Semeval Brown baseline.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from __future__ import print_function\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import matplotlib.pyplot as plt\n",
    "import re\n",
    "import collections\n",
    "import random\n",
    "from time import time\n",
    "\n",
    "from gensim.models import Word2Vec\n",
    "from sklearn.manifold import TSNE\n",
    "from sklearn.decomposition import PCA, FastICA\n",
    "\n",
    "import data_handler as dh\n",
    "import semeval_data_helper as sdh\n",
    "\n",
    "\n",
    "# plot settings\n",
    "% matplotlib inline\n",
    "# print(plt.rcParams.keys())\n",
    "# plt.rcParams['figure.figsize'] = (16,9)\n",
    "\n",
    "import mpld3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# reload(eh)\n",
    "import experiment_helper as eh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "shuffle_seed = 20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creating Data objects...\n",
      "Done creating Data objects\n",
      "7999 total examples :: 7199 training : 800 valid (90:10 split)\n",
      "Vocab size: 22683 Dep size: 50 POS size: 18\n"
     ]
    }
   ],
   "source": [
    "reload(dh)\n",
    "DH = dh.DataHandler('data/semeval_train_sdp__include_8000', valid_percent=10, shuffle_seed=shuffle_seed) # for semeval"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Didn't find common ancestor\n",
      "1790\t\"The imams were removed from a US Airways <e1>flight</e1> awaiting departure from the Minneapolis-St. Paul <e2>airport</e2>.\"\n",
      "\n",
      "(The imams were removed from a US Airways flight awaiting departure from the Minneapolis - St . Paul airport ., flight , airport )\n",
      "Bad sentence: '1790\\t\"The imams were removed from a US Airways <e1>flight</e1> awaiting departure from the Minneapolis-St. Paul <e2>airport</e2>.\"\\r\\n'\n",
      "((The imams were removed from a US Airways flight awaiting departure from the Minneapolis - St . Paul airport ., flight , airport ), None)\n",
      "Skipping this one... '1790\\t\"The imams were removed from a US Airways <e1>flight</e1> awaiting departure from the Minneapolis-St. Paul <e2>airport</e2>.\"\\r\\n'\n",
      "(None, None, None, 4)\n",
      "Num training: 7199\n",
      "Num valididation: 800\n",
      "Didn't find common ancestor\n",
      "8310\t\"Tributes have been paid to the <e1>writer</e1> who created Goodness Gracious Me, the hit BBC television <e2>series</e2>.\"\n",
      "\n",
      "(Tributes have been paid to the writer who created Goodness Gracious Me , the hit BBC television series ., writer , series )\n",
      "Bad sentence: '8310\\t\"Tributes have been paid to the <e1>writer</e1> who created Goodness Gracious Me, the hit BBC television <e2>series</e2>.\"\\r\\n'\n",
      "((Tributes have been paid to the writer who created Goodness Gracious Me , the hit BBC television series ., writer , series ), None)\n",
      "Skipping this one... '8000\\t\"The <e1>surgeon</e1> cuts a small <e2>hole</e2> in the skull and lifts the edge of the brain to expose the nerve.\"\\r\\n'\n",
      "(None, None, None, 3)\n",
      "Num testing: 2717\n"
     ]
    }
   ],
   "source": [
    "# reload(sdh)\n",
    "train, valid, test, label2int, int2label = sdh.load_semeval_data(include_ends=True, shuffle_seed=shuffle_seed)\n",
    "num_classes = len(int2label.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "### WE DONT WANT INDICES THIS TIME ###\n",
    "# # convert the semeval data to indices under the wiki vocab:\n",
    "# train['sdps'] = DH.sentences_to_sequences(train['sdps'])\n",
    "# valid['sdps'] = DH.sentences_to_sequences(valid['sdps'])\n",
    "# test['sdps'] = DH.sentences_to_sequences(test['sdps'])\n",
    "    \n",
    "# train['targets'] = DH.sentences_to_sequences(train['targets'])\n",
    "# valid['targets'] = DH.sentences_to_sequences(valid['targets'])\n",
    "# test['targets'] = DH.sentences_to_sequences(test['targets'])\n",
    "\n",
    "# print(train['targets'][:5]) # small sample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# max_seq_len = max([len(path) for path in train['sdps']+valid['sdps']+test['sdps']])\n",
    "# print(max_seq_len, DH.max_seq_len)\n",
    "# DH.max_seq_len = max_seq_len"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Simple Logistic Regression on the SDPs w/ TFIDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here\n",
      "Tom plans to continue to live the country life as he grows older . 18\n",
      "Fitting 3 folds for each of 144 candidates, totalling 432 fits\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.8s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done  33 tasks      | elapsed:    8.9s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.6s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.6s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   0.8s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   0.8s\n",
      "[CV] count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 1), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   0.7s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done 154 tasks      | elapsed:   37.2s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   0.9s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   0.9s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.6s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.6s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.7s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.6s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.8s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   0.9s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.0s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.0s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.1s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 2), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=True -   1.5s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.2s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.001, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   2.0s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   2.1s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   2.1s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   2.2s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.9s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   2.0s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   2.0s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.9s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.6s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done 357 tasks      | elapsed:  1.8min\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.01, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.6s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.6s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.6s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l1, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.5s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.4s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l1, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=True, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV] count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False \n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=True -   1.3s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.2s\n",
      "[CV]  count__ngram_range=(1, 3), tfidf__sublinear_tf=False, logreg__penalty=l2, logreg__C=0.1, tfidf__norm=l2, tfidf__use_idf=False -   1.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done 432 out of 432 | elapsed:  2.2min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Best score: 0.328\n",
      "Best parameters set:\n",
      "\tcount__ngram_range: (1, 1)\n",
      "\tlogreg__C: 0.1\n",
      "\tlogreg__penalty: 'l2'\n",
      "\ttfidf__norm: 'l2'\n",
      "\ttfidf__sublinear_tf: True\n",
      "\ttfidf__use_idf: False\n"
     ]
    }
   ],
   "source": [
    "from sklearn.pipeline import Pipeline\n",
    "# define baseline pipelines\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer\n",
    "\n",
    "# Feature Extractors\n",
    "cv = CountVectorizer(\n",
    "        input=u'content', \n",
    "        encoding=u'utf-8', \n",
    "        decode_error=u'strict', \n",
    "        strip_accents='unicode', \n",
    "        lowercase=True,\n",
    "        analyzer=u'word', \n",
    "        preprocessor=None, \n",
    "        tokenizer=None, \n",
    "        stop_words='english', \n",
    "        #token_pattern=u'(?u)\\\\b\\w\\w+\\b', # one alphanumeric is a token\n",
    "        ngram_range=(1, 2), \n",
    "        max_df=.9, \n",
    "        min_df=2, \n",
    "        max_features=None, \n",
    "        vocabulary=None, \n",
    "        binary=False, \n",
    "        #dtype=type 'numpy.int64'>\n",
    "        )\n",
    "from sklearn.feature_extraction.text import TfidfTransformer\n",
    "tf = TfidfTransformer(\n",
    "        norm='l2',\n",
    "        use_idf=True,\n",
    "        smooth_idf=True,\n",
    "        sublinear_tf=False\n",
    ")\n",
    "\n",
    "# Final Classifier\n",
    "lr = LogisticRegression(C=.05,\n",
    "                        fit_intercept=True,\n",
    "                        random_state=0,\n",
    "                        class_weight='balanced',\n",
    "#                         multi_class='multinomial',\n",
    "                        #solver='lbfgs',\n",
    "                        n_jobs=1)\n",
    "\n",
    "pipeline = Pipeline([\n",
    "    ('count', cv),\n",
    "    ('tfidf', tf),\n",
    "    ('logreg', lr)\n",
    "    ])\n",
    "\n",
    "param_grid = {\n",
    "    'count__ngram_range':[(1,1),(1,2),(1,3)],\n",
    "    'tfidf__norm':['l1', 'l2'],\n",
    "    'tfidf__use_idf':[True, False],\n",
    "    'tfidf__sublinear_tf':[True,False],\n",
    "    'logreg__C':[.001, .01, .1],\n",
    "    'logreg__penalty':['l1', 'l2']\n",
    "}\n",
    "\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "grid_search = GridSearchCV(pipeline, \n",
    "                           param_grid,\n",
    "                           scoring='f1_macro',\n",
    "                           n_jobs=-1, verbose=1)\n",
    "\n",
    "print(\"Here\")\n",
    "x_data = [sent[0].text for sent in train['sents']]\n",
    "y_data = train['labels']\n",
    "print(x_data[0], y_data[0])\n",
    "grid_search.fit(np.array(x_data), y_data)\n",
    "print()\n",
    "\n",
    "print(\"Best score: %0.3f\" % grid_search.best_score_)\n",
    "print(\"Best parameters set:\")\n",
    "best_parameters = grid_search.best_estimator_.get_params()\n",
    "for param_name in sorted(param_grid.keys()):\n",
    "    print(\"\\t%s: %r\" % (param_name, best_parameters[param_name]))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.metrics import f1_score\n",
    "test_x = [sent[0].text for sent in valid['sents']]\n",
    "test_y = valid['labels']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR Sentence validation f1 macro: 0.3291\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1076: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no true samples.\n",
      "  'recall', 'true', average, warn_for)\n"
     ]
    }
   ],
   "source": [
    "lr_sent = grid_search.best_estimator_\n",
    "lr_sent.fit(x_data, y_data)\n",
    "preds = lr_sent.predict(test_x)\n",
    "lr_sent_score = f1_score(test_y, preds, average='macro')\n",
    "print('LR Sentence validation f1 macro: %2.4f' % lr_sent_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression on the words in SDPs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here\n",
      "country life 18\n",
      "Fitting 3 folds for each of 144 candidates, totalling 432 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "[Parallel(n_jobs=-1)]: Done  42 tasks      | elapsed:    4.0s\n",
      "[Parallel(n_jobs=-1)]: Done 192 tasks      | elapsed:   16.8s\n",
      "[Parallel(n_jobs=-1)]: Done 432 out of 432 | elapsed:   40.6s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Best score: 0.431\n",
      "Best parameters set:\n",
      "\tcount__ngram_range: (1, 3)\n",
      "\tlogreg__C: 0.1\n",
      "\tlogreg__penalty: 'l2'\n",
      "\ttfidf__norm: 'l2'\n",
      "\ttfidf__sublinear_tf: False\n",
      "\ttfidf__use_idf: True\n"
     ]
    }
   ],
   "source": [
    "from sklearn.pipeline import Pipeline\n",
    "# define baseline pipelines\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer\n",
    "\n",
    "# Feature Extractors\n",
    "cv = CountVectorizer(\n",
    "        input=u'content', \n",
    "        encoding=u'utf-8', \n",
    "        decode_error=u'strict', \n",
    "        strip_accents='unicode', \n",
    "        lowercase=True,\n",
    "        analyzer=u'word', \n",
    "        preprocessor=None, \n",
    "        tokenizer=None, \n",
    "        stop_words='english', \n",
    "        #token_pattern=u'(?u)\\\\b\\w\\w+\\b', # one alphanumeric is a token\n",
    "        ngram_range=(1, 2), \n",
    "        max_df=.9, \n",
    "        min_df=2, \n",
    "        max_features=None, \n",
    "        vocabulary=None, \n",
    "        binary=False, \n",
    "        #dtype=type 'numpy.int64'>\n",
    "        )\n",
    "from sklearn.feature_extraction.text import TfidfTransformer\n",
    "tf = TfidfTransformer(\n",
    "        norm='l2',\n",
    "        use_idf=True,\n",
    "        smooth_idf=True,\n",
    "        sublinear_tf=False\n",
    ")\n",
    "\n",
    "# Final Classifier\n",
    "lr = LogisticRegression(C=.05,\n",
    "                        fit_intercept=True,\n",
    "                        random_state=0,\n",
    "                        class_weight='balanced',\n",
    "#                         multi_class='multinomial',\n",
    "                        #solver='lbfgs',\n",
    "                        n_jobs=1)\n",
    "\n",
    "pipeline = Pipeline([\n",
    "    ('count', cv),\n",
    "    ('tfidf', tf),\n",
    "    ('logreg', lr)\n",
    "    ])\n",
    "\n",
    "param_grid = {\n",
    "    'count__ngram_range':[(1,1),(1,2),(1,3)],\n",
    "    'tfidf__norm':['l1', 'l2'],\n",
    "    'tfidf__use_idf':[True, False],\n",
    "    'tfidf__sublinear_tf':[True,False],\n",
    "    'logreg__C':[.001, .01, .1],\n",
    "    'logreg__penalty':['l1', 'l2']\n",
    "}\n",
    "\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "grid_search = GridSearchCV(pipeline, \n",
    "                           param_grid,\n",
    "                           scoring='f1_macro',\n",
    "                           n_jobs=-1, verbose=1)\n",
    "\n",
    "print(\"Here\")\n",
    "x_data = [\" \".join([tok[0] for tok in sdp]) for sdp in train['sdps']]\n",
    "y_data = train['labels']\n",
    "print(x_data[0], y_data[0])\n",
    "grid_search.fit(np.array(x_data), y_data)\n",
    "print()\n",
    "\n",
    "print(\"Best score: %0.3f\" % grid_search.best_score_)\n",
    "print(\"Best parameters set:\")\n",
    "best_parameters = grid_search.best_estimator_.get_params()\n",
    "for param_name in sorted(param_grid.keys()):\n",
    "    print(\"\\t%s: %r\" % (param_name, best_parameters[param_name]))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR SDP validation f1 macro: 0.3663\n"
     ]
    }
   ],
   "source": [
    "lr_sdp = grid_search.best_estimator_\n",
    "lr_sdp.fit(x_data, y_data)\n",
    "preds = lr_sdp.predict(test_x)\n",
    "lr_sdp_score = f1_score(test_y, preds, average='macro')\n",
    "print('LR SDP validation f1 macro: %2.4f' % lr_sdp_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Logistic Regression on SDP with Brown clusters bag"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "brown_clusters = {}\n",
    "lines = open('data/shuffled.en.tok-c50-p1.out/paths', 'r').readlines()\n",
    "for line in lines:\n",
    "    vec = line.split()\n",
    "    brown_clusters[vec[1]] = vec[0]\n",
    "del lines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[u'country life 011011 011011', u'car has player casette 011011 01111111110 011011 000110', u'jam of women 011011 00100 011011', u'view is from bridge 011011 0111111100 0010111 011011', u'hill of corn 011011 00100 0111011']\n"
     ]
    }
   ],
   "source": [
    "# append brown clusters to each sdp sentence\n",
    "x_data = [\" \".join([tok[0] for tok in sdp]) + \" \" + \n",
    "          \" \".join([brown_clusters[str(tok[0])] for tok in sdp if str(tok[0]) in brown_clusters])\n",
    "          for sdp in train['sdps']]\n",
    "print(x_data[:5])\n",
    "test_x = [\" \".join([tok[0] for tok in sdp]) + \" \" + \n",
    "          \" \".join([brown_clusters[str(tok[0])] for tok in sdp if str(tok[0]) in brown_clusters])\n",
    "          for sdp in valid['sdps']]\n",
    "x_data_ends = [\" \".join([tok[0] for tok in sdp]) + \" \" + \n",
    "          \" \".join([brown_clusters[str(tok[0])] for tok in [sdp[0], sdp[-1]] if str(tok[0]) in brown_clusters])\n",
    "          for sdp in train['sdps']]\n",
    "test_x_ends = [\" \".join([tok[0] for tok in sdp]) + \" \" + \n",
    "          \" \".join([brown_clusters[str(tok[0])] for tok in [sdp[0], sdp[-1]] if str(tok[0]) in brown_clusters])\n",
    "          for sdp in valid['sdps']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "country life 011011 011011 18\n",
      "Fitting 3 folds for each of 240 candidates, totalling 720 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "/Users/thomaseffland/.virtualenvs/rel/lib/python2.7/site-packages/sklearn/metrics/classification.py:1074: UndefinedMetricWarning: F-score is ill-defined and being set to 0.0 in labels with no predicted samples.\n",
      "  'precision', 'predicted', average, warn_for)\n",
      "[Parallel(n_jobs=-1)]: Done  42 tasks      | elapsed:    8.4s\n",
      "[Parallel(n_jobs=-1)]: Done 192 tasks      | elapsed:   48.9s\n",
      "[Parallel(n_jobs=-1)]: Done 442 tasks      | elapsed:  2.3min\n",
      "[Parallel(n_jobs=-1)]: Done 720 out of 720 | elapsed:  4.2min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Best score: 0.550\n",
      "Best parameters set:\n",
      "\tcount__ngram_range: (1, 1)\n",
      "\tlogreg__C: 1.0\n",
      "\tlogreg__penalty: 'l2'\n",
      "\ttfidf__norm: 'l2'\n",
      "\ttfidf__sublinear_tf: False\n",
      "\ttfidf__use_idf: True\n"
     ]
    }
   ],
   "source": [
    "from sklearn.pipeline import Pipeline\n",
    "# define baseline pipelines\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer\n",
    "\n",
    "# Feature Extractors\n",
    "cv = CountVectorizer(\n",
    "        input=u'content', \n",
    "        encoding=u'utf-8', \n",
    "        decode_error=u'strict', \n",
    "        strip_accents='unicode', \n",
    "        lowercase=True,\n",
    "        analyzer=u'word', \n",
    "        preprocessor=None, \n",
    "        tokenizer=None, \n",
    "        stop_words='english', \n",
    "        #token_pattern=u'(?u)\\\\b\\w\\w+\\b', # one alphanumeric is a token\n",
    "        ngram_range=(1, 2), \n",
    "        max_df=.9, \n",
    "        min_df=2, \n",
    "        max_features=None, \n",
    "        vocabulary=None, \n",
    "        binary=False, \n",
    "        #dtype=type 'numpy.int64'>\n",
    "        )\n",
    "from sklearn.feature_extraction.text import TfidfTransformer\n",
    "tf = TfidfTransformer(\n",
    "        norm='l2',\n",
    "        use_idf=True,\n",
    "        smooth_idf=True,\n",
    "        sublinear_tf=False\n",
    ")\n",
    "\n",
    "# Final Classifier\n",
    "lr = LogisticRegression(C=.05,\n",
    "                        fit_intercept=True,\n",
    "                        random_state=0,\n",
    "                        class_weight='balanced',\n",
    "#                         multi_class='multinomial',\n",
    "                        #solver='lbfgs',\n",
    "                        n_jobs=1)\n",
    "\n",
    "pipeline = Pipeline([\n",
    "    ('count', cv),\n",
    "    ('tfidf', tf),\n",
    "    ('logreg', lr)\n",
    "    ])\n",
    "\n",
    "param_grid = {\n",
    "    'count__ngram_range':[(1,1),(1,2),(1,3)],\n",
    "    'tfidf__norm':['l1', 'l2'],\n",
    "    'tfidf__use_idf':[True, False],\n",
    "    'tfidf__sublinear_tf':[True,False],\n",
    "    'logreg__C':[.001, .01, .1, .5, 1.],\n",
    "    'logreg__penalty':['l1', 'l2']\n",
    "}\n",
    "\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "grid_search = GridSearchCV(pipeline, \n",
    "                           param_grid,\n",
    "                           scoring='f1_macro',\n",
    "                           n_jobs=-1, verbose=1)\n",
    "\n",
    "print(x_data[0], y_data[0])\n",
    "grid_search.fit(np.array(x_data), y_data)\n",
    "print()\n",
    "\n",
    "print(\"Best score: %0.3f\" % grid_search.best_score_)\n",
    "print(\"Best parameters set:\")\n",
    "best_parameters = grid_search.best_estimator_.get_params()\n",
    "for param_name in sorted(param_grid.keys()):\n",
    "    print(\"\\t%s: %r\" % (param_name, best_parameters[param_name]))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LR SDP validation f1 macro: 0.5969\n"
     ]
    }
   ],
   "source": [
    "lr_sdp_brown = grid_search.best_estimator_\n",
    "lr_sdp_brown.fit(x_data, y_data)\n",
    "preds = lr_sdp_brown.predict(test_x)\n",
    "lr_sdp_brown_score = f1_score(test_y, preds, average='macro')\n",
    "print('LR SDP validation f1 macro: %2.4f' % lr_sdp_brown_score)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "50 Brown clusters: LR SDP validation f1 macro: 0.5970\n",
    "100 Brown on all wiki: LR SDP validation f1 macro: 0.6097\n",
    "100 Brown on all wiki w/ ends only: LR SDP validation f1 macro: 0.5120"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "test_x = [\" \".join([tok[0] for tok in sdp]) + \" \" + \n",
    "          \" \".join([brown_clusters[str(tok[0])] for tok in sdp if str(tok[0]) in brown_clusters])\n",
    "          for sdp in test['sdps']]\n",
    "\n",
    "preds = lr_sdp_brown.predict(test_x)\n",
    "with open('SemEval2010_task8_all_data/test_pred.txt', 'w') as f:\n",
    "    i = 8001\n",
    "    for pred in preds:\n",
    "        f.write(\"%i\\t%s\\n\" % (i, int2label[pred]))\n",
    "        i += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<<< (2*9+1)-WAY EVALUATION (USING DIRECTIONALITY)>>>:\n",
      "\n",
      "Confusion matrix:\n",
      "        C-E1 C-E2 C-W1 C-W2 C-C1 C-C2 E-D1 E-D2 E-O1 E-O2 I-A1 I-A2 M-C1 M-C2 M-T1 M-T2 P-P1 P-P2  _O_ <-- classified as\n",
      "      +-----------------------------------------------------------------------------------------------+ -SUM- skip ACTUAL\n",
      " C-E1 |  72   39    2    0    0    0    2    0    0    1    0    1    0   11    1    1    2    0    2 |  134    0  134\n",
      " C-E2 |   5  153    0    1    0    0    0    0    4    1    0    1    0    6    3    0    3    1   16 |  194    0  194\n",
      " C-W1 |   1    0   71   13    5    2    4    0    4    0    3    1    3   20    6    0    5    1   23 |  162    0  162\n",
      " C-W2 |   0    3   20   54    1    0    6    0    2    2    1   10    1   28    3    0    3    4   12 |  150    0  150\n",
      " C-C1 |   0    1    4    5   89   20   14    0    2    4    0    0    2    2    0    0    0    0   10 |  153    0  153\n",
      " C-C2 |   0    0    1    0    3   23    1    0    0    1    0    1    1    2    1    0    0    0    5 |   39    0   39\n",
      " E-D1 |   0   18    5    3   12    2  186    0   13    0    0    1    2    8    3    2    3    1   32 |  291    0  291\n",
      " E-D2 |   0    0    0    0    0    0    1    0    0    0    0    0    0    0    0    0    0    0    0 |    1    0    1\n",
      " E-O1 |   0    4    5    1    2    1   16    0  109    3    0    0    0   15    4    7   13    0   31 |  211    0  211\n",
      " E-O2 |   0    1    0    1    1    0    0    0    1   40    0    0    0    3    0    0    0    0    0 |   47    0   47\n",
      " I-A1 |   0    2    2    0    0    0    0    0    0    0   12    0    0    1    0    0    1    0    4 |   22    0   22\n",
      " I-A2 |   0    4    6   16    0    1    3    0    2    1    2   49    0   14    1    0    3    4   28 |  134    0  134\n",
      " M-C1 |   0    0    1    0    0    0    0    0    0    0    0    0   19    2    0    1    0    1    8 |   32    0   32\n",
      " M-C2 |   0    1    1    1    0    2    2    0    1    1    0    2    3  161    4    0    2    4   16 |  201    0  201\n",
      " M-T1 |   0    9    4    2    2    0    1    0    2    0    1    2    2   28  108   22    1    0   26 |  210    0  210\n",
      " M-T2 |   0    1    1    0    0    0    2    0    1    0    0    0    0    3   12   27    1    1    2 |   51    0   51\n",
      " P-P1 |   1    6    3    0    0    0    1    0    0    1    3    0    0   12    4    1   59    9    8 |  108    0  108\n",
      " P-P2 |   0    1    4    1    1    1    7    0    4    0    2    3    0   21    3    0   15   38   22 |  123    0  123\n",
      "  _O_ |   6   17   15   14   21    4   33    0   22   10    5   11   11   82   31   10   18   18  126 |  454    0  454\n",
      "      +-----------------------------------------------------------------------------------------------+\n",
      " -SUM-   85  260  145  112  137   56  279    0  167   65   29   82   44  419  184   71  129   82  371   2717    0 2717\n",
      "\n",
      "Coverage = 2717/2717 = 100.00%\n",
      "Accuracy (calculated for the above confusion matrix) = 1396/2717 = 51.38%\n",
      "Accuracy (considering all skipped examples as Wrong) = 1396/2717 = 51.38%\n",
      "Accuracy (considering all skipped examples as Other) = 1396/2717 = 51.38%\n",
      "\n",
      "Results for the individual relations:\n",
      "      Cause-Effect(e1,e2) :    P =   72/  85 =  84.71%     R =   72/ 134 =  53.73%     F1 =  65.75%\n",
      "      Cause-Effect(e2,e1) :    P =  153/ 260 =  58.85%     R =  153/ 194 =  78.87%     F1 =  67.40%\n",
      "   Component-Whole(e1,e2) :    P =   71/ 145 =  48.97%     R =   71/ 162 =  43.83%     F1 =  46.25%\n",
      "   Component-Whole(e2,e1) :    P =   54/ 112 =  48.21%     R =   54/ 150 =  36.00%     F1 =  41.22%\n",
      " Content-Container(e1,e2) :    P =   89/ 137 =  64.96%     R =   89/ 153 =  58.17%     F1 =  61.38%\n",
      " Content-Container(e2,e1) :    P =   23/  56 =  41.07%     R =   23/  39 =  58.97%     F1 =  48.42%\n",
      "Entity-Destination(e1,e2) :    P =  186/ 279 =  66.67%     R =  186/ 291 =  63.92%     F1 =  65.26%\n",
      "Entity-Destination(e2,e1) :    P =    0/   0 =   0.00%     R =    0/   1 =   0.00%     F1 =   0.00%\n",
      "     Entity-Origin(e1,e2) :    P =  109/ 167 =  65.27%     R =  109/ 211 =  51.66%     F1 =  57.67%\n",
      "     Entity-Origin(e2,e1) :    P =   40/  65 =  61.54%     R =   40/  47 =  85.11%     F1 =  71.43%\n",
      " Instrument-Agency(e1,e2) :    P =   12/  29 =  41.38%     R =   12/  22 =  54.55%     F1 =  47.06%\n",
      " Instrument-Agency(e2,e1) :    P =   49/  82 =  59.76%     R =   49/ 134 =  36.57%     F1 =  45.37%\n",
      " Member-Collection(e1,e2) :    P =   19/  44 =  43.18%     R =   19/  32 =  59.38%     F1 =  50.00%\n",
      " Member-Collection(e2,e1) :    P =  161/ 419 =  38.42%     R =  161/ 201 =  80.10%     F1 =  51.94%\n",
      "     Message-Topic(e1,e2) :    P =  108/ 184 =  58.70%     R =  108/ 210 =  51.43%     F1 =  54.82%\n",
      "     Message-Topic(e2,e1) :    P =   27/  71 =  38.03%     R =   27/  51 =  52.94%     F1 =  44.26%\n",
      "  Product-Producer(e1,e2) :    P =   59/ 129 =  45.74%     R =   59/ 108 =  54.63%     F1 =  49.79%\n",
      "  Product-Producer(e2,e1) :    P =   38/  82 =  46.34%     R =   38/ 123 =  30.89%     F1 =  37.07%\n",
      "                   _Other :    P =  126/ 371 =  33.96%     R =  126/ 454 =  27.75%     F1 =  30.55%\n",
      "\n",
      "Micro-averaged result (excluding Other):\n",
      "P = 1270/2346 =  54.13%     R = 1270/2263 =  56.12%     F1 =  55.11%\n",
      "\n",
      "MACRO-averaged result (excluding Other):\n",
      "P =  50.65%\tR =  52.82%\tF1 =  50.28%\n",
      "\n",
      "\n",
      "\n",
      "<<< (9+1)-WAY EVALUATION IGNORING DIRECTIONALITY >>>:\n",
      "\n",
      "Confusion matrix:\n",
      "         C-E  C-W  C-C  E-D  E-O  I-A  M-C  M-T  P-P  _O_ <-- classified as\n",
      "      +--------------------------------------------------+ -SUM- skip ACTUAL\n",
      "  C-E | 269    3    0    2    6    2   17    5    6   18 |  328    0  328\n",
      "  C-W |   4  158    8   10    8   15   52    9   13   35 |  312    0  312\n",
      "  C-C |   1   10  135   15    7    1    7    1    0   15 |  192    0  192\n",
      "  E-D |  18    8   14  187   13    1   10    5    4   32 |  292    0  292\n",
      "  E-O |   5    7    4   16  153    0   18   11   13   31 |  258    0  258\n",
      "  I-A |   6   24    1    3    3   63   15    1    8   32 |  156    0  156\n",
      "  M-C |   1    3    2    2    2    2  185    5    7   24 |  233    0  233\n",
      "  M-T |  10    7    2    3    3    3   33  169    3   28 |  261    0  261\n",
      "  P-P |   8    8    2    8    5    8   33    8  121   30 |  231    0  231\n",
      "  _O_ |  23   29   25   33   32   16   93   41   36  126 |  454    0  454\n",
      "      +--------------------------------------------------+\n",
      " -SUM-  345  257  193  279  232  111  463  255  211  371   2717    0 2717\n",
      "\n",
      "Coverage = 2717/2717 = 100.00%\n",
      "Accuracy (calculated for the above confusion matrix) = 1566/2717 = 57.64%\n",
      "Accuracy (considering all skipped examples as Wrong) = 1566/2717 = 57.64%\n",
      "Accuracy (considering all skipped examples as Other) = 1566/2717 = 57.64%\n",
      "\n",
      "Results for the individual relations:\n",
      "             Cause-Effect :    P =  269/ 345 =  77.97%     R =  269/ 328 =  82.01%     F1 =  79.94%\n",
      "          Component-Whole :    P =  158/ 257 =  61.48%     R =  158/ 312 =  50.64%     F1 =  55.54%\n",
      "        Content-Container :    P =  135/ 193 =  69.95%     R =  135/ 192 =  70.31%     F1 =  70.13%\n",
      "       Entity-Destination :    P =  187/ 279 =  67.03%     R =  187/ 292 =  64.04%     F1 =  65.50%\n",
      "            Entity-Origin :    P =  153/ 232 =  65.95%     R =  153/ 258 =  59.30%     F1 =  62.45%\n",
      "        Instrument-Agency :    P =   63/ 111 =  56.76%     R =   63/ 156 =  40.38%     F1 =  47.19%\n",
      "        Member-Collection :    P =  185/ 463 =  39.96%     R =  185/ 233 =  79.40%     F1 =  53.16%\n",
      "            Message-Topic :    P =  169/ 255 =  66.27%     R =  169/ 261 =  64.75%     F1 =  65.50%\n",
      "         Product-Producer :    P =  121/ 211 =  57.35%     R =  121/ 231 =  52.38%     F1 =  54.75%\n",
      "                   _Other :    P =  126/ 371 =  33.96%     R =  126/ 454 =  27.75%     F1 =  30.55%\n",
      "\n",
      "Micro-averaged result (excluding Other):\n",
      "P = 1440/2346 =  61.38%     R = 1440/2263 =  63.63%     F1 =  62.49%\n",
      "\n",
      "MACRO-averaged result (excluding Other):\n",
      "P =  62.52%\tR =  62.58%\tF1 =  61.57%\n",
      "\n",
      "\n",
      "\n",
      "<<< (9+1)-WAY EVALUATION TAKING DIRECTIONALITY INTO ACCOUNT -- OFFICIAL >>>:\n",
      "\n",
      "Confusion matrix:\n",
      "         C-E  C-W  C-C  E-D  E-O  I-A  M-C  M-T  P-P  _O_ <-- classified as\n",
      "      +--------------------------------------------------+ -SUM- xDIRx skip  ACTUAL\n",
      "  C-E | 225    3    0    2    6    2   17    5    6   18 |  284    44     0    328\n",
      "  C-W |   4  125    8   10    8   15   52    9   13   35 |  279    33     0    312\n",
      "  C-C |   1   10  112   15    7    1    7    1    0   15 |  169    23     0    192\n",
      "  E-D |  18    8   14  186   13    1   10    5    4   32 |  291     1     0    292\n",
      "  E-O |   5    7    4   16  149    0   18   11   13   31 |  254     4     0    258\n",
      "  I-A |   6   24    1    3    3   61   15    1    8   32 |  154     2     0    156\n",
      "  M-C |   1    3    2    2    2    2  180    5    7   24 |  228     5     0    233\n",
      "  M-T |  10    7    2    3    3    3   33  135    3   28 |  227    34     0    261\n",
      "  P-P |   8    8    2    8    5    8   33    8   97   30 |  207    24     0    231\n",
      "  _O_ |  23   29   25   33   32   16   93   41   36  126 |  454     0     0    454\n",
      "      +--------------------------------------------------+\n",
      " -SUM-  301  224  170  278  228  109  458  221  187  371   2547   170     0   2717\n",
      "\n",
      "Coverage = 2717/2717 = 100.00%\n",
      "Accuracy (calculated for the above confusion matrix) = 1396/2717 = 51.38%\n",
      "Accuracy (considering all skipped examples as Wrong) = 1396/2717 = 51.38%\n",
      "Accuracy (considering all skipped examples as Other) = 1396/2717 = 51.38%\n",
      "\n",
      "Results for the individual relations:\n",
      "             Cause-Effect :    P =  225/( 301 +  44) =  65.22%     R =  225/ 328 =  68.60%     F1 =  66.86%\n",
      "          Component-Whole :    P =  125/( 224 +  33) =  48.64%     R =  125/ 312 =  40.06%     F1 =  43.94%\n",
      "        Content-Container :    P =  112/( 170 +  23) =  58.03%     R =  112/ 192 =  58.33%     F1 =  58.18%\n",
      "       Entity-Destination :    P =  186/( 278 +   1) =  66.67%     R =  186/ 292 =  63.70%     F1 =  65.15%\n",
      "            Entity-Origin :    P =  149/( 228 +   4) =  64.22%     R =  149/ 258 =  57.75%     F1 =  60.82%\n",
      "        Instrument-Agency :    P =   61/( 109 +   2) =  54.95%     R =   61/ 156 =  39.10%     F1 =  45.69%\n",
      "        Member-Collection :    P =  180/( 458 +   5) =  38.88%     R =  180/ 233 =  77.25%     F1 =  51.72%\n",
      "            Message-Topic :    P =  135/( 221 +  34) =  52.94%     R =  135/ 261 =  51.72%     F1 =  52.33%\n",
      "         Product-Producer :    P =   97/( 187 +  24) =  45.97%     R =   97/ 231 =  41.99%     F1 =  43.89%\n",
      "                   _Other :    P =  126/( 371 +   0) =  33.96%     R =  126/ 454 =  27.75%     F1 =  30.55%\n",
      "\n",
      "Micro-averaged result (excluding Other):\n",
      "P = 1270/2346 =  54.13%     R = 1270/2263 =  56.12%     F1 =  55.11%\n",
      "\n",
      "MACRO-averaged result (excluding Other):\n",
      "P =  55.06%\tR =  55.39%\tF1 =  54.29%\n",
      "\n",
      "\n",
      "\n",
      "<<< The official score is (9+1)-way evaluation with directionality taken into account: macro-averaged F1 = 54.29% >>>\n"
     ]
    }
   ],
   "source": [
    "%%bash\n",
    "./SemEval2010_task8_all_data/SemEval2010_task8_scorer-v1.2/semeval2010_task8_scorer-v1.2.pl \\\n",
    "SemEval2010_task8_all_data/test_pred.txt SemEval2010_task8_all_data/test_keys.txt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}