Hn6wqFxySklhyOAiCSCTyGSKdnZ2SkMr48ePZsR6bssfjQUuL/6NbqIoHo3pxyEMq8UoaNbpKZdCgQSHbdMcqOMJVFIVi3LhxTFCE+nmlEfWWTAtyRmWjcHoBAKDXpdyNk9+kIyWNAoGwihhSMVlMsBXo43AAgDW/PzzU6WHzTaQJowVBPyM6M47gohLmcADBLofE4egPqbg4hyOorTnlcBAEkSxECqmMGzeOHeuxSR44cIAdJ1pwGOlwiGIgIyMDhYX+TlV2ux2ZmZm6XM/lcrGZNOFGxseaNKq1JBbwr3PChAlBj4cKqSglKAOhEyslDocYUgmRwwEEKlXEkIqtyAaTWb/wAmtbLvhnuQiCwASHvdiGrMrgsKGYw6EkOPjE0QZZHodSWazE4ZAljYo5HNkOacO0REOCgyAOQuQOR0dHh+TT+SGHHMKO9XAB1HxizsnJ0fWa8mubzWYUFRUhOzubOQNGhVTKysok7oMocmK9XmNjIzsOlb8BxJ40qrUkVmTy5MlBj4VyOEKVW4dqjiUNqUiHtynh6s/jEB0OvfI3RKz5gWt72j3o2dUDd38+R/7UAkX3iYVUlByOMN1GLVmBrZoljYbI4fB6BdT0p/mMHAxdhtXpRdSC48cff8T06dPx5JNPsseWL1+OWbNm4cQTT8TSpUujbjhDEISxKDkc4iaTn58v+VSqx6as5hOzXkPH5IjrKikpgdlshtlsZmEOPQWHy+ViTo5cDOglOJT6fCgR62tplODgq11CORyhqlSiCal4u73w9fmTLPUqiRXhB7O5Ozzo+CkgkgumBOfWCIIQVnAMLAgcH5C9ZZaswLVEh8PVFAgVidNkAaCjxz9RF5A2FEsGohIcPp8PjzzyiMR2Xb16NV566SUsX74cL774Ir788ku88cYbut0oQRD6EU5wGBF2iNSDAzBGcIj9RQDpxqeXAOAJ5z6IAkfseRItaipUAH0Fh9qQCqDN4dAjpGLOMMNkU/4E7251G9KDQ0TM4QAAT7tbkpzqGBIcTnG5AfGtV0oa5RNknbKXxmw3sd3aIyaN9l/PXiINFbm4n83Ud8kxE1WVyquvvooJEyZIOti98847OPPMMzF06FAAwEUXXYT//e9/OOOMM4J+3uVyweWStoO1Wq2s/a2e+Hw+yb/pBK0tNUmGtclDKvv372cb04ABAyT5FO3t7arvNdTa+A2spKRE8Xz8JtnS0qLL69PV1cXE1YABA9g5eYcj1rWJ1NcHmh+UlZVJnicKHLfbjZ6eHsWyYDXw1xg4cGDIe+ETVdva2iKuUb42Ne+XEhMnToTJZGKiymQyobCwkP0873A4nU7F8/Zwv5oZNgG+/oRMfvPscQbu1ZprZaEMx7BM9O72v9+uFhf69gdOZiu26vp/js8fcbW54OSmuFoLg6/VHehvhsyM4N8jG7cb97kC6xaxZFng7fLC2+uFz8cNiCuwSc7FO0Q2a3z+zpjN6rwLzYKjra0Nzz//PJYvX46HH36YPb5r1y7MmTOHfV1VVYXq6mrFcyxbtgxPPPGE5LFzzz0X5513ntbbUU2oHgPpAK0tNUnk2uQD0n788Ud2nJOTI/kwsXfvXtTW1mo6v3xtfCMoQRAUz8d/4q2pqdF8TSX4ypvs7Gx2TnF8e1dXF3bt2qX6DyYQ+n3jX0OHwyG5f3EwHQBs2rQJJSUlqq/Hs23bNnZsNptDvkZ8DkxDQ4Pq11Jc265du9hjHo9H03tRWVnJfr6wsBB79uxh3+OFbm1tLUum5WlpGwjAL8j2N9XB0+PfMFsOWAH4Sy72N3ejtrY/ETkzsDFbR1qB/rf8QG0zekoCv+e9tj5dfqdEOrwBd6xhZyN6awOuYbOzGb210v9jTW0WAP4P5PD2oLZ2v+T7HW0OAH43qKGpFbW1MvctA0AX4OxwomZnDSuP9dm9knXVNAZeJ4+rC7W1Cp3EdGb48OGqnqdZcPz973/Hz372s6Aa+Z6eHsmnouzsbPT29sp/HACwePFiLFy4UHojBjocdXV1KC8v1/RHJRWgtaUmybA2uZ3Nf6KtqKjA2LFj2deCIKCiokLVeUOtjd9oxo8fr3i+ESNGsGOLxaL6muHgcx4qKyvZOfkNv7i4OGQ/C55I7xufBDl27FjJ/fNhhcLCwqjXxr9vEyZMCHmewYMDzRdcLlfE68nXxlcJTZo0SVVrc5Hp06czwVFWVia5tli5A/gdJ6X7MnEv7eiR5SyUksltOSZrNioq/PvNnsK9cO/1C+SyI0rR+bH/OMvnQIG5EHvgrxEtGVGCiophqtcRCWulDY3wi4Z8Wz7gMaEVbQCA8nHlyKnIljzfy+WsFhVkBa19aOBXFVnZhaiokJaX1OTWobe5Fya3CUNKhmALtvufWyw9VzdnjBQW5KCiIgfJgibBsWXLFmzatAk33XRT0PeysrIkv6Td3d0hbUO73W6IuAiHmCyWjtDaUpNErk2ewyFvRc2XVXZ2dmq+T/na+LLYsrIyxfPxm1FHR4cur41YQgr4O2GK5+TDN11dXZL1RiLU+8bncAwZMkTyHF7QdHd3R702PvmWX4+cjIwMOBwO9Pb2or29XfX1xLWJ75fFYkFxcbGm+50yZQpefPFFAH5Rwf8s/3f/0UcfxWGHHYbRo0dj3rx5zAXqdQVCAFmZJtaRM9shQJz73ucM2Ph84mj+pMD76m7zwMN1Ac0sydD1/5u9ILAWb6dXMjAuszj4Wk534P6zMoPDEI6MwPddnuDvW/tLY709Xvi6A6+RNdcqea7bEzhPhk19uCMeaBIc3333HWprazFv3jwA/v+oFosF9fX1GD58OHbs2IGZM2cCAKqrqzFy5Ej975ggiJjw+XxBOVS8EyBvjpWIpNFo53+ova4RvTjCzTjRq+RXS/VIfn4+ExzRXkcuGNTAJ47K75FPGn322Wfx7LPPIicnR/IeiDkI8vbfktbmfDfSwX4LxGw3IX9S4H11t7nhPBD4PbeV6FsWa+OSRt0dnoDgMEGxwViocl+RDO5H+lzB3xebf/l6fXC3B9w0eaWOi6s2ztB3yTGjSXCcddZZmD17Nvv64YcfxuDBg3HJJZfghx9+wH333Yc5c+bA4XBgxYoVOP/883W/YYIgYkMuNuQMHDgQVquVfULWY0OuqakB4B8EphS3B/SZcCon1Aatt6CSX4ufKyK/XrTj4oGAw5GVlSUJYSuRn5+PhoYGza8lX9mjpUJF5KijjkJpaSkaGxtx/PHHS77HCw6R0aNHS3pFhGr/bbOaYLMKcHukm3fVb0fC5/Rh4JyByCjNgDnTDF+fz1+lcsDAKhWuLNbT4WYTam35VjbYjSdU9Y2IpEpFSXBwzb9cXNOvIMHBRUvtqSw4MjMzJX8sRNsuNzcXxxxzDM455xwsWrQIPp8PZ5xxBk4//XTdb5ggiNiQh1PkiBtzXl6eLoKju7ubJQ6OGTMm5POMKIsNJTiMcDj4cIc8KVSvdupahID4enZ2dsLn86l2Kjo6Opgo1dKDQyQ3Nxfr16/Hrl27cOSRR0q+F0pw8ITrVeHI8PeY6OF+hXPH5mDqMwFXxVZgg7PBaXhZrI1r/OVu98DdX6ViK1K+TiTBwbsR8rJYQCo4+KZfcsHhTBfBIefOO++UfL148WIsXrw4llMSBGEw8pJYOeJmlpeXh8bGxpi7fvL5IfLNhYfflPUSHKEajvGCQ6+upuK18vPzg3LU9AipeL1elluhRgiIjpEgCOjs7JQIunBEM0dFTllZmWJjMqUx96EEh1IPiawMoKNbGlKRYyvsFxztHvTt63+iCZLmWHrAOxzuVjcLc4Sa1yIVHMEOSAbvcEQSHI284JC2d+cdjowQPUoSRfJkkxAEERe0OByA/xNvLM2q+FLOcILDYrEw0ZHKDoeS+6BHSKW5uZm9D2qEQLQ5MdE2/VKDKocjzLwRsWKlN4zgsBf6r+Ht8aJzk1/cZVTZFcMcsWDNsQL9p+zZ3SvmaYYUNr1cmCQqh4Mb4NbXGNrh4HM47Ek2D54EB0EcZIRzOEwmEyuBFDdlr9cbssRdDWoFBxDYJPVOGrXZbJINWG/B4Xa72T1HEhzROhxaB6pFG6Liq22idThCoUZw9IURHOJjYR0O3mHoL+bIOiy6RmvfbRWw+D4fPlsfLLhNZhPb7Pv2BES8rVCNwxH8fYngiJDDIQmp5MhCKtzPplVIhSCI1COcw1FcXMxsb3liZVZWVlTX0yI4CgoKsGfPHt1DKgMHDpQkJuqdNBppGq4eIRWt802iFRxq26dHg5LgGDVqFDt2ewR4/f2slB0OUXD0+UNFSoPJlEIajigEhyAImHqFX2h8sUHAtucUrpVvhafDA8EbECQhHY5IVSpRh1RSp0qFHA6COMgIJziMKB3lBUdVVVXY54qbZG9vb8hZG2oJV22ht8MRyX3QI6SSCMERbkBcNMgFh7znS6RNWRQhgiDNVZBcQ8FhiMbhWLslcLx9j/Jz+DwOdn3Dkka5RnqNqVmlQoKDIA4ywoVUjMhzEAVHeXl5RJdEz0qVtrY21v0zmQSHHiEVLUmjQPQhFXl5b6zIBUeohFEgfA4HEDqsInc4HMMyYSvVvvP+d6U0jKKUx8T34hCxGxZS4ebQhEkadUqSRhVvJWGQ4CCIg4xwDkcowRHtJtnc3IyWlhYAkcMpgL6CI9wndb2bjEUSHHqHVLTmcGhZYzwdjnCCI1SVitJzJdeQbfiFRxYqPzEMPp+AFz+WPub2BD9P2eEIlTQaECxKgsNiMUEcuaOcNBrYrj0dYRp/kcNBEESyEE+HQ0v+BqCvEAiXi8C3UW9tbY3pOkD6hlSMThrV6nBIuo2G0M1yh6HwyAIttwgA+GojsEc6W01R4Njygnd0dWWxytcVRZZip9EsS/CDiNCHI8myNElwEMRBBu9w8FNMAelmqUdipVbBoWe3Ub7VuPyTel5eHks4jIfg4LuCJnvSqBhSKS4uVkzyjIVIgqMvQuloNCGVwiMKtNwiAOCFVcHhEyXBoeRw2NXkcIToQSaGQCIljUoez5H14eCTRuM7siwiJDgI4iCDFxzyjpjJ5HDoGVKROxxms5mJm3gIDovFwvJXkrksVhAE9rrpHU4BdHA47MrPlVyDczjsxTZkjwrfAl6O1yvgpY+DH1dyHfhuo+yaKqpUspS7+wcEh0qHw5Rpgtkq3cZdbq5ihhwOgiASCR9SkW9c6SQ4wjkcQCCsEg/BAQQco1hDKnl5ecjICOHJc0TjFnV0dDBBaoTgkHcalQ/4jNQcS+JwhAipZJYFfrDwqELF0tlwbNkNNLQEP67W4Yi2DwcQweFwBAsOc3bwFp7Mrc1JcBDEQQbvcMg3R73bf4uCw2q1orKyMuLz45XDAQQER1tbW0ydVAF1gkNMHI01pKI2ryIa8WZkwigQ7HDIB/lJww7BQiErM/BYqJBKRmkGRv2+CkVHF2L0zeHLsJU4EOKlUnY4pOsx202wZCuHPlQJjn4HR21IxaIgOFxJXKWSZIYLQRBGEy+Hw+fzYfv27QD8n2SV5mjISYTD4fP50NnZKVmvVkTBkZ2dDYdDueeD6HBEIzhcLlfYTqZK8OtRK96MLIlVcx96hFQAYNQNIzHqBr974vP5NNwh0Bbi7VHjcNgK7SEdlUjuDRCFw5GjIDiotTlBEMlCOIcjlOCIxm3Yu3cva4nOd5MMhxFJow6HQ1FM6FmpEm6OiogoONxuN5vGqha+k6lah8NqtbJk1WRxOHgxpnT+iGWxKkIqsdLGRbzyA9XMEsEgEiQ4wgyI05Q06gru+2FVcE4UQypJ3NqcBAdBHGSEEhxWq1Wy4fNhCHG8vBZ2797NjtWEUwBjkkbLysoUP3XqJTi8Xi+am5sBhBccsfTi0FqhIlJUVAQArBdKJHiHwwjBsWDBAhx99NEoKyvDBx98EPR9LVUq4RyOWOAFx6CiwLFiSEUmOEI1/QICAslqAaxWZReEryqR9/1QCqkoCQ5qbU4QRNIQKqRSUlICsznwJyE3N5dtyrx4UAsvUsrLy1X9jF6Cw+l0sk021DwQvQRHS0sL+zSqxuEAYhMcWia48oJDTZ6K0SEVq9WKzz//HPX19Zg4cWLQ97WEVMINcIsFieAoDhwrhlRkORyhEkb5nw8VTgHCtzdXCqko5YtQ4y+CSDNcLheuuuoqLFq0KKZJqmp56623MH/+fKxcuTLmc4VyOJQ+OQ8bNgwAUFdXB684VUslvOAYOnSoqp/RK2lUzSd13s2JRXCoLVeNpfmX1rbmIuLkX5fLhe7u7ojPNzqkAvgnEvPClkdLa/Nug/7btXYGhBkvOFQ5HOFCKmGm4IrwYSR5aazZEfyaRapSSTaHI8lSSggiNXjmmWfwr3/9CwBw3HHH4bLLLjP0eldffTXq6+uxZ88e/PDDDzGdi3c4xHCDIAgYPHhw0HOHDRuGH374AR6PBw0NDRgyZIjq69TV1bFjtYIjJyeH3U8sDgefMGq0w6FWcCQypAL4XQ7+HpSIh+AIR6T23wXc7bd1CQC0lbyqQZPDoZA0GgqtDodc4JhMJliyLPD2BIS/YkiFHA6CSC/eeustdvzTTz8Zeq3W1lbU19cDAKqrq2Mu4ZQ7HL/73e8wfPhw/PrXvw56bkVFBTvWGlaJJqRiNpuZyxGL4FCzccZbcOgVUonG4QDU5XGI17FYLJKfjReREisLAy8hWqOrLo4ILzjKigKCRklwWDItMGcEttGwDocoOMJ0/9Qyoh5Q0YcjySwFEhwEoRGn04mPPvqIfc03tzKCHTt2sOPu7u6o+ziI8IIjMzMTDzzwAHbu3InZs2cHPVcMqQBAbW2tpuvwDoeSexIKPQRHpJJYQCo4YgnfxDukEk0OBwCW2BoOUagNGDAgqO19POA/1StVqcRbcEQKqQCAles2qqZKJdocDkA6wA0ALBHKYqm1OUGkOKtXr5bEw40WHGIvCxF+M40GPqQib7wkhxcc0TocpaWlqjpjiujtcFBIxU8kh8Pn87Hcl0SEU4DIORwSwRFdw9aIiILDbpNeL2QrdW6Amz1ESMXjEeDpj4SoFhwq2ptHCqlY468Zw0KCgyA08s4770i+3rVrl+a+ClrQW3DwDkckIRBtSMXj8bD7VJu/ISIKDqfTKblXLWh1OGIRHGorSGIJqfCiRkuog39uJIejvb0dHo//43GyCg5HRiAvoSW6bvsREQVHYa70HvpcyqFMPo8j9Gj6wHFsDof6kIrdBs1t3Y2GBAdBaOTdd9+VfO31erFr1y7DricXHHv37o3pfNE6HFpCKvv27WMdHtXmb4joURqbjA5HLCEVsfFXQUGBpgmuWhwOfh3JKjhMJhOK+l9Go0Iq4nkLcqT3ENLh4EMqMYymB1TkcGSrdziSrUIFIMFBEJrYtWsXNm/eHPS4kWGVRDocZWVlbIPT4nBEU6EioofgEF8jk8kUMgTBXyfZQyqi4JBP942ElqRRfh1G9OBQg5qNudBAweH1Cujoj5YW5EjzSEIJjpz+abSWbAscQ5QFvGrBoTGkojhLpT+HI9kqVAASHAcdb775JsaMGYMHH3ww0beSkvDuxiGHHMKOjRQcfNIooJ/gsNlsIfshiJjNZuZQaHE4oqlQEdGjvTmf/BhqhovFYmGiQw/BkZGREbbsNFqHw+PxsKRWrYJDS9JoUjgcKkIPouDo6gXcntgqtuR09ASO5Q5HqKTRkb8diarfjcTUZybDmqv8u6amrTkAZNgCIRBVIRWFpFFRqJDDQSSc//u//8O2bdtw2223aW7kRABfffUVO/7lL3/Jjo0SHC0tLUGfTPUKqUQKp4iIYZX29nbVAkAs4wVicziiqR4RBEHS1jwceoyoFwXgwIEDw8bMo83h4N9/raWqWkIq1dXV7FhLVZGeRKpSAaSJnG06J47yg9uCQiohBEdGiR2jb65CyfGh3xu1DkemDmWxzOFIspJYgATHQcfOnTsB+DedWMd/H4zwn/JPP/10dmyU4JCHUwD9HA61lSN84igfKgkH/7x453A0NzfD7fb/tQ6VvyEiuimtra1R9TdpaWlhzkFVVfhR6NGGVPjBbUY6HB9//DE7PvbYYzVdRy/EjTnDHjrhkRcceieO8gJGHlLpi6GVelQ5HEohFRVJoy4uaTTZIMFxENHT0yP5o6OmLp+QIuYxFBcXY/DgwSw/IJUER7QOB6A+rBJNW3ORWAWHlm6ZosPh8XjQ09MT9rlK8O/P6NGjwz432pAK//9Uq+Cw2+1M6IRzOPbs2YNNmzYBAKZNmxZRqBmFmuZYRvbi4AWHvEollMOhhmiqVJRCOHzSqMlugtkeukqFQipEQpFP/CTBoQ2v18teQ3ETFjeZvXv3xtyQSwklwRFrSCUWh0Nt4ij/u6bVno9VcKhpay4Sa6UKLzQjCQ49HI5oun+KPxNOcLz99tvseMGCBZqvoRdqmmMVGSg4WiUhFZOqpFE1SB2O0GE3LWWxofJFKGmUSArkmwUJDm3s27eP5b3IBQegLA5ihU8YFSsHOjs7VQ3iCoUoOOLhcGht+gXELgL4cE4yCQ6bzcZeCy2CIxaHAwiEVZqbm0OGjVJJcBTmBjZsIx2OghzAZjVBbLgaKmlUDaqTRjXkcCgJDq9XgJiaRzkcREKRx99JcGiDF2xKgsOIsIooYkwmE44++mj2eCxhlVhCKmocjliafgHSTZWvnFALL/xGjRoV9rnxFBxAIKyiJaQSq8MhCg6Px6N43d7eXqxatQqAX6BNnjxZ8zX0Qs1E1XiFVMRBcaJA0M/hCP28iGWxXGtza06wokjmtuYACY6Ding6HE6nM+YhY8lGvAWHIAhs8xw2bBgqKyvZ96INq3i9XtZNUq3zwAuO6upqNDY2hnxvu7u7sXXr1qibfgHSXhbRCA4tIkAvwWGxWDB8+PCIzxcFR7ySRoHI3UY//vhj9Pb6Z73PmzcvYqm0kfQlXHAEfq8L+q8jhlV0czhi6TSaFRAZ1tzgvuWSSbHkcBCJJF4Ox8qVK1FUVITp06dr7qiYzChVXhgpOJqbm1klUVVVlSQXIlqHQ0uXURGHw8FEwJo1a1BWVoapU6cGtR1/+eWXUVxcjFNPPZU9Fo3DoZfgsNlskvwTJWIRHIIgsGuNGDFCVQdQMY8jESEVQDmPg598PH/+fM3n1wu3JxAOCFUSC8iqVDr1/VCj6HD0C4S4OBySkErw2qzZ4UMqkkmxlMNBJJJ4ORzPPPMMenp6sG7dOjzwwAOGXCMRKDkcI0eOZI+JJcd68fXXX7PjsWPHSvIRohUcWrqM8kycOFHy9ffff4+VK1eyr1taWnDNNdewclSRCRMmaL5Hh8OB7Gx/90atgsPn8zFXaOTIkSGbfonEIjj27t3LKlvUhFOAgMPR19fHnKZI6JU0CigLjm+++QaAv8nbSSedpPn8eqE2zyGeVSpAQPzEJDiimaWi4KiY+ZCKguDgHQ6qUiESSrwcDn5jfuihhzSPNU9WlARHZmYm+wQZa7mqHP6T5+zZsyWCI9qQinw0vVr++te/4pJLLsGMGTPYY+vWrWPHt99+O9sYR48ejdNPPx0333wzLr744qjuU3Q5tAqOuro65uKoEQGxCA6t+RtAdKWx/P9T3q1QS6ReHOLvdWlpadhOqUaj1gWIX5WK9F5iC6kE3IpYGn9ZIySN8jkcyehwJGGU5+Cjr68Pvb29kj9+eiMIQkIER19fH2688Ua88MILhlwrnojrslqtkv4OgwcPRktLC/bt2wdBEHSZ0CgIAhMcGRkZOOmkkySlpvEMqQDAuHHjsGzZMuzYsYMlYn733XcAgPXr1+Mf//gHACArKwvLli3DkUceGVMuwIABA1BTU4OWlhZ4vV5YLMHxaiW0ioB4Cw55aSzfxj0UopDLz8/XNLhNJFxIpbu7m51/yJAhms+tJ2oFR7wcjvxs6b30uRD1/2/1rc0Dx0qCI2tENkwWEwSvgJyxOQB8ku/zrgg5HEQQjY2NOOSQQ1BWVibp9Kc3ra2tQaWURggOn88XJGxefPFFrFmzRvdrxRtRcAwdOlSyAYrOg57dW3/44QfWHvzEE09EdnZ2QkMqIiNGjEBeXh6AgMPxm9/8hiWJ/uEPf9ClaZTocAiCoOn3NBbBofW9468VqRpGJBqHI9rBbSLhkkb5/6uJamcu0qcy7JCZEeiPobvg6AxcP8PuFxYS1yFKl0OvabGOIZmY9sIUjH/wEAxdGPx+SRyOJLQTSHAkmAsuuAA1NTVwuVxYsWKFYddRaklthOBobGxkcXx+U/7oo490v1Y86erqYp8O+aoNALqEOuTw4RSxL0Jubi6ysrJiuk60DoeI2WzGlClTAPh7bXz//fdMKI8cORK/+tWvorovOdEmjmoVHLzDEO+QiprE0VgGt4mEczh4NzLRgkPtpgwYNzFWdDgKuMiSHt1Go6lSCRXCGXBCCSouHQZLZoQqFXI4CJ7vvvsOn3zyCfta7ZyKaFDqn2CE4OCvM2nSJHasd35DvOHfG7ng0KN6RI5S5YDJZGLiJlEOBwAmOADgnnvuYccLFy6M+pxy9BAcY8aMifh8PQSHw+FQHY7Q2m2Un/ESTcKo/OfkgoPPr0qmkEq4KhUgIDhaDBIcfNiGD4FEmziq1r2JlDQaCScljRJKCIKAJUuWSB4zsv5dScz09fVFNT8iHLzgOOKII9ixkYLD5XLhq6++wueff45vv/3WkCm4SgmjInqEOngaGxtZCGrixImS0k5R3LS1tbHeCVqI1eEAgKlTp7LjV199lR3PmzcvqvMpEavgyM3NZZ1Zw2Gz2ZgI0CI43G43q0oaNWqU6v+7WkMqsZbEAuGTRpPK4VC5KQMBQdDTB7gUykejweMR0NX/X4p3ODJVjKiPhF4hlUiQw0Eo8vbbb2P16tWSx4xsxMULDjEGb8Q1+T9ghx9+ODs2SnB4vV4cddRRmDFjBo477jgcfvjhOPvss3W/jlrBoUdI5d1332WfauVtpmMVN9FWqfDwgoP/9D1t2rSozqdENILD6XSipqYGgD/EoTa5L5oR9TU1NaysVW04BdAeUom1JBaQ5qmEC6kkk8PhsId/74yoVFHqweG/l8BxtA4HL6bCuTeRkkYjQTkchCKff/550GNGCg7+D8thhx1m2DX564waNYr9kdQrt0HOli1bWLWEyJtvvgmXK4YaNgXCCQ69Qyo//PADO5b3RdBTcEQb/hg1alRQ+eQpp5yiupJEDdEIjp07d7LkVS0iQPwdPXDgAPv5SPAzbtQmjALSzV/NumLtMgr4XRzxQ4b8/zsfUknUhFiRaHI4AOMFh2REfZR/VuIWUuGrVCKItkRAgiNBKG3A8XI44iU4hg0bJsk5MKLVOd8LQkQQhKDJuLESz5AK/17xjcWA2MWNHiEVs9kcNG9Dz3AKEJ3g0DIqnkd8/zweT9iJqjzV1dXsWP4ehYNv9a4mZ0uPkAoQCKuEcjgKCgok7ksiULspA/EVHJKk0WgdDpX5KRZLYFgcORwHCT6fD2vWrEFHR4dh1+A3i0MOOQSAfxS32u6DWhH/sBQXF0v+6BklOCwWCwYNGmRIySgPLzj4JFW1Y9TVwp9PPh9Eb8EhXstkMgXZ3LGGb/RwOABpWMVkMmH27NlRn0uJaARHNFUjACQ9VRoaGlT9DN9VdsSIEaqvpVVw6BFSAaSCQxT+Xq+XCfNILeDjgTaHI/DpXa/EUYng4JNGdRQcGXZEDPWJgiQawUFJoynIgw8+iCOOOAJTp041TACIG5PD4ZD8wVL7CUsLXq+X9XQoLy+POMwpFsTNcsiQIbBarYaUjPLw4ZTTTz+dHevd3VTcHPLz8yU5MID/PRSrHfRYo3itQYMGBTV6ilXc6OFwANJKlSOOOCKmT99KRCM4tm7dyo6jcTgA9a8pLzi0OBxDhgxhG44aUayXwyH+n/d6vWhqagLgF1diCXs0Q/b0JplCKoU5AVGQyYUmYg2phGv6JSIKhWhCKpQ0moK89NJLAPxxWr0/KYuIf9gGDRpkqAAA/P0SROFUUVFh2PV6e3vZJzIx7GBEyaiI1+vF999/DwCorKyU2Px6vm98MzN5OEVEr9CR0+lkn7KVNoFYX089kkYB4IQTToDd7v/recEFF0R9nlDk5OQwB0at4Pjiiy8A+N01NSWxItE4HGJIxWazaUq2tNvt7HpaHY5YBMfYsWPZ8X333QdA+n8kGRyOTq5gLtLGXMRpfkNCKgY5HJGEFMAJjlhDKiQ4kh+Xy4UNGzawr8VPA3rS29vLMuIHDx5suOCQW81GXU9pmqre4Qaebdu2se6pU6dOlfzR1FNwNDU1sSTUSIKjp6dH0yRQOaITBSgLjmQJqQwdOhSrV6/G888/j1/84hdRnycUJpNJ0zyVPXv2YPPmzQD81VFa8hG0/o4KgsAcjsrKSs3JsuL72tDQEDG5Wa+Qyu9+9zs4HA4AwGOPPYaNGzeGzUtKBFvrAkK9MkL+qhEORzsnOPKyAsd6JI2KgiNSfxEgIDiiuRa1Nk8xNm3aJPkjYITg4D9FxcPh4JPp+MoRva+n9AfMyJAKH06ZMmWK5I+mniEVNX+YeechlnWGazAG+JP7RKGQyJAKAEyfPh0XXHCBrtUpPKLgUFM9wneyPfnkkzVdR6vD0dTUxPrXaAmniIjvqyAIEoGpBP//MxbBUV5ejt///vcA/M7gkiVLJP9HkiGksnGX/1+zGRgT4XYKuaTOVp1G1HcHtDhyHIFjPRwOFlJR43DEkMNBSaMphrzEUuu0SjXwG0U8BId85kM8BYeRIRU+YXTq1KkoKipiY831dDjUCA69nJxwyamA/5O/+Jomqiw2XoiCw+v1Rkw4jkVwaH3v+AoVLQmjIloSR8UPPHl5eVENbuO54YYbUFlZCQBYuXIlnnzySfa9RIdUvF4Bm/v1z6ih/nkp4TDC4ejpCwiXLE6L6xJS0SI4YgmpUA5HaiHmBIgY4XDwn4DjEVKRlwvyvQDi6XAYKTimTJkCk8nErrt7927dynDjKTgiORz8tZqbmyWOhRr0dDiMRm3iqCAITHDk5ORIOtyqQavDEW2Figj/voYTxg0NDazfR1VVlebryHE4HHj44YfZ1/zfhUSHVHbuDbgA4ysjP58XHC06FRPyDkc2918j1pCK2yNAbH6sJaTidEHz3zAn13WVQiopgNzhMEJwxNvhEP+wiBNHrVYrq6pI1ZCKz+dj4rC8vJxtTuJ1+QTWWNEqOPQKqYSyuflrqU1yFNEraTQeqBUcGzZsQGNjIwDg+OOP1+wEZGdns5wPNWJx165d7DiakIpah+Odd95hx3r1OTnzzDODmsnZbDaJ6EoEG2sCx+OHR36+xOFQN3Q3IhLBESqk4tL+IUZLfxFA2t7crbFIkhyOFMLr9WL9+vWSx1JdcMhnPogleeI1jRYcfMmong7Hjh07WHIm3xPCiMRRrTkcRoZUYr1WKoZUgPCCgw+nzJo1K6priRtuPBwO/n0N9zuqNDE4VkwmE5YuXSrJuykvLzd0jpMafgq8pJgwPHKHzAx7YER9u06Co4cTHFncfw2JwxFFSEXasj3y8yO1N/9orYDH3xAU74VyOFKIbdu2BQ0zMyKHI54hlZqaGjbMjG/BLF6zra1Nt2Fn4h/P3Nxc5Ofns8eN6DbKO1G84DAicVRcl9lsDjngSu+Qit1ux8CBAyNeS6ubwrs+6SA4BEHAyy+/zL7Wmr8hIr6m7e3tEYfi8YJj+HAVH8dl8L+joRwOp9OJDz74AID/dZg+fbrm64Ri/PjxuO666xTvJ1FsrAn8XVDjcAABF6InyrwKOSEdDn6WShQhFUmXUQ05HEBwL476/QJO+Z2Aqx8W8O93EIS0tbm2+4wHJDg45OEUIPUdDnmFioh4TUEQNI/lVkIQBLYxDxs2TNJNjy8Z1at7K5+4N27cOHZspMMhNjNTQq+QinitoUOHhvzUGa242bJlC958800A/iTEZNhowqFGcLzwwgv46quvAPjH0Ytde7XChxQivaai4Bg4cGBU7cAHDhzIwj6hfkc//fRTVvI9f/583R2IO++8kzkt0Yo0PREdDqvFnzSqBtGF4J2JWOjmdCbvcMSaNCoJqagQAbyjInc4ttWB5YNsUvg8RQ5HCiFPGAWMFRwZGRkoLCyEw+FgNfJ6C45Q7Z71Fjn79+9nCYlGDzcDQidXqk3IU0tvby/b7MJt0Dk5OZryAJTo6OhAe3t7xGtF+3reeuutrLz0pptuSvkcjq6uLvz2t79lXz/yyCOqJ8TKUZsX09fXxwRlNOEUwO+UiZt9KIfDiHAKT2FhIdatW4cvv/ySlcsmCrdHwNb+l2HMMMBuU/ceipUksTgcfc6As8I7HI5QIZUYHQ6tORxywcHfo5L4kbQ2T0KHIwk1UOLgHY7Bgwdj79692L9/PwRBiPoPmRLiH6xBgwZJcir27NkTd4cD0EdwaBluxnc91ON6fExc75CKmqoRkUGDBqGzszNqwaEmYVS8jkgkN+X777/HK6+8gs7OTrz66qsA/J/mlyxZEtU9xpNIguP//u//2PoXLFgQU2JlJIejo6MDTz31FNauXcsei1ZwAP73d+fOnWhra0NnZ6fEKREEgQkOm81mmAMxYMAA9hobMVhRLTv2BJIj1VSoiMTqcFx6vw8rPgQevwG4ZK6JnScrEzCbA3/vY3U41A5uEwkXUuFdGKV7kSSNJuHunoS3lBgEQWCCY/DgwZgwYQL27t0Lj8eDtrY2SSlpLLhcLrbB8xsHLzj0FDjyHhz89UT0qOYIt1kaUakiXi8jI0OyMQ0ZMgRmsxk+n08Xh0OtCAD869y2bRs6OzvR1dUVNMJdy7XCiRu1IZW+vj7Mnj076P294447WL+SZIZ/Xzdt2iT5f7F161Y88sgjAPz5Lo8++mhM14rkcPzxj39k1xOJpkJFRJ7HwYcFN2/ezCphZs6cGTS7J934KVD0g/EqEkZFRIfD7fG7JDar+p/tcwpY1p8D8e+3BVwy18TcgyyZCxGrw6G5SoUTHPLr8W6OkrNDVSopwq5du5idPWXKFEnCnp5hFXmXURFRALjdbnR16ZR2jYDDUVBQIJnFoLcICOdwGBFSCZXrYLPZ2PX0cDi0tH8eOjQQfBancGpBrbgpLi5mOQDhXs/6+vogsTF58mRcdtllmu8tERQUFLCNeP369XjuuecA+D8cXH/99Wzw2I033hjT5g9Edjj4ShjAHxbhhwVqJVxprNHhlGRj4y4uYbRS/c9lxeA8dHC1Ae3+VBkmOPiEUUBfh8MRoaEZEL5KJZLDQdNiUwR5m2xecOhZqcJv7kqCA9Avj6Ovr49tmHxJLKB/6EFLSCVW2tvbWfKpkggQE0f3798fseIgEloEh9pyRzXXCic4TCYTe03DiUX+9+iss87C6tWr8cUXX8TcsTJemEwmSaOqG264AR0dHXjjjTdYBcewYcN0yUEI53D09vZi48aNAPxVKatWrcKePXsk1VFaCfe7ctAJjprA8QQNUSq+G6jWsEpXb/CxeI5sWWoT73DEJaSiNodDYc00vC1FkAsO3s7V0+HgN1z+k78RgqO6uprFZvlwCqB/NYdawaHn+HZAeWNWU3aoFi2CI9bras0XAfyiSvykL4f/PZowYQKOPvpolpycKpxyyik444wzAPiFwKmnniop6XzkkUeQlZUV4qfVE87h2LBhAysdnzJlCmbOnCn5nY6GUL8rLS0tbOrt2LFjY3ZuUgGxQsVuA0YqV50rIhEcGoUALzg6e/yuWTeXw8FjsZhg608+iHdIRZ7D0cMluSqV6CZ7DgcJjn74CpXJkycbFlKRl8SKGCE45C3NeWL9NC5HPIfJZAoa163XYDMRLYIjVvcmWocjGsHBl/qqyRcREbtsytFr8FeieeSRR1hFzWeffcZ+h2bNmoWzzjpLl2uUlJSwZlhyh4NvoT9hwgRdrse/v2L7cgB47733WCXRweBuCIKA6YcAE0cAk0YCVg15GHxIRavD0dkjPe5zAWLerNzhAAJCISqHQ7PgCLwG4UIqSmt2Ug5H8iMIAvujUlxcjPLy8rQIqYSqUAGArKwsltOhZ0hl0KBBsNulvmF2dja7Fr+pRkskJ0BP90b8+ZycHNYxNRSxiLj29nZ8/fXXAPzVD3zjNCXUiLh0ERzDhw/HvffeK3msoKAAf/3rX3VLrjabzSgtLQUQ7HAYIThGjx7NEnc/+ugjJjIOtnCKyWTCM7ea8eNyM755XNt7qVdIpc8FdHQHvg4nOGIti9U1pBImadRmha6VlXpBggNg5a9AYAhYOoRUQvXgEBE35vr6eng8Gpv2czidTvapMJQLIAqe+vp61tAoWiLlOujlcIRrZqZELCGVDz74gL0HajYaNXkx6SI4AODXv/416urqsGHDBmzYsAF1dXW6lFfziK9pU1OTpPuuGG41m81RNxaTk5GRgdmzZwPwf6D59ttv4fF48O677wLwC6oZM2bocq1UQesGKXE4tIZUpA2l0cT1PpSHVICAUIim02hMjb/kIRVOcChWqfT/GU/GhFGABAeA4PwNAGkXUpE7HEBgg/T5fKivr4/6OnxFRiTBAUgt5GiIl8Nx4MABNntETVfOgoIC9qlV63W1frI92AQH4K8CmjBhAiZMmKC55FgNYh6H1+tl1T1OpxM//fQTAH9OhR75IiL8+/zWW2/hyy+/RFtbGwB/7kqqJPcmiqzMgEDRHFKR5ZI3tASOFR0OUXDEXKUS+flhq1QiNf7qFyjJGE4BSHAAiCw49AypiJuz1WqVbAJGOhwDBgxQtOj12pjV5DnwDgsvhGK9XiSHI5Z18e5IpJwKwP8JTbx2XV2d6mZKXq+XTQbNycnBcccdF/FneHcslFhMN8FhNPxrKiZubtiwgSXlin8b9IJvVPbWW2/hX//6F/t6/vz5ul4rHdHT4WjkHA4lwZEZU0gl8HdAc+OvMIKjxxnIOxERHY5kTBgFSHAAkCaMin9UsrOz2acZvRwOp9OJLVu2APDPfeD7R+gtOLq6utgnX6VwCqDfxqxGcPAOR6yCQ3Q48vLyFJsi5eXlMYEVS0iFd2KqqqpU/YwoTHp7e1W/j2vWrGGfqGfPnq1qqFplZSU7DpUXQ4JDG7wAuOGGG9DX1xdySKAelJWVYdq0aQD8fUZWrFgBACgqKsKpp56q67XSET70odV5kDscjZzDoRRSER0OlxvwerV1ZY1lPH24kIrPB7hlczdFgZKMbc0BEhwAAg5HXl6epF2xmMehl+DYuHEj+7Q0efJkyff0Fhz8ZqkUTgH0y3VQ0z+Cvwc+t0QrPp+PCY5wYQ7Rvamrq2MJeVqJlAOjRDR5HNEkCo4YMYIJ1lACTvw9slqtad+tUg9OP/10nHDCCQD8jQAffPBBScKo/P+sHii933fddVfEpGEi1qRRqWhobA18HS5pFFAeGR+OWEIqckelW7bOPqc070VMGiWHI0nZv38/2xgmT54scR3EsEpzc7MuI9zlpbc8BQUF7Np6CI5QLc154hlS0cvh2L9/P1wu///CcGEO8T7cbnfYYVzhiEZwRFMaywsOtfNAMjIy2Bq3b9+uGL4Rf4+KioqSMmM92TCZTPjrX//KymPvuecevPjii+x78RAc48ePx1VXXaX7ddIRPUMqkhwOR/D/lViaf8U0SyVMWSwA9LmlWzgLqVAOR3LyzTffsGP5HxRRcPh8PrS0tCBWwgkOs9mMoqIiAPrkjITrwSGiV0hFTcOqnJwclugYi+BQ241TDzElCg6TyaR6UJfW0tg9e/bgxx9/BAAcfvjhrDRTDWKYp729XXEejig4KJyinvHjx+P6668H4A+BikmcY8aMMSRRdfLkyZIE4D//+c+wWpP042mSEYvDETakouBCxNLePKaQilv6QUIurHplDocYgqEqlSTlvffeY8czZ86UfE/v0lhecBx22GFB3xcbZu3duzdmRyVShQrgX5/YUEmPkIrD4Qi7uYnCp6mpic2t0YrabpyxiilBEJjgqKioUD3KXWtIhZ/Rccopp2i6x3CVP06nk5Ufk+DQxp133okjjjiCfZ2Tk2PYCHez2YwHHngAJSUluPHGGw2bDJuOSB0ObXkVXXLBwSeNKjTjjWWAW0xVKpFCKu6A4PD5BHj6t420cDhcLhf++Mc/Yv78+Zg5cyYuueQS9ukMAJYvX45Zs2bhxBNPxNKlSxM68lgNgiCw6gCbzYaTTjpJ8n09S2O9Xi9++OEHAP6EP6Xps3wYIFT3SLXw4YBQCY98VcXu3bujer+09KrQI6yidsBZrPkp+/fvZ6JIbThFfk9qhM6HH37IjrVuNuFeT0oYjZ68vDx8/fXXcDqdzOW4+OKLDbvez3/+c+zfvx8PPPCAYddIR2LK4ZBXqUQqi43B4dDeaZS7r1bg8TcE7G70/20OCqlwDoebn6OSpCaZJsHh9XoxePBgPPXUU/j444/xs5/9DL/+9a/R09OD1atX46WXXsLy5cvx4osv4ssvv8Qbb7xh1H3rwrZt29gY6GOPPRa5ubmS7+tZGrtjxw72iTNULFjPltziBjRkyJCwo8jFa3Z1daG1tTXk80LR2trK1hWpV4UeiaNqW43HGlKJJn8DCM7h+NOf/oT58+crVpL4fD7mcOTm5ko+VauBBIex2O122O12ltNBJBd6hlT2c4arYpUKJxS0Ohz889XkcPDPWfEhcPXDAub+ToDPJwRdm3c4kn1SLABo0kEOhwNXXHEF+3rOnDn485//jNraWrzzzjs488wz2Yjuiy66CP/73//Y4CUel8vFEv/YjVitQS2x9UCsUFCqVBDdDQCYO3du0HP4kMr27dujrnYApO2RDzvsMMVz8RtoTU1NxA0o1NpaW1tZTH/UqFFh71t+zUjtu+WITZEA/2Yb7lq807Jt27awzw21Nn7jHjJkSMhz8KPia2pqNL93mzdvZseRXkOejIwMlJSU4MCBA/jmm29YPweLxYLXX38dQGBNP/74I3POZs6cCYvFouk++cFe8teTF8hFRUUx/e5qIdz/t1SH1pZcZHKbandf6HtXWpvc4eAj2A67f3Pn4TfwjTUCpo5R7waLU10tZsBiDj63HJvCrrypBmjvCv65XqeJrauPc15s1vi+l3yxRThiMl52796Njo4OlJeXY9euXZgzZw77XlVVVcj+AMuWLcMTTzwheezcc8/FeeedF8vthEUpnv7qq6+y40MPPTTIVeA34zfeeAMXXnhh1Nf/9NNP2fGQIUMUHQx+kuePP/6II488UtW55WsTQzeAvyNlOLeEL5dct26dYqgnHP/973/Z8dixY8Nei3da1q9fr8rFka9t7dq1APyzYEwmU8hz+Hw+2Gw2uN1uVFdXa3aMxOsAQH5+vqafLysrw4EDByTt4j/66CNs27ZN0mPjlVdeYcdTpkzRfI9msxlmsxk+nw+bNm2S/LzY70V8nh7zcrQQ65TeZIbWlhy0tFoA+D9YHGjtRm1tcOI0D7+21o7BAJRtgI62faitlX4gLi/MAuD/AHrNwz7kWJowdZS62EpH1yAAdmTafaitjfz6HjgQWBfP2g31AKSDMfvcJrauRu718Hoivx56Mnz4cFXPi1pw9PX14bbbbsMll1yCnJwc9PT0SDaU7Oxs9Pb2Kv7s4sWLsXDhQumNGOhw1NXVoby8XKLCuru7sWbNGgB++/2kk04Kyj+oqKjA2LFjsWXLFqxbtw45OTlR29M7d+5kx3PmzJF0NRThGwt1dHRIwgJa1rZ69Wp2PHny5LDnmThxIjvu7e2NeE054rVMJhMuuuiisFUWpaWlMJlMEAQBe/fuDXstpbU1NzezrpqTJ0+OWDkydOhQ7Nq1Cw0NDZrXxefQHHPMMZp+fsSIERLnB/C/trW1tTj55JPZ2nhRc95552m+R8D/O7pr1y7s3r1bkkPD/z6MHDkyqnNHQ6jfyXSA1pZc5BdxX5izUVGhHDpWWltfmF4aIysHQf7f5ZqhwOotwEsfA30uM654tAzrngAqBymfg0c0TxwZZlX/DzNCFEP5bEOCHutzmtm6fNz2WZAX+vVIJFEJDo/Hg5tvvhnl5eUsxJKVlSUZytXd3S35xM4jxkbjifhpUOSTTz5hYZ25c+eGjNMuWLAAW7Zsgc/nwwcffBAklNQgCAKrUBk4cCCGDBmimFzJd4+sq6tT/R9fvja+YmH06NFhz8MrUy3XBPx5Jhs2bADgL+nky/uUyMrKwrBhw1BbW4vt27fDZDJF7A/Br239+vXs8alTp0a812HDhmHXrl0sz0SeoxMOMSfCbrejsrJS0+sSKrfkvffeYy6g0+nE559/DsDveI0bNy6qXhmjRo3Crl270NHRgebmZpZ3xOfjDBgwIO6biPx3Mp2gtSUHOVkCAH+YodcZ2dbn19bVFzrckJtlgtks/b9oNgP/uVVAa6eAj9YCbV3Asx+acPslkf/P9jr913JkqAs9FOYJyLALQRUq+5pNENcr0uc2sXV5vIHXI8OmPswRTzTfkc/nw2233QaTyYQ777yT/ZEcPny4ZKOrrq6WxJiTjY8//pgdz507N+Tz5AOWomH79u0siW/y5MkhN5aysjI2sCkWC5zPP4iU8Mgrbq1D1d5++212rLZDppjo2NbWprnBGZ8Ho6bNdLSJo16vl70WVVVVmpMG+d/7yy67jP08nzP03HPPMQdw1qxZUTfmCpU4SkmjRLpjswLif00tjb98PiGo2oNHqSwWADLsJtx3ZeD/6b5mdXkcYlWLmgoV//NMWHazCYtOAS7gCif3KBRK9rpCJI2mS2vze++9F83Nzbj//vslDWrmzZuHV199FXv27EFzczNWrFihumtiIuDt7HBjoGfMmMESKd977z3WmlwL/MYsL73lMZvNrMohlkZcYqt2h8MRsgeHSGVlJQuF8Q6CGmIRHID2ShWtgiPaqp/du3cz90tLhYrIwoULMWfOHJx77rn485//zH6/tm3bhurqajQ2NuLPf/4ze/61116r+RoiJDiIgxWTycR6cWipUpH3spCj1PhLpJAzSVs71V1PrCxRU6Ei8rNZJiy/xYxpYwKCYs/+YIHTxwkOF7c1pUVZ7L59+/D6669j48aNmDVrFo499lgce+yx+P7773HMMcfgnHPOwaJFi3DOOefgyCOPxOmnn27UfceEz+djIY5hw4ahpKQk5HNtNhtryNTW1oYvv/xS8/W0zMoQN8m2tjZ0dHRovlZbWxtL1p00aVLEroVmsxmTJk0C4K/mUFsa293djZUrVwLwhwTEc0QilqmxvJAaM2ZMxOdH63BEWxIrMmDAALz33nt48cUXkZubKxHe7777Lm655RZ0dXUBAC699FIcfvjhmq8hwlf+kOAgDjbEElYtgqOzJ/z3wzkRvOBo64p8LUEIlLKqdThCXa9eIQc0pOBIh7LYQYMGSZwBOYsXL8bixYtjvimj2b59O/uDr+aT8oIFC1g1xltvvRXUkTQc7e3t+OyzzwD4kwnHjh0b9vnyTXLChAmqrwVIu5mqnW45efJkJqTWr1/PBliFY+XKlXA6/V7hggULVIcEom3+1drayhJvDzvsMFXtn6PtNhqr4JAzb9481qlyyZIlrFwtPz8f9913X0znVuNwiC3zCSLdYA6HhpCKvMuo5HyZCMrf4Mnn8jDlDocgCPjvSr+Dcuk8/3m0tjWXwwuOPQqtoPpChVSSVHAkX1ZJHNBqzZ9yyilsQ+VzP9TwwQcfsPJINRtzrC25+bVNmTJF1c/wjch4wRIKQRDwyCOPsK/VhlMA6QauJaQSzZhwPglXrbgRBAHPPfcc+/qQQw5Rd4NhmDhxIguV8bXxd911l6S5XDQMHz6cldq+++672LdvH4CA4MjNzY17gjZBxItoHA55Dw7J+SKIAovFhPz+KhK54FizGbjwLgFX/EnAm/72O5qbfsmRCA7FHI7AFu7iO42S4Ege+M1LzaZcXFzMNsoNGzZoyuPQOnpcT8GhxeEQUSM43nvvPdZXZNSoUZL+K5GorKxkSZRaHA6t7xngT94UK6XUrAvw91v5+uuvAQDjxo3T3P1TCZPJhCeeeALHHHMMJk+ejMmTJ+OSSy7B1VdfHfO5bTYbLr30UgD+brE333wzABrcRhwc8A6H2tEM8i6jPKESRnkKQwiOzVya2PfbA9UzIo4YBccBhfFToXM4knM69EEpOGLZlF0ul6QKJBxer5dVJuTk5OC4446L+DN8SCWaShVxY87IyMC4ceNU/cz48eNZiCLSxuzz+dimBvhHeIuVNWqw2WysFDfUWHU5fM8UQP17ZrFYcOihhwLwV01FGhjn8Xhwyy23sK/vvfde3dpaz5kzB59//jm+++47rF27FrfffrtuU0Hvvvtu1rDtmWeewZdffsmmG5PgINIZ0eHw+aQbbjjChVSU5qjIEUVAa5dU5PCVL+JsFq2D20JdKxQUUklyfD4f25SHDh2q2tLW6gIAwJo1a1iL8dmzZ0u6TIYiFoejo6ODhSkmTZqkWghkZGRg/PjxAPwdKsVyzebmZkybNg1Dhw5lFSwrVqxgA/umTZuGc845R9M9AoGwSnd3NwsBKOHxeDBnzhzk5OTg5ZdfBgBkZmaqFlKA9H3jO7Aq8eyzzzIxOWPGDJx22mmqr5NIiouLcffdd7OvL7/8cha6IcFBpDPSibHqfiZsSEWD4PB6peKFr34Rp8/qGVJRItWSRg86wbFz505W/aH2kzIQneB477332PH8+fNV/UwsA9z4+1IbdhAR1+f1elkzrz/84Q9Yt24d6uvrcemll6K9vV3ibtx///1RN6sSCRdW+fLLLyXj2wF/gzEtzoCW923ZsmXsONq1JYqrrrqKdY3lHTgSHEQ6E80At7AhFQ2CA5CGVbr7Am6HXg5HjiPQa0QJvg8HX31DDkeSEE1SJRCd4OBHj8+ePVvVzzgcDjY0TqvDEU1ipYh8fT/++CP+9a9/SR475phjsHfvXgB+ARWup0g41PbiWLVqFTuePn06zjnnHPzlL3/RdC2171tLSwsbtDZ69Ggce+yxmq6TaKxWK5YuXRr0OFWoEOlMNIKDdzgKZG3EYxEcPQoOR6yCw2QysZwRJXiH47MfAoJnXKX2a8WDg1pwaNmUS0pK2ATS9evXR5zE197ezvIOxo4dK5leGgnR5aivr5cMAItEtGsDpBvzF198gV//+tdBaxTng9jtdsXNTS3henH09vZCEAQIgsAEh81mw8qVK/HSSy+p7vchMnHiRJaHEU5wvP/++/D2j4zUUnWTTJxwwglBIS5yOIh0JpqQCu9wlMn0uJaQCiB3OALHeoVU5NeTIwoOn0/Ah/0dK3KzgCPHR3ctoznoBMe3337LjqPdlDs6OrBr166wz/3kk0/YBnbyySdruo6YOOrz+diwskgIgsB6pNjtdpaToRZ+I//Pf/7DNvvhw4cHTfG98cYbY2pbHyqkcvvttyMrKwuXXXYZNm7ciD179gAAjj/+eE1zUHgyMzNZaeumTZtY7xA5WquJkpWHHnoImZmBv5okOIh0JjqHI+AEDJL991DncARcBYng6JUed/cKMocjuhBteMHh38J/2AHsb/M/duIUwEZVKonnu+++wyeffAIAKC8vR1lZmaaf1xJW4cMpWgUHv5mHa7Qmv97WrVsB+O9Ta++FvLw8HHbYYUGPP/TQQ3j00UfZhj9s2DDWxCpaysvLWQKtGFL5/vvv8X//938A/LkUS5YsYc+PVQCI75vH4wma4io+/u677wLwvw7HHHNMTNdLJBUVFbjpppvY13wnUoJIN6JKGuWEQXSCI3AcyuEA/C5HrCEV+fVExKRQMYfjg8DnaMyenpxiAziIBIcgCPjtb3/LvuaP1aJFcIjJjhaLRVNnUkA6b0XcCMMhL1X91a9+pel6IsuXL8ell16K8847D+eddx7+/ve/48wzz8SgQYPw/vvv47rrrsMHH3yArKysqM4vYrFYmKiqrq6G1+vFkiVLJCVmojAE1CfchiLS+/bVV1+xlu6nnHKKpjLfZOTWW2/F3Xffjfvuu0917hBBpCJZmYHNNZqk0cGyqRaqQipcTkVYwdFiXEilJN//r5MJjsDfztnTo7tOPEjSES/689Zbb7FNbOTIkbjmmms0n4PfuFavXo3vv/8ew4cPZ8PdfD4ftm3bhvr6euY2HHnkkcjLy9N0nZkzZ8LhcKC3txfvvPMOBEEIWzHxwgsvsI108uTJQSEQtUyaNAlPPfWU4veOOuooHHXUUVGdV4lRo0axEMfdd9/NRrXLOeSQQ2KeOsy/b6tWrcK0adMwevRoJpzSJZwiYrPZ8Ic//CHRt0EQhiNxOKJIGh1ULB35rt3hEAD4/zbLJ9A2tgK9MbY2l18PAMxmf7Lr3gN+h6O7F1jtLyzE8EHAyCHRXSceHBQOhyAIkjDAAw88EFW752HDhrEGS5999hmmTJmCiooKbNq0CQBwxRVX4JBDDsGsWbPYz2gNpwD+vIMTTzwRgH9gXrj+EU6nE7fffjv7+oEHHoDZnPxvK584+sc//pEdK00hjhU+VPT8889j8uTJGDNmDJqamuDz+fD6668D8GeEz507N+brEQQRHyQ5HFGEVORJo9mOyOGIkFUqsus3tsTeaVR+PcAvisR197lM+GR9oAfH7OlI6nL+5N+ZdGDjxo2sN8GRRx6Js846K6rzmEymoMFmHR0dWLJkCVatWoV///vfQT9z6qmnRnUtfuMLF1Z56qmn2FCzk046KSqBkwiOPPLIoMdOPvlk3Hjjjbj++uvZY2effXbM1yooKJC4HACwZ88e3HbbbVi2bBnLIznmmGPCTg4mCCK5iLUPR1CVigoXQiI4uImxSg6HLiGVHKmAyM4M3KdPMGFloDgRJ09LXrEBHCQhFb6fw8UXXxyTAvzLX/6CQw45BC0tLXjzzTdRX1+Pjz76SJIbcNZZZ6G0tBQnnHCC5gZcIrzgeOeddxQTNXfv3o2//e1vAPx5EfxAtWTnjDPOwN///nfWZKygoAC/+c1vYDKZ8MADD6CyshJWq1WXWSYA8Nxzz+Gpp55Cd3c3nn32WXR2duKJJ57Aiy++yJ7DOy0EQSQ/sXQazcoEG8QmomqWisqk0YYWgeVaAPqFVLIypefawrVrmpTkOeIHneCINQFxyJAhrJrixBNPxLnnngsgMCzr8MMPx0svvRRzWEMcZb9lyxZ89dVXePvtt4Naoy9dupS1Ib/uuuvY3JBUwGw2h8yjsVqtuO6666KaJROKsWPH4sEHHwTgHyB30003QRAEtLW1AQDOP//8IPeKIIjkJhaHI8fh71khOZ8KUVCgIWmUzwnRS3DwDgcAVHOdE+RVN8lG2guOhoYGNvvj0EMPlbQOj5Wzzz4bxx13HD777DMA/pDLY489plsOxdy5c7FlyxZ4vd6wyYwDBgygT+caWLJkCf71r3+huroaAJCVlYWHHnoowXdFEIRWHJKk0UACZzjEHI7cLL/o4FHjcFgsJuRlC+jojlwWy1fB6FWlInc4ahr8/+bnqMtBSSRpn8MhTmsF9K9AMJlMWLp0KUtyvPLKKzF9un41SaJ7Eon777+fVcoQkcnIyJCEn+644w5NnWAJgkgOYunDoeRwqKlSAbiJsf2Cw+sV4HRJnxOUNKqnw8Hdp8ffXxKDk9zdAA4Ch8NIwQH4KyA+/fRTbN68GRdffLGu5z7qqKPwv//9D998843i9wVBQGlpKRYtWqTrdQ8GTjvtNLz77rtobW3FBRdckOjbIQgiCrSGVNyegDDIcQQLDDUhFcAvAmobgLb+EfVydwMwrvGXPKQiIu8pkoykteBwOp2s42dJSQkOP/xwQ64zY8YMzJgxw5BzL1iwIKRQ8vl8qK2tTeoyqGTmlFNOSfQtEAQRA1odDr4kNjcLMJtNyHYIrMJETUgFCDT/cnv8QkdJ7HT2SEMuRoVURFJBcKR1SOWzzz5DV5e/bmnu3LlsiBdBEASR+mh1OPimX2L+Ri4nMtSGVOSJo0oOBwDUNgaOo3U4crOkI+r9IZXgD5mpEFJJa8HBd5CMtTqFIAiCSC60Nv7ie3CIgoNPHFXT2hwILo0NJTjEhE4gesFhMpkkAifboXyuIQOS3+lO65DKnXfeiSOOOAIvvvgizZQgCIJIM/junVodDjFhlE8c1Zo0CvgFhy3ETsrfU0YMI5oKc4Dmdv9xVkaIHI4UcDjSWnAUFhbiggsuwFFHHYX8/PzIP0AQBEGkDFarCXabAJc7eofjkArg++3A0AHqXQj/iHr/DJbWTmnux4CCwKh4kQy7P18kWniBk+0wpWwOR1oLDoIgCCK9ycrwzxLpVZM0KnE4/ALgwWtNGFcJzD1CvSgI1d4cAEYMDhYc0c5RUbpeVoZy6CcVBEda53AQBEEQ6Y24+aoJqSg5HINLTLj1YhOmjFHvQITL4Rg5OPj50VaoKF0v26EsYORzYZIREhwEQRBEyiLmM6gJqYh5EABQlBv6eZGQj6jnrz1CQXCUxBjRlwiOzGCHoyQfyLAnf9IoCQ6CIAgiZdHicLR0Cuy4OAYREORwcM7J8EHSjX9AAfDIL2ITA/KQijyHIxXCKQAJDoIgCCKFEQVHrxPw+YSwzzXG4ZCGVIrzgKtO8yeKXjoP2PwfE06eHpvgmNYf7jGZ/BNh5VUqqSI4KGmUIAiCSFn4zXfjLmDiyNDPbe4IHOvqcPQFhE62A/jnDWb87dcCLBZ9whxnHAu8fo8JJQVA1VATNtdIhVUqlMQC5HAQBEEQKcyw0sDxcb8U8PF3oV0O3uEozov+mkGdRrmQitjLQy+xAfirZ04/1oSjJ/rPKc/hSBWHgwQHQRAEkbLcd6UJk0f5j9u6gDk3CKiuVxYdLf2zTSwW/zj3aLFZTazKpUUWUlE7jyUW5DkcqdBlFCDBQRAEQaQwZcUmfPZXE+Ye4f/a7QFe+0z5uaLDUZiDmIdeDijw/7u/TSY4VHYrjYWgHA4KqRAEQRCE8eRkmfCHiwMCorZR2eEQczhiyd8QKe3ve9HSAbRzzb/iITioSoUgCIIgEkTloMAxPzRNxOX2j4wHYsvfECktVL6e2gFwsWCxmGDnZrOQ4CAIgiCIOFFWBLYJ1yoIjha+QkVnwbFzb+A4Hg4HEAirmM3AwIL4XDNWSHAQBEEQKY/ZbMKwgf7jmgZAEKRhFTFhFNA3pAIAfS7/vxl2fatTwiGW5g4p8Q+xSwVIcBAEQRBpgRhW6ezxl6vy8D04Ymn6JVJaGLzJx8vdAIDfXwRUDHTjjsXxu2asUOMvgiAIIi2oLAsc1+wDCnMFbKkFbF5ZD4782B2BUoVhafEUHJfNB2ZN2IuKior4XTRGyOEgCIIg0oLKMr5SBbj5nwLGLwIWPzxQ2mVU5xwOkXgKjlSEBAdBEASRFvAOx659wPL3/MdfbHRIEjv1zuEQiUeFSipDgoMgCIJIC/jS2JXrBDS1Br5e/WPgWJ8cjuDHyOEIDwkOgiAIIi2o4OaqfPCt9HtrtgSO9XA48rL9VSk88WhrnsqQ4CAIgiDSgkHFgK2/FMLjlX7P5Q4c65HDYTKZglwOcjjCQ4KDIAiCSAssFpNkemwo9HA4gOCwCgmO8JDgIAiCINIGPnFUCUcG4MjQp1GWPHGUQirhIcFBEARBpA2RBEeRDuEUEbnDIZ/iSkghwUEQBEGkDXwvDiAwRl5Ej/wNEXI4tEGCgyAIgkgbKmQOx6JTpF/rKjhk7c2zM1NjpkmiIMFBEARBpA18SKV8IDB7uvT7eiWMAsBAShrVBAkOgiAIIm0YMwywWPzHJ0wGxldKv69H0y+RoCoVCqmEhYa3EQRBEGnDwEITnroR+PQHAX9cbEJpkYD8bC/au/0qRE+HQ57DQUmj4SHBQRAEQaQVi+aasGiuP5/C5xNQNdiNddv7BUeefnkW5HBog0IqBEEQRFozanCgzaieDkdhLmC1BL6mHI7wkOAgCIIg0pqpo53seFyFfuc1m02SxFESHOGhkApBEASR1px+VDfM9hIU5ppw+Dh9S1dLC4G9B/zHFFIJDwkOgiAIIq2xWoBfnet3JPSGTxwlhyM8FFIhCIIgiCg5e6ZfxEyq8vf9IEJDDgdBEARBRMnlC0yYNRUYXGKMg5JOkOAgCIIgiBioHERCQw0UUiEIgiAIwnBIcBAEQRAEYTgkOAiCIAiCMBwSHARBEARBGA4JDoIgCIIgDIcEB0EQBEEQhkOCgyAIgiAIwyHBQRAEQRCE4ZDgIAiCIAjCcHQVHK2trViyZAmOOeYYnHXWWVizZo2epycIgiAIIkXRVXA88MADKC4uxkcffYQlS5bg97//Pdrb2/W8BEEQBEEQKYhus1R6enrwySef4I033kBmZiZmzpyJkSNH4tNPP8Vpp50mea7L5YLL5ZLeiNUKu92u1+0wfD6f5N90gtaWmtDaUhNaW2qSrmtLpnWZzeq8C90Ex+7du5GVlYXS0lL2WFVVFXbu3Bn03GXLluGJJ56QPHbuuefivPPO0+t2gqirqzPs3ImG1paa0NpSE1pbapKua0uGdQ0fPlzV83QTHL29vcjOzpY8lp2drRhSWbx4MRYuXCi9EQMdjrq6OpSXl6tWYakCrS01obWlJrS21CRd15aK69JNcDgcDnR3d0se6+7uRlZWVtBz7Xa7IeJCCbPZrFp9pRq0ttSE1paa0NpSk3RdWyquSzdZNGzYMPT09KCpqYk9Vl1djREjRuh1CYIgCIIgUhTdBEdWVhZmzpyJxx9/HH19ffj888+xY8cOzJw5U69LEARBEASRopgEQRD0OllrayvuuOMOrFu3DqWlpbjppptwxBFH6HV6giAIgiBSFF0FB0EQBEEQhBKpkdpKEARBEERKQ4KDIAiCIAjDIcFBEARBEIThkOAgCIIgCMJwSHAQBEEQBGE4JDgIgiAIgjAcEhwEQRAEQRhOWggOj8cDAEjHliJKw+8IgtAft9sNAPB6vQm+E/05cOAA+ztJEIkiZQWHIAjo7u7GH/7wB7z88ssAAJPJlOC70o+Ghgb84he/wAMPPICOjo5E346uNDQ04G9/+xteffVVtLS0JPp2dGXfvn14/PHH8c4776TV+9bQ0ID//Oc/+Pjjj9HV1ZXo29ENQRDQ29uLW265BY899hgAwGKxJPiu9KOhoQHXXnst7rnnHjQ2Nib6dnSloaEBDz30EJYtW4bq6upE346u7Nu3D88++yy++OKLRN+KrqSs4DCZTDhw4AA+/vhjfPvtt9iyZQuA9HA5tm7diquuugrl5eW4/vrrkZ2dDSA91vbSSy/hZz/7GTo6OvDyyy9j6dKl2LFjR6JvSxc+/fRTXHDBBWhqasKzzz6L+++/H59++mmibytmHn/8cZx//vnYvXs3/vGPf+DRRx/Ftm3bEn1bumAymeB0OrFmzRqsXbsWa9euBeAf/Z3q1NfX4/rrr8ewYcNwxx13YNCgQQDS4+/I//73P1x44YUQBAHff/89/v73v+Obb75J9G3FhPi+vPPOOzj//POxdetW3H///bjvvvuwYcOGBN+dPug2nj4R1NTUYOTIkaioqMD777+PsWPHpoXLsW7dOsyYMQM33XQTAKC5uRnFxcUpvzaPx4M1a9bghhtuwPz581FdXY377rsPPT09ib41XVi7di3OO+88XHfdddi7dy/ef/99LF++HBMnTkRRUVGiby8qPvnkE2zZsgXLli3DiBEjsGPHDvzlL3/B3r17MXr06ETfni7U1tZizJgxmDp1Kl544QVMmzYNZnPKfhZjrF27FmPGjMHNN98MwC9AhgwZkvJ/RwBgzZo1uPzyy3HhhRdi3759WLp0acqHn8X3Zd26dbjyyitx0UUXYceOHXjllVfw/PPPo6KiAnl5eQm+y9hImf9VYnwVCORsDBkyBBMnTsSwYcOwZ88e9mky1WKw/Nq8Xi86Ojowfvx4bNmyBRdddBFuu+023HPPPVi/fn3ibjJK+LXV19ejtrYWxcXFAICRI0fCarWm/KdJQRDQ19eH7u5u5OTkwOv1YvDgwZg9ezYGDhyIf/7zn4m+xaiprKzEnDlzMGLECLjdblRVVQFAWlnYQ4YMgd1ux6hRo9DX14fXXnsNAFI250H8/+R2uzF06FBs3bqV/R259dZb8d577yX4DmOjubkZzc3NyMjIAAAMGjQIvb297OtUg/89a2lpQWNjI8rKyiAIAqqqqjB79mz4fD68+OKLCbxLfUh6h6OtrQ1/+tOfIAgCxo0bh3PPPReZmZkAgE2bNsHj8WDOnDnYuXMn1q1bhzFjxsBqtaKkpCTBdx6ZUGvbu3cvGhoasH37dixYsADTpk3DM888g//85z/Izc3FyJEjE33rEVFaW0VFBSZMmID3338f27Ztw7p167B161Y8/fTT2Lp1K0466aSUed9eeeUVzJkzB0OHDoXJZEJmZibMZjNqampYDsDgwYNx6qmn4vHHH8eOHTvYZp2syNcF+AVHZWUlAMBqtbJ/xd9BQRBS4hOz0tpEtm/fjqKiIhx11FGoqanBZ599hqlTpyIjIwOlpaVJv0b52kR3pqmpCa2trXjjjTewYMECnHjiiXjllVfw3HPPYejQoZgwYUKC7zwySu9bcXExDjvsMPz000+or69HdXU11q9fD0EQsH79epx11lkoLy9P8J1HprW1FX/7299gMplQVVWFM888E0VFRfD5fFi3bh1mzZoFAJgwYQJ27NiBr776irlUqUpSOxxdXV24/vrrYbVaceyxx2LVqlX4/e9/z5LxKisrYTab4XA4cPjhh+Obb77BhRdeiK+//jrBdx4ZpbXdeOONAIDLLrsMb7/9NjZs2IBzzjkHVVVVuPLKK1FYWIjvv/8+wXcemVBr8/l8uOmmm3DOOefgs88+Q25uLlatWoWTTjoJP/zwQ0rkO3z66ae4/PLL8c9//hMfffQRnE4n+95FF12E999/H5s3bwbgt0grKysxfPhw1NTUJOiO1RFqXWJcWdx0u7q6sG/fPiYMk3kjFgm1NtEJKC8vR1tbG6xWK4466ig0NDTgnHPOwapVqwAk9xrD/T6edtpp+OCDD/D1119j7ty5GDhwIC6++GJMmjQJK1euTOBdqyPc2i699FJcccUV2LVrF7KysvDpp59i4cKFaG9vx+uvv564m1bJW2+9hQsuuABWqxWjRo3CW2+9hXvvvRcAcPnll+PVV19FW1sbAMBms2Hs2LGwWCwpHzZKasGxY8cOWK1W3HXXXZg3bx7+/ve/o76+Hm+++SY8Hg/27dsHANiwYQMeeeQR9PX1YcKECZg+fTqA5E6OUlpbQ0MDnn76aQwZMgRnnHEGWlpa2BqGDh2K3bt3s0+Zqbi2FStWwO12o6CgAC6XCzfeeCNMJhMWLFiAlpYWFgpL5rW1trbiggsuwD333IMPPvgAu3btAuC/ZzH88Oc//5mJ4qFDh6ZEgmWodck3261bt7L/ZwCwfv167N27N+73qwX52nbu3AkAzAmoq6tDUVER1q5dixtvvBFutxtjx47FUUcdBSB1fx+HDh2KM888E319fejs7AQAZGdno62tDfn5+ex5yUqotQH+SiKLxYKenh5cfPHFAIAjjjgCPp8PdrsdQPKuraurCzU1Nbjuuutw880347zzzsMtt9yCDRs2oLm5GVOmTMFRRx2F22+/nf3MhAkTsGHDhpTPd0sqwdHY2Iimpib2dX5+Pmpra5nSczgcuPjii/HZZ59h+/btGD58OF5++WXccsstuPLKK/HHP/4Rdrsdq1evBpBcn0zUru2LL77A9u3bcd1116GrqwtPP/009uzZg/r6eng8HmYVpuLaVq9ejbq6OuTn52PLli3MCairq0NXVxfLok/mtZ1yyik48cQTMXv2bAwZMgRvvPEGurq62D3ffPPN6OrqwuOPP46ffvoJtbW1yMzMxODBgxO1BEXUrktEXF9dXR3mzp2L+vp6XHDBBXjwwQeTLgcn0trefPNNydpGjhyJ119/HXfeeScuv/xyPProo6ioqGC5Dqn8+3jttdeisLAQ//3vf7Fhwwbs3bsXu3fvRkVFBYDUXBvgv++ioiLs2LEDO3bsYB9At23bxkIOybo2h8OBE044ATNnzmTf7+joQH5+PqtIvP3227Fx40Y8+eST2Lt3L6qrqzFs2DCUlZUl5P71wiQkgQx0u9244447sH79egwYMADHHnss5s+fj8LCQtx6662YOHEiLrnkEvb8q6++GtOmTcMFF1yAtWvXYurUqcjNzQUAbNu2Lamy57Wu7ZprrsHEiRNx7bXX4vPPP8eqVatQW1uL2tpa/PznP5c8N9FEs7ZDDz0U11xzDf71r3/h3XffRVlZGbZt25YSa5s3bx4GDx4Mr9cLi8WCHTt24KabbsJvfvMbzJgxAz6fDxaLBT/99BPee+89/Pjjj6ivr8fChQtx6aWXJnpJAKJbl8lkgs/ng9lsxl133YX//e9/yM7OxkUXXYTLL7880UtiRLs2t9uNH374AWPGjGF/R5ItVh7L7+PGjRvx0UcfMQF8wQUXJM3vIxDb2l577TW8+eabsFgsqKmpwcKFC7F48eJEL4mhtLYFCxYw4SCGKtevX48nn3wSf/7zn2E2m2GxWLB69Wq89957qK2tRV1dHRYvXoxFixYleEUxIiQBb775pnDttdcKbrdb+PDDD4Xbb79d+NWvfiUIgiA8/fTTwo033ihUV1ez53/wwQfCaaedJjmH1+uN6z2rRevaPvzwQ2HBggWSc+zYsUPo7e2N632rIZq1nXrqqezrmpoaYdWqVUJ3d3fc7z0S8rXdcccdwi9/+Uv2fY/HIwiCIDz88MPCr371K6GpqSnoHLt370669y3WdS1ZskS499570/Y9S5W/I9Gsbe/evUn3+ygIsa/twIEDwldffZWSv5M+n08QBEF46KGHhHvuuUfxHJs3bxZ6enricr9Gk7CQSl9fH4ux7dixA3l5ebBarTjppJNw+eWXY/v27Xj77bexYMECZGRk4Pnnn2c/m5+fj6qqKvT29rJzJFPdfCxry8vLw+jRo9HT08PKpUaOHMkqcxJNrGsbNWoUuru74fP5UFFRgRNOOAFZWVmJWo6EcGu77LLLUFNTg5deeglAoPT6iiuuwP79+7F+/Xq8/vrreOqpp9j5ysvLk+J902Nd//rXvwAA999/P37/+9+n7XuWKn9HolnboEGDkuL3EdBnbU8++SQAf9XKkUcemZK/k263G4IgYOPGjZg9ezYA4L333pMkvo4dOxYOhyPu6zCCuJfF7t69Gw899BCysrLgcDhw4403Ijc3FxaLBZ2dncjNzUV5eTkuv/xy/O1vf8O8efNw1lln4d5778Vdd92FU045BY899hgmTZqUdG+CnmtLlv88InquTYxTJgtq13bZZZfh73//O84++2yWmJabm4vDDz8ct9xyCzIzM1mztmTAiHUly4aVru8ZQGs72NbW09ODgoICtLW1YcmSJdiwYQOrWEw34irnX3/9dVx99dUYPXo0LrroImzduhVPPfUUqqqq8O2330p6/Z9wwgkoLy/Ha6+9hsMOOwx33XUXMjIy8I9//ANTp07Fb3/723jeekRobX7SeW3HH388RowYgVdeeQWAv2HPP//5T6xYsQJXXXUVPv/8cyxYsCBRS5GQrusCaG0itLb0WNvOnTvx+eef4+6770ZVVRVWrVqFU045JVFLMZZ4xm/+9re/Ca+99hr7ura2Vjj66KMFl8slXHvttcKjjz4qtLa2CoIgCN3d3cKdd94pPPvss5K4qsvliuctq4bW1ioIwsGxthUrVrDY62effSY0Nzcn4M7Dk67rEgRaG60t+Yj2b6QgCMLGjRuFxx9/nH0/nYmrw3H22WezUiC32w2LxYLhw4fD4/Hg8ssvx7p16/DJJ5/A6XQiKysLbW1tKCgokMRVbTZbPG9ZNbS2g2dt+fn5rOTu2GOPTco5Kem6LoDWRmtLPqJdG+DP0bjyyitRUFCQwBXEh7jmcJSWlgLwlwLZbDYcOHAAJpMJdrsdkydPxmmnnYb3338fq1atgsfjwd69ezFu3Lh43mLU0NpobclEuq4LoLXR2pKPaNY2fvx4AMmVqGw0CZmlIirWNWvWYPjw4Wz2xNlnn41jjjkGX3zxBTo7O5OqL4NaaG20tmQiXdcF0NpobclHOq9NDxIiOMRmLtu2bcPJJ58MAHjxxRfR1dWFSy+9FGeffXYibksXaG2pSbquLV3XBdDaUhVa28FLQrwci8UCj8eDvr4+NDY24oorrsDTTz+dEtMLI0FrS03SdW3pui6A1paq0NoOXhI2nn7nzp34+uuvsX37dlx44YVsAE86QGtLTdJ1bem6LoDWlqrQ2g5SElUe43a7hWeffVbo6+tL1C0YBq0tNUnXtaXrugSB1paq0NoOTpJieBtBEARBEOnNwVOPQxAEQRBEwiDBQRAEQRCE4ZDgIAiCIAjCcEhwEARBEARhOCQ4CIIgCIIwHBIcBEEQBEEYDgkOgiAIgiAMhwQHQRCGceWVV2LatGm48sorE30rBEEkGBIcBEEQBEEYDgkOgiAIgiAMJ2HD2wiCSC86Ojpw77334vPPP0dBQQEWL14c9Jy33noLzz33HPbs2QMAGDhwIA455BDcfffd8b5dgiDiDM1SIQhCF373u9/h448/BgBUVFSgqakJANDb24spU6bghhtuwMKFCyEIAsrLy2G327Fv3z709PRg7dq1ibx1giDiADkcBEHEzJ49e5jYWLRoEX75y1+ipqYG559/PntOXV0dBEHAsGHD8PLLL8NsNsPr9eKHH35I1G0TBBFHSHAQBBEz1dXV7PjEE08EAFRWVmLUqFHYsmULAGDSpEnIy8vD7t27cdJJJ2HYsGEYPXo05s6dm5B7JggivlDSKEEQcaGkpAQvvvgifvnLX+LII49ET08PXnvtNVx99dX46aefEn17BEEYDDkcBEHEzIgRI9jxJ598gvHjx6O2thbbt29nj+/fvx9tbW1YtGgRe+ycc85BTU0N1q9fjwkTJsT1ngmCiC8kOAiCiJny8nIcf/zx+OSTT7Bs2TJ8/PHHaGxshMVigdfrBQDs3LkT1113HQoLCzFgwAB0d3ejvr4eAFBVVZXI2ycIIg5QSIUgCF247bbbcOKJJyIjIwNdXV246qqrJK7FkCFDMHv2bGRnZ6O2thatra0YPXo0br31Vhx55JEJvHOCIOIBlcUSBEEQBGE45HAQBEEQBGE4JDgIgiAIgjAcEhwEQRAEQRgOCQ6CIAiCIAyHBAdBEARBEIZDgoMgCIIgCMMhwUEQBEEQhOGQ4CAIgiAIwnBIcBAEQRAEYTgkOAiCIAiCMBwSHARBEARBGM7/A8Gji1vVLP3BAAAAAElFTkSuQmCC",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 640x480 with 1 Axes>"
       ]
@@ -1900,25 +425,15 @@
    "source": [
     "ice_heater = IceCreamHeaterDataset().load()\n",
     "ice_train, ice_test = ice_heater[\"ice cream\"].split_after(0.8)\n",
-    "heater_cov, _ = ice_heater[\"heater\"].split_after(0.8)\n",
+    "heater_past, heater_future = ice_heater[\"heater\"].split_after(0.8)\n",
     "horizon = len(ice_test)\n",
     "\n",
     "model = NeuralProphetDarts(n_lags=horizon, n_forecasts=horizon)\n",
-    "model.fit(ice_train, heater_cov)\n",
-    "preds_cov = model.predict(horizon)\n",
-    "\n",
-    "model = NeuralProphetDarts(n_lags=horizon, n_forecasts=horizon)\n",
-    "model.fit(ice_train)\n",
-    "preds_no_cov = model.predict(horizon)\n",
-    "\n",
-    "print(\"MAPE with lagged regressor: \", mape(ice_test, preds_cov))\n",
-    "print(\"MAPE without lagged regressor: \", mape(ice_test, preds_no_cov))\n",
-    "\n",
-    "# for some reason results vary a lot here for model with lagged regressors\n",
+    "model.fit(ice_train, future_covariates=heater_past)\n",
     "\n",
-    "ice_train.plot(label=\"train\")\n",
-    "preds_cov.plot(label=\"cov\")\n",
-    "preds_no_cov.plot(label=\"no cov\")"
+    "preds = model.predict(horizon, future_covariates=heater_future)\n",
+    "preds.plot()\n",
+    "ice_heater.plot()"
    ]
   }
  ],

From 1d293fe874bfe8f68f7c876abd1c925c037eae3d Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Mon, 19 Dec 2022 15:33:01 +0100
Subject: [PATCH 07/15] Update requirements

---
 requirements/core.txt | 1 +
 1 file changed, 1 insertion(+)

diff --git a/requirements/core.txt b/requirements/core.txt
index 32d3ad8802..15b0e013b6 100644
--- a/requirements/core.txt
+++ b/requirements/core.txt
@@ -8,6 +8,7 @@ numpy>=1.19.0
 pandas>=1.0.5
 pmdarima>=1.8.0
 prophet>=1.1.1
+neuralprophet>=0.4.2
 requests>=2.22.0
 scikit-learn>=1.0.1
 scipy>=1.3.2

From 67ea1d9b58724aefe3f203ecee910a9139bc8415 Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Fri, 23 Dec 2022 09:28:00 +0100
Subject: [PATCH 08/15] Test with newer version

---
 requirements/core.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/requirements/core.txt b/requirements/core.txt
index 15b0e013b6..0fec2f3cf2 100644
--- a/requirements/core.txt
+++ b/requirements/core.txt
@@ -8,7 +8,7 @@ numpy>=1.19.0
 pandas>=1.0.5
 pmdarima>=1.8.0
 prophet>=1.1.1
-neuralprophet>=0.4.2
+neuralprophet>=0.5.0
 requests>=2.22.0
 scikit-learn>=1.0.1
 scipy>=1.3.2

From b180e877b43e4a6a81b52bdb5485c7e5ccf4edbf Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Fri, 23 Dec 2022 09:43:50 +0100
Subject: [PATCH 09/15] Test rollback

---
 requirements/core.txt | 1 -
 1 file changed, 1 deletion(-)

diff --git a/requirements/core.txt b/requirements/core.txt
index 9285b875ba..ccc0ebe5f4 100644
--- a/requirements/core.txt
+++ b/requirements/core.txt
@@ -8,7 +8,6 @@ numpy>=1.19.0
 pandas>=1.0.5
 pmdarima>=1.8.0
 prophet>=1.1.1
-neuralprophet>=0.5.0
 pyod>=0.9.5
 requests>=2.22.0
 scikit-learn>=1.0.1

From 632b850482f793fa39381127e8ab2f7f5dab2526 Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Fri, 23 Dec 2022 10:18:51 +0100
Subject: [PATCH 10/15] Manually add tensorboardX

---
 requirements/core.txt  | 1 +
 requirements/torch.txt | 1 +
 2 files changed, 2 insertions(+)

diff --git a/requirements/core.txt b/requirements/core.txt
index ccc0ebe5f4..9285b875ba 100644
--- a/requirements/core.txt
+++ b/requirements/core.txt
@@ -8,6 +8,7 @@ numpy>=1.19.0
 pandas>=1.0.5
 pmdarima>=1.8.0
 prophet>=1.1.1
+neuralprophet>=0.5.0
 pyod>=0.9.5
 requests>=2.22.0
 scikit-learn>=1.0.1
diff --git a/requirements/torch.txt b/requirements/torch.txt
index 621887fd74..0442a38c12 100644
--- a/requirements/torch.txt
+++ b/requirements/torch.txt
@@ -1,2 +1,3 @@
 pytorch-lightning>=1.5.0
 torch>=1.8.0
+tensorboardX>=2.5.1
\ No newline at end of file

From 14f215aebb03ef4ae548f874e58e5d65464abeed Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Fri, 23 Dec 2022 10:37:52 +0100
Subject: [PATCH 11/15] Remove unused imports

---
 examples/04-RNN-examples.ipynb         | 5407 ++++++++++++++++++++++--
 examples/06-Transformer-examples.ipynb |   10 +-
 examples/08-DeepAR-examples.ipynb      |   10 +-
 requirements/torch.txt                 |    3 +-
 4 files changed, 5179 insertions(+), 251 deletions(-)

diff --git a/examples/04-RNN-examples.ipynb b/examples/04-RNN-examples.ipynb
index ffee0bf3f2..5d73e99b85 100644
--- a/examples/04-RNN-examples.ipynb
+++ b/examples/04-RNN-examples.ipynb
@@ -49,7 +49,6 @@
     "from sklearn.preprocessing import MinMaxScaler\n",
     "from tqdm import tqdm_notebook as tqdm\n",
     "\n",
-    "from tensorboardX import SummaryWriter\n",
     "import matplotlib.pyplot as plt\n",
     "\n",
     "from darts import TimeSeries\n",
@@ -177,32 +176,4692 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "67f94871ce5541c59f858b450b669fa5",
+       "model_id": "c7a2b0b218f2416db5b36aa52eff22d6",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "  0%|          | 0/300 [00:00<?, ?it/s]"
+       "Sanity Checking: 0it [00:00, ?it/s]"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b9f8919931bc49698190a2549d1e0ad6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Training: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8e3a06bf76514bac9e6f91520017de19",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "66475d0902a143b4a5a88044f4cf3549",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6ed5f7f4f2a04c0294205b719ae8f99c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8e2decec69404a3a91399ddb9dc495a6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5d11c47b9b184251bcba1cb772119660",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0a2731eef0414fe1ae2bb6abdeeef714",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "11ffd467e6d448af8d3a615d011d67c8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "df940bc838d045b19f8d2b8e99788ff4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "546df9e867df4672bae60d8b28b991f0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0960b43274b5473e82bacf010ec4807f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4fb34d611c374ff88a8f919494b3316d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "71f174e53fc24968a81d55a127ae55d5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7e9f628e17934f4cba676f31d08f707a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "598b58d70bce41b893457a2b76869f2a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fc2a21f161aa40b98d81d78859570b77",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cbeff95d9cbd48fea15a44a5f2511f85",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d98198a025dd42b59b158f4b29ecd949",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ea1d84f4d1464111a12470d1b8ccf0e9",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6e2bcf143531447fb4bcb86f23ea9102",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "115b53a9936944ebabbf8302fa0989bb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a2d339410c084d63a55e709ac86aa9e2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0b50ba5e3a26423bb289a576077f9d41",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2a892240a03e4cf9a25ba518e9c269b7",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f08fbba0dc67495790efdf6588b089ee",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ef6a691e3b864e90a8fb3b6be54fe9d6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4658b5a2966942ceb1d18556ee1b72f0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a411cd355b4c47d9984b07693e26fbe3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ec97aa9dc2fa4c2197a3a3db4863d424",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "71dfadde525a4a51865f384bb7f08fcd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "aad92c19817443aeb3677bff3e26a2f4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "95c5c18ee500477d9527b4f330dce8be",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4d049b01d4374fca9c3c419f52232eea",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "69495091c7a348a3b752834fd2fd77cb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a30798535b794a38be09f0ae21e9ca36",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cde87b67ee87470dbcfa20bebc21c821",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "abffaceb4d6e4160aad935ddc58d5ecc",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "88d5c1f2ea9745c18128209f484d08d1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e8b7b82080054562a5d9063aaaef1887",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3954935dda9c4b1c9a57a97b9adca746",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "94b95e7c22a04797b4b74a1f3e5a5502",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "143f4638cca940fbbfaf18db75362695",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8717d88b1e574c848d7bfa2f7b3dd2e0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "bbd45a2894d944fdb27f23697606ac56",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8d5253cb9b83449a8bf085218901a7d6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2804f8b0de194a05ab6adc1b8e9c54ea",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "45fdfa03c5bc4b1f8a4124f1e3a6c895",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d23abc1282da4047bfccb2c8b80f5524",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cb513f17313740199ea9d18c82b61a14",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cad338261c6d4ca89679fa23637f54d6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1e7fd1f0faa6489583d51c8e899e4250",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ab031c6264ea400ba226cf7bab610e2a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "dccdb930b93049abb28007bae2e8055f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5073b0cb2bfb4f1da8aa994afe229add",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3c7781262f6b460b90622c13cfeb32b7",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "bcd5882d30a74c68bcc34cbbcbae8b9d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c3a4e8c1326347dab6477a011ae8e66f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0a382ffe80f94bd2a14675094a8f9f50",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "02a42d5969cb4330b409f8efc29c0513",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c069b8c73af24d829962f774a11e32aa",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5f9fcef5de934c849bd554c98a7d42dd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5258630080f94c24a20da23b7a0cc8b8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2d259ede15ab4ab187af485b4421fdce",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fda44dde08504077bda88bf63cab6c9d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7bb9e8eeb43a4e0e8567225050eb6e3e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f4c4a118de7a4dd9aa579f8ca6fb760c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "56262d5d771e44218bd566803eed140f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e8201aa82caf4e2894b8f58482019c86",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "540db41815824aafba7375a512030fa3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5f55ba0905be470886213d0d90e8fa5f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "11ac34e355bb496883f14b3a11701dfd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6e549cdaa6b74db5b24003682ad3a4cc",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8f6aa4c07a1a43418f1165ed3c9622a9",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b569370c27644a6a84bc48c815b16df3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4ff4e200f30342e094b2e01e1f1e4b76",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "44bedf27335d47b4964973d72fd44f72",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "94a7a11572c0401ea5636e077f1f2a98",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4d0a14179b7b4ce288c63cb50420a951",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "23c88f36efc3464f98e67ede34be4f5a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d0566b047fb04c1f81e08772bfbe6a2a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1c329717e3e949429d2161a7fb0c41f0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "11e6b0bd575c45159fa083f7fc1c43e8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6a4f675a34774281ba567d30dfee46a7",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ce28c6b71e6543e5b342f835b6c0301e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d78c5a629431459aa8f50dacd584cca9",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5299ce9c5bdd4ce8927e9d9b631d53c4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0f0ea40ba66d4d2fb5b5369d048210d0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e581ebe5b0504c04a2bbaa2eba7a1db7",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7b3805c34d3a4999965ffd2e65c0e140",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4547118afe67422ea049b356049ffc7d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "25394712755c47ad9294dc1567a08645",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "14d11bdcec224906bac596d53ced1a71",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ff28c714648741b7aae33003f2172617",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "737076a0e4524ceba90848244df7cd1d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "43c91d944b8d43e9a863294d85c09122",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e14460a7bd1842e988be749a8b901222",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c3e28e62edcc4517a733951b6b951b42",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b7462f91e1004ba199fea0b3a7a771bc",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f1c453a42b1e46bcb8f6cbf587c5004b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9475d5a5c1e94f319ba009a5e30b5cad",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1d68d44e49b040a29b07cae5adc9147a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ee1815d7355d4b04b264de0b8a4cc974",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ca5894574606425a9b74e756c276cbb2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a5a11a951e014bdeb64675b6c5bde63f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7d0cc33fb1c449b2902809be38080a47",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "74c45b99a6cc49de8ae026f991b06dc0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f60384809baa4126be53a3c222847ef6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "bccdcdd9d6ef46c5991d7bce918ce762",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d0b8f0212cca48b48c3ae9e7eca0319b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fae4eda597c74a338732423e6d2e822e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6fc5b8dc7a6449f5883b369e2f5d208b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "94d1f31ab7f94b3dba1e2931bf947ba1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ff107cc427b8430e8a644752aed9c7e5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cdb267245dfd46a28a3e9e11c40b372b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7a25392fe06941428bc7a19c3e503bbd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6b5460e73c8d4d0ea72011a4e6b74265",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "42d45d7adbac45b794a4f8de6d4f5c92",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "efbe1f4a4b38495eaeaf0dd6ca1c307f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "339f28522e284b1fa3ad9e01ae42ff95",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0cd4f70557454077a757fc0fb3bc3210",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4fded45496ea4d6c93e48c846983dc89",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "124b8af31209409b8e0ddef4156a512a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b516abc5f6834110a9cbdf8d3113ad45",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1e6f3eeb3aa242e984912d04564e4e91",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "67223694ae874c63868257638182d398",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fa35d716206f4699b76b333de729ca97",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ddcc518a0e6549f093bab66640388010",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "be7f62e5023b4a76873c032f691e5af1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2c8c84f3b023457f96f4442e35975d1d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fb18f7695da6427dbefcdc88248afd37",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3ba9aa5920144a7d82490af041c5cb41",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a27ffb2af68d49b5ad4b65da8505fe44",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4f72831cd0694d01828447ef95d62aa4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5693dad51fc3426c9cb114d8fd97acf3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1c611627188e4178a3d5cdd302b1d945",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "edb73b00c02843dfa23d7562f49a0b5c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d921948ba84b4bb3b86a4cd204b5b31a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "28a514c9c8bc4d6da12c4158b0fc1264",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "486f9807cf144630aac5b9d84f59e373",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "584a4d011b0c4ed1b0921fc9f44a682e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a5d1f5cc4a544e90a67d10119a28ba7c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2398b6e478b2489fa2b95becadd021be",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e7dcef61094e4749bdca6dbb509f80bf",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fd996b3870554324b178d81adaf4156d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e415e6b0e7c9464a933c456c99e247eb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3c16168c6b294e43906b196fb7797b76",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "46adbca22bad4d46af57db493f1e09eb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "51e55f4a437e43b8b7f10a5717c40699",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "366c7440c0cd42108f0aac17525da564",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c1ba705cd1e5468b8fa988a39c7ef813",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "07d237a3065a4802b5708aea234d9cce",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "208ab33aa0dc461693c0e0f3d0b8f385",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ed99a1686e7e4312a6499db2196ddc2e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ad76bc09b44440dd89499e3775138071",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b158f404494a42b19e60fc68f3e2f6e6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ceaf0403f1204650b565aeecd3b1b84a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7f1893f9fa814ca1910777f79baec55d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e2096e8e8c9840169829b14a6322f6ef",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d20bcd20696349208b9c7dea003bda62",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "92cd48f838884f9dafb3ee625ed19d7a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "71be8611e81f4014a54e81323a9bb545",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "42cfe5a59ff740eb8645f3162721c93a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "61333e94dab44e7ca9e2cdf4e6308ece",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "262addb6339a47dc8eed6b3e63edd9cd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a40efa8b623347cc844b7af3c47f2f7f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "30aa06990879489cb7774838240b4659",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6c405b7203854f238f08b7d5a33f2967",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "006510203f1c4bbd9b6a4bafc17d9072",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b5f2c1998c724c509b76ccc3a136e7d2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1b8e6fc3c0514ecca279f335775c017d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f0b0c693deea47ef89cde48d763fda2e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "bd0924fc31644743b67170b80f17fed2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6537e582d33a446bab166e8da78da21f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b6cf8ac857124faba021ab72f81f9444",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5844e9399e4b4a33a45d1447dd4129b8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7f7cd00b585f4dd6b0ee5bdb526b230a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "57fca4204e8144b69e46dbbbeb213770",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f1320d0156d14d05af3544134a9dfacd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5db556c0ffc64c1cb1e0da5f1d02c1f8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "847d31fb63aa4f71a0b77db06d0d220b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7544f67b54614cedb0dff0026f5f768e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "179625b3cbe6438292a8280971d51fcc",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "73ac8264e0744727adb2d9e3b10ac65c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c3605ca96d66423aabc2b6d3fb564600",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a5db4888e7a24c94a8f3f5cb1594ef40",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d639bdaccbb3478bafe0cc6fb7cf5b78",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b9fe9bcfbf444f67a65719ca26d83310",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3ed8520d85a14a068023fca88049c8b4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cb93e7a37ddb4aafa942107ca16f6e3c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "aa6566831bd94790ad4b4fa40e19ce67",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2103c9af6950436eb2792d2898bbcf7f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "209fe5d37d984e1f92e8a42c04ed0122",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "83b2f77afaa54837ad8fe231251354d1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3a8923cce08942ef84793095a7153a57",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fd2181e60a52475aadded34cc073e697",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cb0287a9af3d4355995e36049be82a4e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f3c310e58e97435ea5e623932e5035f5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "95a2daa1127a49afb52e094f890e7d86",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c711ce407fbb4defbca12d2afee285a0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "41ea9e6d7618476bb973908e97d257e6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b2da3aca958842a2b2e1f81bd03d3273",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ef2ba7bd9b13401f81c4d5b3ca44daf8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b98fcceba36740939a76bd75b75b68ac",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4de6995e31374f0ebf889933084792db",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "bb04b39a393147feb5b00caf08fdfee4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c648868406bc49879145f4f3b7222337",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "36935abada854a48a870f3b9d12e1028",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8d6a0efa8a35432f888ed3817610af5d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "17ba1f1f24cb44ce9efca853a17d2f67",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6ae01e9de998449cb00386834189fd72",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "12c05281982440a6a371a7210e657d53",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e335073a456a45d9a35f7b788704b9a4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "33b526363bcd49ccb99157e35f0b8246",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c056fc078e7a40b7a2f03436e349c95c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "944e4da732e143b1a022f1bbb65349eb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9329d52741cd49b5a4aad34f235042f5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1cf31d0a49af418ab16647e1ffce0b76",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f343bb36c0bf42fbb59fef924eee03cd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7c250a34bd3f4192891b051c11724ef0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2d4950cf98d84cc588dc75966a505e45",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "18ee9c687e134a09853fec79f4442b05",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c38226c5c4e34ff994b6666daf6df1d1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4ad4be58d3a2429098c34e5c945de80d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "987b7615b72e47019e5690ffab9ab310",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d6dc0ddcaeb34048aa38d7e2e890069d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5bb35074cc9b47f8b3dd027279c5b9a8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "bf1d2218ffc745e38930bdc3dd861cca",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "88585797975349bbbf9133a304a39ba8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7b9e4f1ddcb240e8809fc7a1b7e6229e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6ca9577c5f9c466c96b370f1ed10c274",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c82cdf35d527454cb4547623b93a30ab",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "dccfd9968dc9438bb90722eae58e4302",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9d81a32c78d04205b4a9c9f1de70a9a1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0ccffd7c4a6a4679adc97a7d799d024d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a82f82ad50ca474fa112195a273e689d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "da7f6dfb8ec74232b064a0d874098239",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "56aee7947968450e90d5873d88d1e7f5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "710bedcfdc2a4e7f8110be10aee229fc",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c3edbcf0e8d5461ab018400d8903a9b2",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c265293ac8ad45468efbd6ada40e2527",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "beeec4f0cd2a405ba2555e4c7bf6b7c4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1fed89bb315541aeb89761c2648350c3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "941cba3083394fbc9347810428200ec0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f62991edebef47449cd8e1efe26712ad",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1463461321a04f8bacd4b370e72c9f83",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9f0e4d8ef87e4f2b816d20dc5a29f575",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "773eab53e031492b9a5873095484aa2f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "240192d392dd44b4a9aff393f3fa56fe",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7c547a2f1f01499e99dea03969d80485",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "661890776ab24351a57e9c6a96271209",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1ec4a7eea94e4ae58c3e2a3779c4c947",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c5cea38d6bd348e98a565a238c9cef0d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5e8e0e5f9e9d4114833b2ef188c832b4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7c642631044541979cf05adee88f584f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4feccd7e38e4476882010f7b7d4d4779",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "aa84acfe997043a5a80d449f5dfb57cf",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "fcd4ac27f9ad482282fcf6ae84716c87",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3a9f8d3a2fbe4d3c92d83160bd22f5b1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4ead7bc17ce74f08872af3e107af37c5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3c2623f0e3db4ddba68f92bbe234daec",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d3b7b2d6d1f64a3caefe754bae4bbafa",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c7de3ab3671a402c8bd5dd1dcb285a0c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6757e2c68a29416ababdcd6cce86cfec",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f5ee8061564a427f9913c4816f9fa835",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "344f35ea564443c382cbc74d7a9ce5b6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f890f3758e924a36aa96845039464f5b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "66607a21fe4e4e3094ddf6b6a5c36e6c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5e48c940e86f416e8671a9cb073a8405",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "94da7b88361449cf87cb9f4305b248a3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ae5d61d985da4c6f8c8b56639b2fc692",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3d24db062a3d422798517308c7cfb2cf",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "10c463476a024f0884664ca990c077bc",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "878ee172ce5d465886c0acd44b90df90",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e3fa7cece5f64179aa1cfc4f3d27eacb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b0974b8c828d458db07c1aec8ee0c394",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7ebf78b97309447b87784c2931364d29",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a3b73d1159f0427d8ad758256f10714a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4ee222c4cfb24bc381237f7c68d73346",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "98b238887bb94bce9969e3a56dbce768",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b513c98520d34892b2a6c853bff8d283",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "88b7971202df42d6a74af290b34522f3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "53eb71f54bb849d1b14c5034dbd64604",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d1a9bf1f03e64b0fb06754e02ede30f3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "54f4ab408aff4ba68e47d5fbf6dea0ab",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4132bbcf19724b83b2499c27c6a62cb4",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "25e179f218294387a806175da8302ac1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4e02718a0d974485b6f7d161d3e69081",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4fa1ee149bf844a3b28ae78654d07403",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b91b11378a574459a8e78af36b139b46",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "786a8da3d6a04ef5b4b005f7f510ac6c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "33289f3a28a649939dbf1a733cfe1223",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3293ff7388b84be5ab7b50689924b359",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f2168225758e449b9bd26586dfd0f2f0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6deeaa0e1a574873ae66dd3f6da6ba02",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "83944599bd1d48c49518e37e785c78de",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c35e51f2cd974ab58cf0c4059dc23e7f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "71dbdd8673b54d35ba4fe159ac14d850",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f88ada84c8224cc785be2f79acfe4074",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0041158c057e4d5c954e71276279ff5b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "615bacf3bbad4d7fa3aff53798e92e7d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "00a11e3179a34b8e9fdb94feca5fadc1",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<darts.models.forecasting.rnn_model.RNNModel at 0x7fc24a2538e0>"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "my_model.fit(\n",
+    "    train_transformed,\n",
+    "    future_covariates=covariates,\n",
+    "    val_series=val_transformed,\n",
+    "    val_future_covariates=covariates,\n",
+    "    verbose=True,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Look at predictions on the validation set\n",
+    "Use the \"current\" model - i.e., the model at the end of the training procedure:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "889a59d2364442e2b20d3a404b80a7f7",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "def eval_model(model):\n",
+    "    pred_series = model.predict(n=26, future_covariates=covariates)\n",
+    "    plt.figure(figsize=(8, 5))\n",
+    "    series_transformed.plot(label=\"actual\")\n",
+    "    pred_series.plot(label=\"forecast\")\n",
+    "    plt.title(\"MAPE: {:.2f}%\".format(mape(pred_series, val_transformed)))\n",
+    "    plt.legend()\n",
+    "\n",
+    "\n",
+    "eval_model(my_model)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Use the best model obtained over training, according to validation loss:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "26b9c1aad455499b84cb70faf032fa96",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "best_model = RNNModel.load_from_checkpoint(model_name=\"Air_RNN\", best=True)\n",
+    "eval_model(best_model)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Backtesting\n",
+    "Let's backtest our `RNN` model, to see how it performs at a forecast horizon of 6 months:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9728af169af34a5a81dc269068311d1b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "  0%|          | 0/19 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ab1c1e8f99404778911eb8c430c1e877",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "c94ad799621b4bf6b00675f7e668e69f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "81e7b6f37af34766bdbafe6e6fb2103c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d45d1dccd370425c843b811bfe3c425d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "89f412e3ea154d1e9c4e074eebf0cf75",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3400cdf5d2e64c06b23a50e6bcb0c391",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "851e13108fd041e0a7f9a0feb4c92fbd",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "362ec7c2447042de9a56d484422d2258",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e08208e0c8eb4654b1df92c36c54cc33",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "94d16a9bb675442e86e8305f3c0fba12",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "41a386c8e07145399a9505a069024e09",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b51977de317e4903ae1b1e1eb60d07ee",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b38ddffd53ba44b38f0266f60b1a5244",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "96825d3bd1bf49b1babbb4f7f1e6fdbb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a1cbd8b561e54ead95e89f26341a391c",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "70adb57a62cb40f294746c16f130ab93",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "93aa4e07e7a84445b52e80ef56712a58",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "47a9c9c72c114d35b97ff9a50f50b276",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "827209f55b4b4a7da548f1173d6d5291",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Predicting: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "backtest_series = my_model.historical_forecasts(\n",
+    "    series_transformed,\n",
+    "    future_covariates=covariates,\n",
+    "    start=pd.Timestamp(\"19590101\"),\n",
+    "    forecast_horizon=6,\n",
+    "    retrain=False,\n",
+    "    verbose=True,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Training loss: 0.0003, validation loss: 0.0021, best val loss: 0.0021\r"
+      "MAPE: 2.71%\n"
      ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     }
    ],
    "source": [
-    "my_model.fit(\n",
-    "    train_transformed,\n",
-    "    future_covariates=covariates,\n",
-    "    val_series=val_transformed,\n",
-    "    val_future_covariates=covariates,\n",
-    "    verbose=True,\n",
+    "plt.figure(figsize=(8, 5))\n",
+    "series_transformed.plot(label=\"actual\")\n",
+    "backtest_series.plot(label=\"backtest\")\n",
+    "plt.legend()\n",
+    "plt.title(\"Backtest, starting Jan 1959, 6-months horizon\")\n",
+    "print(\n",
+    "    \"MAPE: {:.2f}%\".format(\n",
+    "        mape(\n",
+    "            transformer.inverse_transform(series_transformed),\n",
+    "            transformer.inverse_transform(backtest_series),\n",
+    "        )\n",
+    "    )\n",
     ")"
    ]
   },
@@ -210,59 +4869,444 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Look at predictions on the validation set\n",
-    "Use the \"current\" model - i.e., the model at the end of the training procedure:"
+    "## Monthly sunspots\n",
+    "Let's now try a more challenging time series; that of the monthly number of sunspots since 1749. First, we build the time series from the data, and check its periodicity."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(True, 125)"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "series_sunspot = SunspotsDataset().load()\n",
+    "\n",
+    "series_sunspot.plot()\n",
+    "check_seasonality(series_sunspot, max_lag=240)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x500 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plot_acf(series_sunspot, 125, max_lag=240)  # ~11 years seasonality"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
    "metadata": {},
+   "outputs": [],
+   "source": [
+    "train_sp, val_sp = series_sunspot.split_after(pd.Timestamp(\"19401001\"))\n",
+    "\n",
+    "transformer_sunspot = Scaler()\n",
+    "train_sp_transformed = transformer_sunspot.fit_transform(train_sp)\n",
+    "val_sp_transformed = transformer_sunspot.transform(val_sp)\n",
+    "series_sp_transformed = transformer_sunspot.transform(series_sunspot)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {
+    "scrolled": true,
+    "tags": []
+   },
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e538227a1a7c42178beeb5cee8f15093",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Sanity Checking: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0aaa3486197746c3b8712936764b7b80",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Training: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f6ab1f9e0c38415d8ed3b958964363be",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1eadbd799c9a4032bb43de381b06949d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a216f8cce56a4fe9aba29dd1fa14724d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "19ca383944f24a28bf0409a7339a4189",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cdbf1d12e7d1485dac86d09ac9ed0c0a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "29b956d2e39e4c1ba3efc30a32790aea",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "33d0f5675ece456ab6b220fdae8a4e45",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9687ef2aba614d3aaa47be9955b872d9",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "50ec10b43f93401b9f7583e1d49532ad",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e6a66d5813fe414d8b3cdaed52644703",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "95515d97500349e7a9e940979eaab0c6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0997672c2c8b4041b9fb46d45630f0fe",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8d73fe1c244044dead4789cde9e65d1d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "3fcb9619b62a415fa4308bce13b29d53",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9845d8e613dd42e192e524cb2fdb8f9e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "46172cb139bd4108ad888326e1baee44",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "64c21241e79c4091b8b1352361a81515",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "5fd008cb92c64aa895ff3eb9c85e104e",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "618a6a325bbf42ba95a528fcf3318b78",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a77dc34ad45e46218e65936b56b7e5d0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "dab02faba9834a21950ee089ee5ce8eb",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b28a222ffedd411d841f805e42a5baf3",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4197f211aa714c1a8a07d582d4b12f8a",
+       "version_major": 2,
+       "version_minor": 0
+      },
       "text/plain": [
-       "<Figure size 576x360 with 1 Axes>"
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
-     "metadata": {
-      "needs_background": "light"
-     },
+     "metadata": {},
      "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "def eval_model(model):\n",
-    "    pred_series = model.predict(n=26, future_covariates=covariates)\n",
-    "    plt.figure(figsize=(8, 5))\n",
-    "    series_transformed.plot(label=\"actual\")\n",
-    "    pred_series.plot(label=\"forecast\")\n",
-    "    plt.title(\"MAPE: {:.2f}%\".format(mape(pred_series, val_transformed)))\n",
-    "    plt.legend()\n",
-    "\n",
-    "\n",
-    "eval_model(my_model)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "Use the best model obtained over training, according to validation loss:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {
-    "tags": []
-   },
-   "outputs": [
+    },
     {
      "data": {
-      "application/vnd.code.notebook.stdout": [
-       "loading model_best_290.pth.tar\n"
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "f68c64575c904adba089df32816c2254",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
      "metadata": {},
@@ -270,205 +5314,125 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6231cbcbcfe44fa89eeb1d192b94e301",
+       "version_major": 2,
+       "version_minor": 0
+      },
       "text/plain": [
-       "<Figure size 576x360 with 1 Axes>"
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "best_model = RNNModel.load_from_checkpoint(model_name=\"Air_RNN\", best=True)\n",
-    "eval_model(best_model)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "### Backtesting\n",
-    "Let's backtest our `RNN` model, to see how it performs at a forecast horizon of 6 months:"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {
-    "tags": []
-   },
-   "outputs": [
+    },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "1da9ae4ed7244f24bfbca41d85e51dbc",
+       "model_id": "00f257f6493b4ae6b28234d14f5939a2",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "  0%|          | 0/19 [00:00<?, ?it/s]"
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "backtest_series = my_model.historical_forecasts(\n",
-    "    series_transformed,\n",
-    "    future_covariates=covariates,\n",
-    "    start=pd.Timestamp(\"19590101\"),\n",
-    "    forecast_horizon=6,\n",
-    "    retrain=False,\n",
-    "    verbose=True,\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "metadata": {
-    "tags": []
-   },
-   "outputs": [
+    },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "MAPE: 2.43%\n"
-     ]
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "15899f7724b94f418ce6330555c08552",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Validation: 0it [00:00, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
     },
     {
      "data": {
-      "image/png": "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",
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "34b1e13992fb425ca6f7695210203b6f",
+       "version_major": 2,
+       "version_minor": 0
+      },
       "text/plain": [
-       "<Figure size 576x360 with 1 Axes>"
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "plt.figure(figsize=(8, 5))\n",
-    "series_transformed.plot(label=\"actual\")\n",
-    "backtest_series.plot(label=\"backtest\")\n",
-    "plt.legend()\n",
-    "plt.title(\"Backtest, starting Jan 1959, 6-months horizon\")\n",
-    "print(\n",
-    "    \"MAPE: {:.2f}%\".format(\n",
-    "        mape(\n",
-    "            transformer.inverse_transform(series_transformed),\n",
-    "            transformer.inverse_transform(backtest_series),\n",
-    "        )\n",
-    "    )\n",
-    ")"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## Monthly sunspots\n",
-    "Let's now try a more challenging time series; that of the monthly number of sunspots since 1749. First, we build the time series from the data, and check its periodicity."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "metadata": {},
-   "outputs": [
+    },
     {
      "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "1a402be5c8644c5e9754057e30ef6d4d",
+       "version_major": 2,
+       "version_minor": 0
+      },
       "text/plain": [
-       "(True, 125)"
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
-     "execution_count": 11,
      "metadata": {},
-     "output_type": "execute_result"
+     "output_type": "display_data"
     },
     {
      "data": {
-      "image/png": "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",
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d7655be515b74a17906e4a8addffcd90",
+       "version_major": 2,
+       "version_minor": 0
+      },
       "text/plain": [
-       "<Figure size 432x288 with 1 Axes>"
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "series_sunspot = SunspotsDataset().load()\n",
-    "\n",
-    "series_sunspot.plot()\n",
-    "check_seasonality(series_sunspot, max_lag=240)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "metadata": {},
-   "outputs": [
+    },
     {
      "data": {
-      "image/png": "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",
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6aacc12a087749b6b5b0487c0ee4847d",
+       "version_major": 2,
+       "version_minor": 0
+      },
       "text/plain": [
-       "<Figure size 720x360 with 1 Axes>"
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "plot_acf(series_sunspot, 125, max_lag=240)  # ~11 years seasonality"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "train_sp, val_sp = series_sunspot.split_after(pd.Timestamp(\"19401001\"))\n",
-    "\n",
-    "transformer_sunspot = Scaler()\n",
-    "train_sp_transformed = transformer_sunspot.fit_transform(train_sp)\n",
-    "val_sp_transformed = transformer_sunspot.transform(val_sp)\n",
-    "series_sp_transformed = transformer_sunspot.transform(series_sunspot)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {
-    "scrolled": true,
-    "tags": []
-   },
-   "outputs": [
+    },
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "5f159c4b53ca4bf0a2687c68292b9fc3",
+       "model_id": "837e280161a143fdb4a1749cbde1486a",
        "version_major": 2,
        "version_minor": 0
       },
       "text/plain": [
-       "  0%|          | 0/100 [00:00<?, ?it/s]"
+       "Validation: 0it [00:00, ?it/s]"
       ]
      },
      "metadata": {},
      "output_type": "display_data"
     },
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Training loss: 0.0087, validation loss: 0.0208, best val loss: 0.0208\r"
-     ]
+     "data": {
+      "text/plain": [
+       "<darts.models.forecasting.block_rnn_model.BlockRNNModel at 0x7fc26f9c8c70>"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [
@@ -501,38 +5465,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "538b4ddfae114d4eb971440bfc42be63",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/49 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "application/vnd.jupyter.widget-view+json": {
-       "model_id": "5e86ab53af1247b98b4d4bb77db2cad8",
-       "version_major": 2,
-       "version_minor": 0
-      },
-      "text/plain": [
-       "  0%|          | 0/49 [00:00<?, ?it/s]"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "# Compute the backtest predictions with the two models\n",
     "pred_series = my_model_sun.historical_forecasts(\n",
@@ -558,28 +5493,9 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "RNN MAPE: 71.86777453068686\n",
-      "ETS MAPE: 116.63591326597222\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 432x288 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
     "val_sp_transformed.plot(label=\"actual\")\n",
     "pred_series.plot(label=\"our RNN\")\n",
@@ -592,7 +5508,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "Python 3.9.15 ('darts')",
    "language": "python",
    "name": "python3"
   },
@@ -606,7 +5522,12 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.11"
+   "version": "3.9.15"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "679853c720796769b87b8a23d7322371925113d3257e0f93c82aede63e3ea4ff"
+   }
   }
  },
  "nbformat": 4,
diff --git a/examples/06-Transformer-examples.ipynb b/examples/06-Transformer-examples.ipynb
index 584a53d03e..f4e61493f5 100644
--- a/examples/06-Transformer-examples.ipynb
+++ b/examples/06-Transformer-examples.ipynb
@@ -49,7 +49,6 @@
     "from sklearn.preprocessing import MinMaxScaler\n",
     "from tqdm import tqdm_notebook as tqdm\n",
     "\n",
-    "from tensorboardX import SummaryWriter\n",
     "import matplotlib.pyplot as plt\n",
     "\n",
     "from darts import TimeSeries\n",
@@ -567,7 +566,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3.9.15 ('prophet')",
    "language": "python",
    "name": "python3"
   },
@@ -581,7 +580,12 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.5"
+   "version": "3.9.15"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "2f14e79e1646dc5b749c3dc6e0dfef5e568c2efea6b930caf0398818dd8806ea"
+   }
   }
  },
  "nbformat": 4,
diff --git a/examples/08-DeepAR-examples.ipynb b/examples/08-DeepAR-examples.ipynb
index 3511ffbad5..43a59c41ee 100644
--- a/examples/08-DeepAR-examples.ipynb
+++ b/examples/08-DeepAR-examples.ipynb
@@ -49,7 +49,6 @@
     "from sklearn.preprocessing import MinMaxScaler\n",
     "from tqdm import tqdm_notebook as tqdm\n",
     "\n",
-    "from tensorboardX import SummaryWriter\n",
     "import matplotlib.pyplot as plt\n",
     "\n",
     "from darts import TimeSeries\n",
@@ -370,7 +369,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3.9.15 ('prophet')",
    "language": "python",
    "name": "python3"
   },
@@ -384,7 +383,12 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.5"
+   "version": "3.9.15"
+  },
+  "vscode": {
+   "interpreter": {
+    "hash": "2f14e79e1646dc5b749c3dc6e0dfef5e568c2efea6b930caf0398818dd8806ea"
+   }
   }
  },
  "nbformat": 4,
diff --git a/requirements/torch.txt b/requirements/torch.txt
index 0442a38c12..458e09f5bc 100644
--- a/requirements/torch.txt
+++ b/requirements/torch.txt
@@ -1,3 +1,2 @@
 pytorch-lightning>=1.5.0
-torch>=1.8.0
-tensorboardX>=2.5.1
\ No newline at end of file
+torch>=1.8.0
\ No newline at end of file

From 78c5e760d6a84a61c6d2c76974ac715aafcfc707 Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Tue, 7 Mar 2023 14:35:09 +0100
Subject: [PATCH 12/15] Require neural prophet with updated requirements

---
 requirements/core.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/requirements/core.txt b/requirements/core.txt
index ba6db036d4..0ef8f03594 100644
--- a/requirements/core.txt
+++ b/requirements/core.txt
@@ -8,7 +8,7 @@ numpy>=1.19.0
 pandas>=1.0.5
 pmdarima>=1.8.0
 prophet>=1.1.1
-neuralprophet>=0.5.0
+neuralprophet>=0.5.2
 pyod>=0.9.5
 requests>=2.22.0
 scikit-learn>=1.0.1

From 4116e17ad3b839a745353032c9e7aa0c5f3f6af9 Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Wed, 8 Mar 2023 15:15:17 +0100
Subject: [PATCH 13/15] Revert changes from the notebooks

---
 examples/06-Transformer-examples.ipynb | 10 +++-------
 examples/08-DeepAR-examples.ipynb      | 10 +++-------
 2 files changed, 6 insertions(+), 14 deletions(-)

diff --git a/examples/06-Transformer-examples.ipynb b/examples/06-Transformer-examples.ipynb
index 02a61cd9a9..dcee081139 100644
--- a/examples/06-Transformer-examples.ipynb
+++ b/examples/06-Transformer-examples.ipynb
@@ -49,6 +49,7 @@
     "from sklearn.preprocessing import MinMaxScaler\n",
     "from tqdm import tqdm_notebook as tqdm\n",
     "\n",
+    "from tensorboardX import SummaryWriter\n",
     "import matplotlib.pyplot as plt\n",
     "\n",
     "from darts import TimeSeries\n",
@@ -566,7 +567,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3.9.15 ('prophet')",
+   "display_name": "Python 3",
    "language": "python",
    "name": "python3"
   },
@@ -580,12 +581,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.15"
-  },
-  "vscode": {
-   "interpreter": {
-    "hash": "2f14e79e1646dc5b749c3dc6e0dfef5e568c2efea6b930caf0398818dd8806ea"
-   }
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,
diff --git a/examples/08-DeepAR-examples.ipynb b/examples/08-DeepAR-examples.ipynb
index 43a59c41ee..3511ffbad5 100644
--- a/examples/08-DeepAR-examples.ipynb
+++ b/examples/08-DeepAR-examples.ipynb
@@ -49,6 +49,7 @@
     "from sklearn.preprocessing import MinMaxScaler\n",
     "from tqdm import tqdm_notebook as tqdm\n",
     "\n",
+    "from tensorboardX import SummaryWriter\n",
     "import matplotlib.pyplot as plt\n",
     "\n",
     "from darts import TimeSeries\n",
@@ -369,7 +370,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3.9.15 ('prophet')",
+   "display_name": "Python 3",
    "language": "python",
    "name": "python3"
   },
@@ -383,12 +384,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.9.15"
-  },
-  "vscode": {
-   "interpreter": {
-    "hash": "2f14e79e1646dc5b749c3dc6e0dfef5e568c2efea6b930caf0398818dd8806ea"
-   }
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,

From 8c449b1f4982ca1d51a6be216834d72bf9705393 Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Wed, 8 Mar 2023 15:16:28 +0100
Subject: [PATCH 14/15] Add model import in module init file

---
 darts/models/__init__.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/darts/models/__init__.py b/darts/models/__init__.py
index 60ac60b572..450b26c113 100644
--- a/darts/models/__init__.py
+++ b/darts/models/__init__.py
@@ -15,6 +15,7 @@
 from darts.models.forecasting.fft import FFT
 from darts.models.forecasting.kalman_forecaster import KalmanForecaster
 from darts.models.forecasting.linear_regression_model import LinearRegressionModel
+from darts.models.forecasting.neural_prophet_model import NeuralProphet
 from darts.models.forecasting.random_forest import RandomForest
 from darts.models.forecasting.regression_ensemble_model import RegressionEnsembleModel
 from darts.models.forecasting.regression_model import RegressionModel

From 35b88a6bf7708bf3cacc54f73f818806488e160b Mon Sep 17 00:00:00 2001
From: Blazej Nowicki <blazej.nowicki@unit8.co>
Date: Wed, 8 Mar 2023 16:43:51 +0100
Subject: [PATCH 15/15] Add docstring

---
 .../forecasting/neural_prophet_model.py       | 62 +++++++++++++++++--
 1 file changed, 58 insertions(+), 4 deletions(-)

diff --git a/darts/models/forecasting/neural_prophet_model.py b/darts/models/forecasting/neural_prophet_model.py
index dd3eed7142..08c3649cf2 100644
--- a/darts/models/forecasting/neural_prophet_model.py
+++ b/darts/models/forecasting/neural_prophet_model.py
@@ -4,7 +4,7 @@
 """
 
 import warnings
-from typing import Optional, Sequence, Union
+from typing import Dict, List, Optional, Sequence, Tuple, Union
 
 import neuralprophet
 import pandas as pd
@@ -16,8 +16,50 @@
 
 
 class NeuralProphet(ForecastingModel):
-    def __init__(self, n_lags: int = 0, n_forecasts: int = 1, **kwargs):
-        super().__init__()
+    def __init__(
+        self,
+        n_lags: int = 0,
+        n_forecasts: int = 1,
+        add_encoders: Optional[Dict] = None,
+        **kwargs,
+    ):
+        """Neural Prophet
+
+        This class provides a basic wrapper around `NeuralProphet <https://github.com/ourownstory/neural_prophet>`_.
+        It extends approach similar to Facebook Prophet model with auto-regressive feed-forward neural network
+        It supports also supports past and future covariates. For more parameters refer to the original documentation.
+
+        Parameters
+        ----------
+        n_lags
+            Number of lagged values provided to AR-Net. If equal to 0 then only trend
+            and seasonality will be used for forecasting.
+
+        n_forecast
+            Output size chunk of the AR-Net. Limits how far into the future is is possible to forecast.
+
+        add_encoders
+            A large number of future covariates can be automatically generated with `add_encoders`.
+            This can be done by adding multiple pre-defined index encoders and/or custom user-made functions that
+            will be used as index encoders. Additionally, a transformer such as Darts' :class:`Scaler` can be added to
+            transform the generated covariates. This happens all under one hood and only needs to be specified at
+            model creation.
+            Read :meth:`SequentialEncoder <darts.dataprocessing.encoders.SequentialEncoder>` to find out more about
+            ``add_encoders``. Default: ``None``. An example showing some of ``add_encoders`` features:
+
+            .. highlight:: python
+            .. code-block:: python
+
+                add_encoders={
+                    'cyclic': {'future': ['month']},
+                    'datetime_attribute': {'future': ['hour', 'dayofweek']},
+                    'position': {'future': ['relative']},
+                    'custom': {'future': [lambda idx: (idx.year - 1950) / 50]},
+                    'transformer': Scaler()
+                }
+            ..
+        """
+        super().__init__(add_encoders=add_encoders, **kwargs)
         # TODO improve passing arguments to the model
 
         raise_if_not(n_lags >= 0, "Argument n_lags should be a non-negative integer")
@@ -43,7 +85,7 @@ def fit(
 
         raise_if_not(
             past_covariates is None or self.n_lags > 0,
-            "Past covariates are only supported when auto-regression is enabled (n_lags > 1)",
+            "Past covariates are only supported when auto-regression is enabled (n_lags > 0)",
         )
 
         self.training_series = series
@@ -226,5 +268,17 @@ def _future_covariates_checks(self, future_covariates: Optional[TimeSeries]):
     def uses_future_covariates(self):
         return True
 
+    def _model_encoder_settings(
+        self,
+    ) -> Tuple[
+        Optional[int],
+        Optional[int],
+        bool,
+        bool,
+        Optional[List[int]],
+        Optional[List[int]],
+    ]:
+        return (None, None, True, True, None, None)
+
     def __str__(self):
         return "Neural Prophet"