modeling.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1f1ba075",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import xgboost\n",
    "from sklearn.model_selection import TimeSeriesSplit, cross_val_score\n",
    "from xgboost import XGBRegressor, XGBRFRegressor\n",
    "from lightgbm import LGBMRegressor\n",
    "from catboost import CatBoostRegressor\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_percentage_error\n",
    "from sktime.performance_metrics.forecasting import MeanAbsoluteScaledError\n",
    "from sklearn.dummy import DummyRegressor\n",
    "from copy import deepcopy\n",
    "import shap\n",
    "import datetime\n",
    "\n",
    "sns.set(rc={'figure.figsize':(16,9)})\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2024476",
   "metadata": {},
   "source": [
    "# 0.1 Global  Variables\n",
    "\n",
    "Section reserved for global/config variables \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d9707214",
   "metadata": {},
   "outputs": [],
   "source": [
    "METRICS = [mean_squared_error, mean_absolute_percentage_error]\n",
    "SCORING = \"neg_mean_absolute_percentage_error\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "20a126a4",
   "metadata": {},
   "source": [
    "# 0.2 Functions Definition \n",
    "\n",
    "Section reserved for functions that will be used through the jupyter file\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7b174fb3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def cross_train(models,cv,X_train,Y_train):\n",
    "    best_model_score = -np.inf\n",
    "    best_model = None\n",
    "    for idx,(name,model) in enumerate(models):\n",
    "        result = cross_val_score(model, X_train, Y_train,  scoring=SCORING, cv=cv)\n",
    "        names.append(name)\n",
    "        results.append(result)\n",
    "        if result.mean() > best_model_score:\n",
    "            best_model = idx\n",
    "            best_model_score = result.mean()\n",
    "            \n",
    "\n",
    "    for i in range(len(names)):\n",
    "        print(names[i],results[i].mean())\n",
    "     \n",
    "    \n",
    "    print(\"Best Model:\", names[best_model])\n",
    "    return best_model\n",
    "\n",
    "def infer(model, X_test):\n",
    "    \n",
    "    inference_results = []\n",
    "    y_pred = model.predict(X_test)\n",
    "    \n",
    "\n",
    "    return y_pred\n",
    "\n",
    "def smape(target, preds):\n",
    "    n = len(preds)\n",
    "    masked_arr = ~((preds == 0) & (target == 0))\n",
    "    preds, target = preds[masked_arr], target[masked_arr]\n",
    "    num = np.abs(preds - target)\n",
    "    denom = np.abs(preds) + np.abs(target)\n",
    "    smape_val = (200 * np.sum(num / denom)) / n\n",
    "    return smape_val\n",
    "\n",
    "METRICS.append(smape)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d88f7429",
   "metadata": {},
   "source": [
    "# 1. Temporal K-Fold \n",
    "\n",
    "As we are dealing with temporal data, we can't use the standard K-Fold methodology to validate our model. The problem with standard K-Fold is the **Data Leakage** (Usage of training data *after* the test data) and unnatural behavior of the model, we can't use future to predict the past. Therefore, we have to use a special K-Fold technique to validate our model **Temporal K-Fold**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "bdf0b3b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>total_accidents</th>\n",
       "      <th>day_of_the_month</th>\n",
       "      <th>total_accidents_lag_1</th>\n",
       "      <th>total_accidents_lag_2</th>\n",
       "      <th>total_accidents_lag_3</th>\n",
       "      <th>total_accidents_lag_4</th>\n",
       "      <th>total_accidents_lag_5</th>\n",
       "      <th>total_accidents_lag_6</th>\n",
       "      <th>total_accidents_lag_7</th>\n",
       "      <th>...</th>\n",
       "      <th>month_9</th>\n",
       "      <th>month_10</th>\n",
       "      <th>month_11</th>\n",
       "      <th>month_12</th>\n",
       "      <th>week_of_the_month_0</th>\n",
       "      <th>week_of_the_month_1</th>\n",
       "      <th>week_of_the_month_2</th>\n",
       "      <th>week_of_the_month_3</th>\n",
       "      <th>week_of_the_month_4</th>\n",
       "      <th>week_of_the_month_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2015-06-01</td>\n",
       "      <td>39</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2015-06-02</td>\n",
       "      <td>49</td>\n",
       "      <td>2</td>\n",
       "      <td>39.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2015-06-03</td>\n",
       "      <td>46</td>\n",
       "      <td>3</td>\n",
       "      <td>49.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2015-06-04</td>\n",
       "      <td>28</td>\n",
       "      <td>4</td>\n",
       "      <td>46.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015-06-05</td>\n",
       "      <td>40</td>\n",
       "      <td>5</td>\n",
       "      <td>28.0</td>\n",
       "      <td>46.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1670</th>\n",
       "      <td>2019-12-27</td>\n",
       "      <td>37</td>\n",
       "      <td>27</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1671</th>\n",
       "      <td>2019-12-28</td>\n",
       "      <td>27</td>\n",
       "      <td>28</td>\n",
       "      <td>37.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1672</th>\n",
       "      <td>2019-12-29</td>\n",
       "      <td>15</td>\n",
       "      <td>29</td>\n",
       "      <td>27.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1673</th>\n",
       "      <td>2019-12-30</td>\n",
       "      <td>23</td>\n",
       "      <td>30</td>\n",
       "      <td>15.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1674</th>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>10</td>\n",
       "      <td>31</td>\n",
       "      <td>23.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1675 rows × 69 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            date  total_accidents  day_of_the_month  total_accidents_lag_1  \\\n",
       "0     2015-06-01               39                 1                    NaN   \n",
       "1     2015-06-02               49                 2                   39.0   \n",
       "2     2015-06-03               46                 3                   49.0   \n",
       "3     2015-06-04               28                 4                   46.0   \n",
       "4     2015-06-05               40                 5                   28.0   \n",
       "...          ...              ...               ...                    ...   \n",
       "1670  2019-12-27               37                27                   42.0   \n",
       "1671  2019-12-28               27                28                   37.0   \n",
       "1672  2019-12-29               15                29                   27.0   \n",
       "1673  2019-12-30               23                30                   15.0   \n",
       "1674  2019-12-31               10                31                   23.0   \n",
       "\n",
       "      total_accidents_lag_2  total_accidents_lag_3  total_accidents_lag_4  \\\n",
       "0                       NaN                    NaN                    NaN   \n",
       "1                       NaN                    NaN                    NaN   \n",
       "2                      39.0                    NaN                    NaN   \n",
       "3                      49.0                   39.0                    NaN   \n",
       "4                      46.0                   49.0                   39.0   \n",
       "...                     ...                    ...                    ...   \n",
       "1670                   14.0                   33.0                   44.0   \n",
       "1671                   42.0                   14.0                   33.0   \n",
       "1672                   37.0                   42.0                   14.0   \n",
       "1673                   27.0                   37.0                   42.0   \n",
       "1674                   15.0                   27.0                   37.0   \n",
       "\n",
       "      total_accidents_lag_5  total_accidents_lag_6  total_accidents_lag_7  \\\n",
       "0                       NaN                    NaN                    NaN   \n",
       "1                       NaN                    NaN                    NaN   \n",
       "2                       NaN                    NaN                    NaN   \n",
       "3                       NaN                    NaN                    NaN   \n",
       "4                       NaN                    NaN                    NaN   \n",
       "...                     ...                    ...                    ...   \n",
       "1670                   21.0                   34.0                   50.0   \n",
       "1671                   44.0                   21.0                   34.0   \n",
       "1672                   33.0                   44.0                   21.0   \n",
       "1673                   14.0                   33.0                   44.0   \n",
       "1674                   42.0                   14.0                   33.0   \n",
       "\n",
       "      ...  month_9  month_10  month_11  month_12  week_of_the_month_0  \\\n",
       "0     ...        0         0         0         0                    0   \n",
       "1     ...        0         0         0         0                    0   \n",
       "2     ...        0         0         0         0                    0   \n",
       "3     ...        0         0         0         0                    0   \n",
       "4     ...        0         0         0         0                    0   \n",
       "...   ...      ...       ...       ...       ...                  ...   \n",
       "1670  ...        0         0         0         1                    0   \n",
       "1671  ...        0         0         0         1                    0   \n",
       "1672  ...        0         0         0         1                    0   \n",
       "1673  ...        0         0         0         1                    0   \n",
       "1674  ...        0         0         0         1                    0   \n",
       "\n",
       "      week_of_the_month_1  week_of_the_month_2  week_of_the_month_3  \\\n",
       "0                       1                    0                    0   \n",
       "1                       1                    0                    0   \n",
       "2                       1                    0                    0   \n",
       "3                       1                    0                    0   \n",
       "4                       1                    0                    0   \n",
       "...                   ...                  ...                  ...   \n",
       "1670                    0                    0                    0   \n",
       "1671                    0                    0                    0   \n",
       "1672                    0                    0                    0   \n",
       "1673                    0                    0                    0   \n",
       "1674                    0                    0                    0   \n",
       "\n",
       "      week_of_the_month_4  week_of_the_month_5  \n",
       "0                       0                    0  \n",
       "1                       0                    0  \n",
       "2                       0                    0  \n",
       "3                       0                    0  \n",
       "4                       0                    0  \n",
       "...                   ...                  ...  \n",
       "1670                    1                    0  \n",
       "1671                    1                    0  \n",
       "1672                    1                    0  \n",
       "1673                    0                    1  \n",
       "1674                    0                    1  \n",
       "\n",
       "[1675 rows x 69 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"./dataset/primary.csv\")\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "413b559d",
   "metadata": {},
   "source": [
    "**We will remove 2019 in our data, because it will our out-of-sample data to test our model**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "377719a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"date\"] = pd.to_datetime(df[\"date\"])\n",
    "train, test = df[df[\"date\"].dt.year < 2019],df[df[\"date\"].dt.year >= 2019]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "fa799a13",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1310, 69)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "aa5ef418",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>total_accidents</th>\n",
       "      <th>day_of_the_month</th>\n",
       "      <th>total_accidents_lag_1</th>\n",
       "      <th>total_accidents_lag_2</th>\n",
       "      <th>total_accidents_lag_3</th>\n",
       "      <th>total_accidents_lag_4</th>\n",
       "      <th>total_accidents_lag_5</th>\n",
       "      <th>total_accidents_lag_6</th>\n",
       "      <th>total_accidents_lag_7</th>\n",
       "      <th>...</th>\n",
       "      <th>month_9</th>\n",
       "      <th>month_10</th>\n",
       "      <th>month_11</th>\n",
       "      <th>month_12</th>\n",
       "      <th>week_of_the_month_0</th>\n",
       "      <th>week_of_the_month_1</th>\n",
       "      <th>week_of_the_month_2</th>\n",
       "      <th>week_of_the_month_3</th>\n",
       "      <th>week_of_the_month_4</th>\n",
       "      <th>week_of_the_month_5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1310</th>\n",
       "      <td>2019-01-01</td>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>16.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1311</th>\n",
       "      <td>2019-01-02</td>\n",
       "      <td>33</td>\n",
       "      <td>2</td>\n",
       "      <td>14.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1312</th>\n",
       "      <td>2019-01-03</td>\n",
       "      <td>29</td>\n",
       "      <td>3</td>\n",
       "      <td>33.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>32.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1313</th>\n",
       "      <td>2019-01-04</td>\n",
       "      <td>23</td>\n",
       "      <td>4</td>\n",
       "      <td>29.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>43.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1314</th>\n",
       "      <td>2019-01-05</td>\n",
       "      <td>26</td>\n",
       "      <td>5</td>\n",
       "      <td>23.0</td>\n",
       "      <td>29.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1670</th>\n",
       "      <td>2019-12-27</td>\n",
       "      <td>37</td>\n",
       "      <td>27</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>50.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1671</th>\n",
       "      <td>2019-12-28</td>\n",
       "      <td>27</td>\n",
       "      <td>28</td>\n",
       "      <td>37.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1672</th>\n",
       "      <td>2019-12-29</td>\n",
       "      <td>15</td>\n",
       "      <td>29</td>\n",
       "      <td>27.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1673</th>\n",
       "      <td>2019-12-30</td>\n",
       "      <td>23</td>\n",
       "      <td>30</td>\n",
       "      <td>15.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>44.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1674</th>\n",
       "      <td>2019-12-31</td>\n",
       "      <td>10</td>\n",
       "      <td>31</td>\n",
       "      <td>23.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>27.0</td>\n",
       "      <td>37.0</td>\n",
       "      <td>42.0</td>\n",
       "      <td>14.0</td>\n",
       "      <td>33.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>365 rows × 69 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           date  total_accidents  day_of_the_month  total_accidents_lag_1  \\\n",
       "1310 2019-01-01               14                 1                   16.0   \n",
       "1311 2019-01-02               33                 2                   14.0   \n",
       "1312 2019-01-03               29                 3                   33.0   \n",
       "1313 2019-01-04               23                 4                   29.0   \n",
       "1314 2019-01-05               26                 5                   23.0   \n",
       "...         ...              ...               ...                    ...   \n",
       "1670 2019-12-27               37                27                   42.0   \n",
       "1671 2019-12-28               27                28                   37.0   \n",
       "1672 2019-12-29               15                29                   27.0   \n",
       "1673 2019-12-30               23                30                   15.0   \n",
       "1674 2019-12-31               10                31                   23.0   \n",
       "\n",
       "      total_accidents_lag_2  total_accidents_lag_3  total_accidents_lag_4  \\\n",
       "1310                   13.0                   18.0                   43.0   \n",
       "1311                   16.0                   13.0                   18.0   \n",
       "1312                   14.0                   16.0                   13.0   \n",
       "1313                   33.0                   14.0                   16.0   \n",
       "1314                   29.0                   33.0                   14.0   \n",
       "...                     ...                    ...                    ...   \n",
       "1670                   14.0                   33.0                   44.0   \n",
       "1671                   42.0                   14.0                   33.0   \n",
       "1672                   37.0                   42.0                   14.0   \n",
       "1673                   27.0                   37.0                   42.0   \n",
       "1674                   15.0                   27.0                   37.0   \n",
       "\n",
       "      total_accidents_lag_5  total_accidents_lag_6  total_accidents_lag_7  \\\n",
       "1310                   32.0                   27.0                   13.0   \n",
       "1311                   43.0                   32.0                   27.0   \n",
       "1312                   18.0                   43.0                   32.0   \n",
       "1313                   13.0                   18.0                   43.0   \n",
       "1314                   16.0                   13.0                   18.0   \n",
       "...                     ...                    ...                    ...   \n",
       "1670                   21.0                   34.0                   50.0   \n",
       "1671                   44.0                   21.0                   34.0   \n",
       "1672                   33.0                   44.0                   21.0   \n",
       "1673                   14.0                   33.0                   44.0   \n",
       "1674                   42.0                   14.0                   33.0   \n",
       "\n",
       "      ...  month_9  month_10  month_11  month_12  week_of_the_month_0  \\\n",
       "1310  ...        0         0         0         0                    1   \n",
       "1311  ...        0         0         0         0                    1   \n",
       "1312  ...        0         0         0         0                    1   \n",
       "1313  ...        0         0         0         0                    1   \n",
       "1314  ...        0         0         0         0                    1   \n",
       "...   ...      ...       ...       ...       ...                  ...   \n",
       "1670  ...        0         0         0         1                    0   \n",
       "1671  ...        0         0         0         1                    0   \n",
       "1672  ...        0         0         0         1                    0   \n",
       "1673  ...        0         0         0         1                    0   \n",
       "1674  ...        0         0         0         1                    0   \n",
       "\n",
       "      week_of_the_month_1  week_of_the_month_2  week_of_the_month_3  \\\n",
       "1310                    0                    0                    0   \n",
       "1311                    0                    0                    0   \n",
       "1312                    0                    0                    0   \n",
       "1313                    0                    0                    0   \n",
       "1314                    0                    0                    0   \n",
       "...                   ...                  ...                  ...   \n",
       "1670                    0                    0                    0   \n",
       "1671                    0                    0                    0   \n",
       "1672                    0                    0                    0   \n",
       "1673                    0                    0                    0   \n",
       "1674                    0                    0                    0   \n",
       "\n",
       "      week_of_the_month_4  week_of_the_month_5  \n",
       "1310                    0                    0  \n",
       "1311                    0                    0  \n",
       "1312                    0                    0  \n",
       "1313                    0                    0  \n",
       "1314                    0                    0  \n",
       "...                   ...                  ...  \n",
       "1670                    1                    0  \n",
       "1671                    1                    0  \n",
       "1672                    1                    0  \n",
       "1673                    0                    1  \n",
       "1674                    0                    1  \n",
       "\n",
       "[365 rows x 69 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "673297fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "cv = TimeSeriesSplit(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "26443620",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[((120,), (119,)),\n",
       " ((239,), (119,)),\n",
       " ((358,), (119,)),\n",
       " ((477,), (119,)),\n",
       " ((596,), (119,)),\n",
       " ((715,), (119,)),\n",
       " ((834,), (119,)),\n",
       " ((953,), (119,)),\n",
       " ((1072,), (119,)),\n",
       " ((1191,), (119,))]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating expanding windows \n",
    "[(el[0].shape, el[1].shape) for el in cv.split(train)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "053df5df",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating model\n",
    "xgb = XGBRegressor()\n",
    "X_train = train.drop(columns=[\"date\",\"total_accidents\"]).values\n",
    "Y_train = train[\"total_accidents\"].values\n",
    "X_test = test.drop(columns=[\"date\",\"total_accidents\"]).values\n",
    "Y_test = test[\"total_accidents\"].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2088627d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 26.4 s, sys: 85.6 ms, total: 26.5 s\n",
      "Wall time: 2.47 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "cv_scores = cross_val_score(xgb, X_train, Y_train, cv=cv, \n",
    "                            scoring=SCORING, n_jobs=1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "0bd2c328",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([-0.36455083, -0.46240373, -0.27590307, -0.19881295, -0.31193501,\n",
       "        -0.20510137, -0.29093712, -0.25890438, -0.37815769, -0.25350498]),\n",
       " -0.300021113301684)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv_scores, cv_scores.mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8393138d",
   "metadata": {},
   "source": [
    "**Seems that all things are working. Next section will develop a model selection based on multiple models and multiple metrics. This test above is only a sanity check** "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e92f9426",
   "metadata": {},
   "source": [
    "# 2. Model Selection \n",
    "\n",
    "## Conclusions\n",
    "\n",
    "It seems that LGBM performs better than other algorithms in cross-validation. Although, the other algorithms are close to the LGBM, with difference less than 0.1, we have to pick one to understand the most importante features.  A DummyRegressor was added only for baselines. MAPE shows 24% error and we can use this metric as principal metric because we do not have 0 values on test and neither extreme values. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "e79636c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "models = []\n",
    "results = []\n",
    "names = []\n",
    "\n",
    "models.append((\"XGBoostRF\",XGBRFRegressor()))\n",
    "models.append((\"XGBoost\",XGBRegressor()))\n",
    "models.append((\"LGBM\",LGBMRegressor()))\n",
    "models.append((\"CatBoost\",CatBoostRegressor(silent=True)))\n",
    "models.append((\"DummyRegressorMean\", DummyRegressor(strategy=\"mean\")))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0539f3c6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XGBoostRF -0.2973368482949419\n",
      "XGBoost -0.300021113301684\n",
      "LGBM -0.29543210465172065\n",
      "CatBoost -0.2957550689593118\n",
      "DummyRegressorMean -0.3954595668013417\n",
      "Best Model: LGBM\n"
     ]
    }
   ],
   "source": [
    "winner = cross_train(models,cv,X_train,Y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "449cd55d",
   "metadata": {},
   "source": [
    "**Fit the best model with the whole data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "3dde1413",
   "metadata": {},
   "outputs": [],
   "source": [
    "model_fitted = models[winner][1].fit(X_train,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "aa13c330",
   "metadata": {},
   "outputs": [],
   "source": [
    "preds = infer(model_fitted, X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9c31a29",
   "metadata": {},
   "source": [
    "## 2.1 Measure results "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e8bb9f52",
   "metadata": {},
   "outputs": [],
   "source": [
    "metric_results = []\n",
    "for metric in METRICS:\n",
    "    metric_value = metric(Y_test, preds)\n",
    "    metric_results.append(metric_value)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c06b239c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(data=[metric_results], columns=[metric.__name__ for metric in METRICS])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "e2c1afc6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>mean_squared_error</th>\n",
       "      <th>mean_absolute_percentage_error</th>\n",
       "      <th>smape</th>\n",
       "      <th>model</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>75.270992</td>\n",
       "      <td>0.240852</td>\n",
       "      <td>21.094886</td>\n",
       "      <td>LGBM</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   mean_squared_error  mean_absolute_percentage_error      smape model\n",
       "0           75.270992                        0.240852  21.094886  LGBM"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"model\"] = models[winner][0]\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "da7b79c0",
   "metadata": {},
   "source": [
    "**Seems like LGBM with random forest has the best result regarding forecasting year 2019**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "22112788",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>total_accidents</th>\n",
       "      <th>label</th>\n",
       "      <th>day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1310</th>\n",
       "      <td>14</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1311</th>\n",
       "      <td>33</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1312</th>\n",
       "      <td>29</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1313</th>\n",
       "      <td>23</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1314</th>\n",
       "      <td>26</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1670</th>\n",
       "      <td>37</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1671</th>\n",
       "      <td>27</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1672</th>\n",
       "      <td>15</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1673</th>\n",
       "      <td>23</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1674</th>\n",
       "      <td>10</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-31</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>365 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      total_accidents label        day\n",
       "1310               14  True 2019-01-01\n",
       "1311               33  True 2019-01-02\n",
       "1312               29  True 2019-01-03\n",
       "1313               23  True 2019-01-04\n",
       "1314               26  True 2019-01-05\n",
       "...               ...   ...        ...\n",
       "1670               37  True 2019-12-27\n",
       "1671               27  True 2019-12-28\n",
       "1672               15  True 2019-12-29\n",
       "1673               23  True 2019-12-30\n",
       "1674               10  True 2019-12-31\n",
       "\n",
       "[365 rows x 3 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compare_df = test[[\"total_accidents\"]]\n",
    "compare_df[\"label\"] = \"True\"\n",
    "compare_df[\"day\"] = pd.date_range(\"2019-01-01\", \"2019-12-31\", freq=\"d\")\n",
    "compare_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "a82267de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>total_accidents</th>\n",
       "      <th>label</th>\n",
       "      <th>day</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>14.000000</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33.000000</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>29.000000</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>23.000000</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>26.000000</td>\n",
       "      <td>True</td>\n",
       "      <td>2019-01-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>725</th>\n",
       "      <td>32.970074</td>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-12-27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>726</th>\n",
       "      <td>23.276949</td>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-12-28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>727</th>\n",
       "      <td>18.547169</td>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-12-29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>728</th>\n",
       "      <td>27.956541</td>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-12-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>729</th>\n",
       "      <td>32.762806</td>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-12-31</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>730 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     total_accidents label        day\n",
       "0          14.000000  True 2019-01-01\n",
       "1          33.000000  True 2019-01-02\n",
       "2          29.000000  True 2019-01-03\n",
       "3          23.000000  True 2019-01-04\n",
       "4          26.000000  True 2019-01-05\n",
       "..               ...   ...        ...\n",
       "725        32.970074  LGBM 2019-12-27\n",
       "726        23.276949  LGBM 2019-12-28\n",
       "727        18.547169  LGBM 2019-12-29\n",
       "728        27.956541  LGBM 2019-12-30\n",
       "729        32.762806  LGBM 2019-12-31\n",
       "\n",
       "[730 rows x 3 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_result = pd.DataFrame(columns=[\"total_accidents\",\"label\",\"day\"])\n",
    "df_result[\"total_accidents\"] = preds\n",
    "df_result[\"label\"] = models[winner][0]\n",
    "df_result[\"day\"] = pd.date_range(\"2019-01-01\", \"2019-12-31\", freq=\"d\")\n",
    "    \n",
    "df_total_result = pd.concat([compare_df,df_result],ignore_index=True)\n",
    "df_total_result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "5592a634",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>label</th>\n",
       "      <th>day</th>\n",
       "      <th>total_accidents</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-01-06</td>\n",
       "      <td>30.090249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-01-13</td>\n",
       "      <td>27.013378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-01-20</td>\n",
       "      <td>29.121975</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-01-27</td>\n",
       "      <td>28.938439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LGBM</td>\n",
       "      <td>2019-02-03</td>\n",
       "      <td>30.050424</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>101</th>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-08</td>\n",
       "      <td>50.714286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-15</td>\n",
       "      <td>41.571429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>103</th>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-22</td>\n",
       "      <td>41.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>104</th>\n",
       "      <td>True</td>\n",
       "      <td>2019-12-29</td>\n",
       "      <td>30.285714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>105</th>\n",
       "      <td>True</td>\n",
       "      <td>2020-01-05</td>\n",
       "      <td>16.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>106 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    label        day  total_accidents\n",
       "0    LGBM 2019-01-06        30.090249\n",
       "1    LGBM 2019-01-13        27.013378\n",
       "2    LGBM 2019-01-20        29.121975\n",
       "3    LGBM 2019-01-27        28.938439\n",
       "4    LGBM 2019-02-03        30.050424\n",
       "..    ...        ...              ...\n",
       "101  True 2019-12-08        50.714286\n",
       "102  True 2019-12-15        41.571429\n",
       "103  True 2019-12-22        41.000000\n",
       "104  True 2019-12-29        30.285714\n",
       "105  True 2020-01-05        16.500000\n",
       "\n",
       "[106 rows x 3 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_total_result.set_index(\"day\").groupby(\"label\").resample(\"W\").mean().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "9f9c9cd4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='day', ylabel='total_accidents'>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=df_total_result.set_index(\"day\").groupby(\"label\").resample(\"W\").mean().reset_index(), x=\"day\", y=\"total_accidents\", hue=\"label\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2d6e2ff7",
   "metadata": {},
   "source": [
    "# 3 Interpretation - Shap\n",
    "\n",
    "In this section we will see the importance/shap value for each one of the 10 top most importante features. This gave us the power to interpret the model and make some analysis related to results. We will choose only the best model according to the calculed metrics above\n",
    "\n",
    "## Conclusions\n",
    "\n",
    "- Seems that feature is_weekend is the most important feature\n",
    "    - When is_weekend is high (bool == 1) we have less accidents. Probably people spend more time on their houses and don't have to commute to the work (Less cars in the strees)\n",
    "- total_acidentes_lag_1 (t-1), when day before has a high number of accident, we have more accidents in a day t. Probably because an street or avenue have a problem or maybe hidden road sign.\n",
    "- Day of the year shows that high values, more accidents. It the trends we see before on data exploration. In the second semester we have more holidays, probably people drink more and as we said before, alcohool and drink do not mix well.\n",
    "- Day of the week 5(Saturday), probably is a combination with alcohool and parties proably\n",
    "- day_of_the_month, in the beginning of the month people probably have more money available to put gas on car and hang gout\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "719a1c49",
   "metadata": {},
   "outputs": [],
   "source": [
    "explainer = shap.TreeExplainer(model_fitted)\n",
    "Xd = xgboost.DMatrix(X_train, label=Y_train)\n",
    "shap_values = explainer.shap_values(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "e9b4e53a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x684 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shap.summary_plot(shap_values, train.drop(columns=[\"date\",\"total_accidents\"]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "53d3c3b9",
   "metadata": {},
   "source": [
    "# 4. Model Degradation \n",
    "\n",
    "In this section, we will measure how much time is necessary to model lose its quality/performance. We will use all the training dataset to fit the model and create a test set with monthly gap. First gap we will be zero, so we do not have interval between training and testing datasets. Next iteration gap == 1, so we will jump one month and pick next ones and creating testing set, until we see a big degradation on metrics we are evaluating. \n",
    "\n",
    "## Conclusions\n",
    "\n",
    "In the final of 5 months, model degrades approximate 10% related to mase metric. The right time to retrain the model, or a upper bound for degradation, is related to business problem, infrastructure and computional cost for it. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "948cf011",
   "metadata": {},
   "outputs": [],
   "source": [
    "if winner == 0:\n",
    "    model_not_yet_fitted = XGBRFRegressor()\n",
    "if winner == 1:\n",
    "    model_not_yet_fitted = XGBRegressor()\n",
    "if winner == 2:\n",
    "    model_not_yet_fitted = LGBMRegressor()\n",
    "if winner == 3:\n",
    "    model_not_yet_fitted = CatBoostRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "c866498d",
   "metadata": {},
   "outputs": [],
   "source": [
    "month_result = []\n",
    "for i in range(1,6+1):\n",
    "    date_begin = datetime.datetime(2019, i, 1)\n",
    "    model = deepcopy(model_not_yet_fitted)\n",
    "    model_fitted = model.fit(X_train,Y_train)\n",
    "    X_test = test[(test.date >= date_begin)].drop(columns=[\"date\",\"total_accidents\"]).values\n",
    "    Y_test = test[(test.date >= date_begin)][\"total_accidents\"].values\n",
    "    preds = infer(model_fitted, X_test)\n",
    "    metric_value = mean_absolute_percentage_error(Y_test, preds)\n",
    "    month_result.append(tuple([i-1,metric_value]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "332fdd49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>gap</th>\n",
       "      <th>mape</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0.240852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0.234113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0.236164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0.223103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0.221332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>0.215008</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   gap      mape\n",
       "0    0  0.240852\n",
       "1    1  0.234113\n",
       "2    2  0.236164\n",
       "3    3  0.223103\n",
       "4    4  0.221332\n",
       "5    5  0.215008"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_degradation = pd.DataFrame.from_records(month_result,columns=[\"gap\",\"mape\"])\n",
    "df_degradation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "dd57d955",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='gap', ylabel='mape'>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=df_degradation,x=\"gap\",y=\"mape\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "349517f4",
   "metadata": {},
   "source": [
    "# 5. Next Steps\n",
    "\n",
    "- Use differentation and log transformation on data and remove the small trend on series. (Can improve the model a little)\n",
    "    - They are simple and can easily return to original values\n",
    "- Create models for specific parts of the day\n",
    "    - It is possible it has more accidents on rush hours.\n",
    "- Test autoregressive models like ARIMA or Prophet \n",
    "- Create models for each region(South, Southeast and etc) as we have neighborhood data available. \n",
    "- Imputation to use other models \n",
    "- Use a feature selection technique to improve model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "62a1eedc",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}