Connect-4 Classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Connect-4 Classification\n",
    "_____\n",
    "\n",
    "Dataset : https://archive.ics.uci.edu/ml/datasets/Connect-4 \n",
    "\n",
    "This database contains all legal 8-ply positions in the game of connect-4 in which neither player has won yet, and in which the next move is not forced. \n",
    "\n",
    "x is the first player; o the second. \n",
    "\n",
    "The outcome class is the game theoretical value for the first player.\n",
    "\n",
    "Attribute Information: (x=player x has taken, o=player o has taken, b=blank) \n",
    "\n",
    "The board is numbered like: \n",
    "<pre>\n",
    "6 . . . . . . . \n",
    "5 . . . . . . . \n",
    "4 . . . . . . . \n",
    "3 . . . . . . . \n",
    "2 . . . . . . . \n",
    "1 . . . . . . . \n",
    "a b c d e f g \n",
    "\n",
    "\n",
    "1. a1: {x,o,b} \n",
    "2. a2: {x,o,b} \n",
    "3. a3: {x,o,b} \n",
    "4. a4: {x,o,b} \n",
    "5. a5: {x,o,b} \n",
    "6. a6: {x,o,b} \n",
    "7. b1: {x,o,b} \n",
    "8. b2: {x,o,b} \n",
    "9. b3: {x,o,b} \n",
    "10. b4: {x,o,b} \n",
    "11. b5: {x,o,b} \n",
    "12. b6: {x,o,b} \n",
    "13. c1: {x,o,b} \n",
    "14. c2: {x,o,b} \n",
    "15. c3: {x,o,b} \n",
    "16. c4: {x,o,b} \n",
    "17. c5: {x,o,b} \n",
    "18. c6: {x,o,b} \n",
    "19. d1: {x,o,b} \n",
    "20. d2: {x,o,b} \n",
    "21. d3: {x,o,b} \n",
    "22. d4: {x,o,b} \n",
    "23. d5: {x,o,b} \n",
    "24. d6: {x,o,b} \n",
    "25. e1: {x,o,b} \n",
    "26. e2: {x,o,b} \n",
    "27. e3: {x,o,b} \n",
    "28. e4: {x,o,b} \n",
    "29. e5: {x,o,b} \n",
    "30. e6: {x,o,b} \n",
    "31. f1: {x,o,b} \n",
    "32. f2: {x,o,b} \n",
    "33. f3: {x,o,b} \n",
    "34. f4: {x,o,b} \n",
    "35. f5: {x,o,b} \n",
    "36. f6: {x,o,b} \n",
    "37. g1: {x,o,b} \n",
    "38. g2: {x,o,b} \n",
    "39. g3: {x,o,b} \n",
    "40. g4: {x,o,b} \n",
    "41. g5: {x,o,b} \n",
    "42. g6: {x,o,b} \n",
    "43. Class: {win,loss,draw}\n",
    "</pre>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import os\n",
    "from sklearn.tree import DecisionTreeClassifier, export_graphviz\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn import cross_validation, metrics\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.naive_bayes import BernoulliNB\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.svm import SVC\n",
    "from time import time\n",
    "from sklearn import preprocessing\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import roc_auc_score , classification_report\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import precision_score, recall_score, accuracy_score, classification_report\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "index=[]\n",
    "board = ['a','b','c','d','e','f','g']\n",
    "for i in board:\n",
    "    for j in range(6):\n",
    "        index.append(i + str(j+1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "column_names  = index +['Class']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# read .csv from provided dataset\n",
    "csv_filename=\"connect-4.data\"\n",
    "\n",
    "# df=pd.read_csv(csv_filename,index_col=0)\n",
    "df=pd.read_csv(csv_filename, \n",
    "              names= column_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>a1</th>\n",
       "      <th>a2</th>\n",
       "      <th>a3</th>\n",
       "      <th>a4</th>\n",
       "      <th>a5</th>\n",
       "      <th>a6</th>\n",
       "      <th>b1</th>\n",
       "      <th>b2</th>\n",
       "      <th>b3</th>\n",
       "      <th>b4</th>\n",
       "      <th>...</th>\n",
       "      <th>f4</th>\n",
       "      <th>f5</th>\n",
       "      <th>f6</th>\n",
       "      <th>g1</th>\n",
       "      <th>g2</th>\n",
       "      <th>g3</th>\n",
       "      <th>g4</th>\n",
       "      <th>g5</th>\n",
       "      <th>g6</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>...</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>win</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>...</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>win</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>o</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>...</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>win</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>...</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>win</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>o</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>...</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>b</td>\n",
       "      <td>win</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 43 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  a1 a2 a3 a4 a5 a6 b1 b2 b3 b4  ...  f4 f5 f6 g1 g2 g3 g4 g5 g6 Class\n",
       "0  b  b  b  b  b  b  b  b  b  b  ...   b  b  b  b  b  b  b  b  b   win\n",
       "1  b  b  b  b  b  b  b  b  b  b  ...   b  b  b  b  b  b  b  b  b   win\n",
       "2  b  b  b  b  b  b  o  b  b  b  ...   b  b  b  b  b  b  b  b  b   win\n",
       "3  b  b  b  b  b  b  b  b  b  b  ...   b  b  b  b  b  b  b  b  b   win\n",
       "4  o  b  b  b  b  b  b  b  b  b  ...   b  b  b  b  b  b  b  b  b   win\n",
       "\n",
       "[5 rows x 43 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['win', 'draw', 'loss'], dtype=object)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Class'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#Convert animal labels to numbers\n",
    "le = preprocessing.LabelEncoder()\n",
    "for col in df.columns:\n",
    "    df[col] = le.fit_transform(df[col])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>a1</th>\n",
       "      <th>a2</th>\n",
       "      <th>a3</th>\n",
       "      <th>a4</th>\n",
       "      <th>a5</th>\n",
       "      <th>a6</th>\n",
       "      <th>b1</th>\n",
       "      <th>b2</th>\n",
       "      <th>b3</th>\n",
       "      <th>b4</th>\n",
       "      <th>...</th>\n",
       "      <th>f4</th>\n",
       "      <th>f5</th>\n",
       "      <th>f6</th>\n",
       "      <th>g1</th>\n",
       "      <th>g2</th>\n",
       "      <th>g3</th>\n",
       "      <th>g4</th>\n",
       "      <th>g5</th>\n",
       "      <th>g6</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 43 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   a1  a2  a3  a4  a5  a6  b1  b2  b3  b4  ...    f4  f5  f6  g1  g2  g3  g4  \\\n",
       "0   0   0   0   0   0   0   0   0   0   0  ...     0   0   0   0   0   0   0   \n",
       "1   0   0   0   0   0   0   0   0   0   0  ...     0   0   0   0   0   0   0   \n",
       "2   0   0   0   0   0   0   1   0   0   0  ...     0   0   0   0   0   0   0   \n",
       "3   0   0   0   0   0   0   0   0   0   0  ...     0   0   0   0   0   0   0   \n",
       "4   1   0   0   0   0   0   0   0   0   0  ...     0   0   0   0   0   0   0   \n",
       "\n",
       "   g5  g6  Class  \n",
       "0   0   0      2  \n",
       "1   0   0      2  \n",
       "2   0   0      2  \n",
       "3   0   0      2  \n",
       "4   0   0      2  \n",
       "\n",
       "[5 rows x 43 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for col in df.columns:\n",
    "    df[col] = pd.get_dummies(df[col])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>a1</th>\n",
       "      <th>a2</th>\n",
       "      <th>a3</th>\n",
       "      <th>a4</th>\n",
       "      <th>a5</th>\n",
       "      <th>a6</th>\n",
       "      <th>b1</th>\n",
       "      <th>b2</th>\n",
       "      <th>b3</th>\n",
       "      <th>b4</th>\n",
       "      <th>...</th>\n",
       "      <th>f4</th>\n",
       "      <th>f5</th>\n",
       "      <th>f6</th>\n",
       "      <th>g1</th>\n",
       "      <th>g2</th>\n",
       "      <th>g3</th>\n",
       "      <th>g4</th>\n",
       "      <th>g5</th>\n",
       "      <th>g6</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 43 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   a1  a2  a3  a4  a5  a6  b1  b2  b3  b4  ...    f4  f5  f6  g1  g2  g3  g4  \\\n",
       "0   1   1   1   1   1   1   1   1   1   1  ...     1   1   1   1   1   1   1   \n",
       "1   1   1   1   1   1   1   1   1   1   1  ...     1   1   1   1   1   1   1   \n",
       "2   1   1   1   1   1   1   0   1   1   1  ...     1   1   1   1   1   1   1   \n",
       "3   1   1   1   1   1   1   1   1   1   1  ...     1   1   1   1   1   1   1   \n",
       "4   0   1   1   1   1   1   1   1   1   1  ...     1   1   1   1   1   1   1   \n",
       "\n",
       "   g5  g6  Class  \n",
       "0   1   1      0  \n",
       "1   1   1      0  \n",
       "2   1   1      0  \n",
       "3   1   1      0  \n",
       "4   1   1      0  \n",
       "\n",
       "[5 rows x 43 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "X = df[index]\n",
    "y = df['Class']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>a1</th>\n",
       "      <th>a2</th>\n",
       "      <th>a3</th>\n",
       "      <th>a4</th>\n",
       "      <th>a5</th>\n",
       "      <th>a6</th>\n",
       "      <th>b1</th>\n",
       "      <th>b2</th>\n",
       "      <th>b3</th>\n",
       "      <th>b4</th>\n",
       "      <th>...</th>\n",
       "      <th>f3</th>\n",
       "      <th>f4</th>\n",
       "      <th>f5</th>\n",
       "      <th>f6</th>\n",
       "      <th>g1</th>\n",
       "      <th>g2</th>\n",
       "      <th>g3</th>\n",
       "      <th>g4</th>\n",
       "      <th>g5</th>\n",
       "      <th>g6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 42 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   a1  a2  a3  a4  a5  a6  b1  b2  b3  b4 ...  f3  f4  f5  f6  g1  g2  g3  g4  \\\n",
       "0   1   1   1   1   1   1   1   1   1   1 ...   1   1   1   1   1   1   1   1   \n",
       "1   1   1   1   1   1   1   1   1   1   1 ...   1   1   1   1   1   1   1   1   \n",
       "2   1   1   1   1   1   1   0   1   1   1 ...   1   1   1   1   1   1   1   1   \n",
       "3   1   1   1   1   1   1   1   1   1   1 ...   1   1   1   1   1   1   1   1   \n",
       "4   0   1   1   1   1   1   1   1   1   1 ...   1   1   1   1   1   1   1   1   \n",
       "\n",
       "   g5  g6  \n",
       "0   1   1  \n",
       "1   1   1  \n",
       "2   1   1  \n",
       "3   1   1  \n",
       "4   1   1  \n",
       "\n",
       "[5 rows x 42 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# split dataset to 60% training and 40% testing\n",
    "X_train, X_test, y_train, y_test = cross_validation.train_test_split(X, y, test_size=0.4, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(40534, 42) (40534,)\n"
     ]
    }
   ],
   "source": [
    "print (X_train.shape, y_train.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature importances with forests of trees\n",
    "\n",
    "This examples shows the use of forests of trees to evaluate the importance of features on an artificial classification task. The red bars are the feature importances of the forest, along with their inter-trees variability."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "features = index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Feature ranking:\n",
      "1. feature 24 - e1 (0.065720) \n",
      "2. feature 18 - d1 (0.065125) \n",
      "3. feature 19 - d2 (0.063892) \n",
      "4. feature 30 - f1 (0.059698) \n",
      "5. feature 6 - b1 (0.058818) \n",
      "6. feature 12 - c1 (0.054548) \n",
      "7. feature 36 - g1 (0.051978) \n",
      "8. feature 13 - c2 (0.051923) \n",
      "9. feature 25 - e2 (0.041976) \n",
      "10. feature 0 - a1 (0.039567) \n",
      "11. feature 7 - b2 (0.038988) \n",
      "12. feature 31 - f2 (0.037212) \n",
      "13. feature 1 - a2 (0.032235) \n",
      "14. feature 20 - d3 (0.030173) \n",
      "15. feature 14 - c3 (0.029111) \n",
      "16. feature 37 - g2 (0.029028) \n",
      "17. feature 32 - f3 (0.023717) \n",
      "18. feature 2 - a3 (0.023221) \n",
      "19. feature 8 - b3 (0.022005) \n",
      "20. feature 3 - a4 (0.020436) \n",
      "21. feature 38 - g3 (0.019726) \n",
      "22. feature 26 - e3 (0.018618) \n",
      "23. feature 15 - c4 (0.013379) \n",
      "24. feature 39 - g4 (0.012765) \n",
      "25. feature 21 - d4 (0.012557) \n",
      "26. feature 33 - f4 (0.010049) \n",
      "27. feature 9 - b4 (0.009960) \n",
      "28. feature 4 - a5 (0.009011) \n",
      "29. feature 27 - e4 (0.008681) \n",
      "30. feature 40 - g5 (0.008212) \n",
      "31. feature 16 - c5 (0.005241) \n",
      "32. feature 10 - b5 (0.004916) \n",
      "33. feature 34 - f5 (0.004350) \n",
      "34. feature 28 - e5 (0.003893) \n",
      "35. feature 41 - g6 (0.003219) \n",
      "36. feature 17 - c6 (0.002979) \n",
      "37. feature 35 - f6 (0.002731) \n",
      "38. feature 5 - a6 (0.002675) \n",
      "39. feature 22 - d5 (0.002574) \n",
      "40. feature 11 - b6 (0.002352) \n",
      "41. feature 29 - e6 (0.001449) \n",
      "42. feature 23 - d6 (0.001293) \n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHUAAANDCAYAAAAjFiMTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEaUhLuoQEREREREREaWh/wBEAnM9rrC9egAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xb954ef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "\n",
    "# Build a classification task using 3 informative features\n",
    "\n",
    "# Build a forest and compute the feature importances\n",
    "forest = ExtraTreesClassifier(n_estimators=250,\n",
    "                              random_state=0)\n",
    "\n",
    "forest.fit(X, y)\n",
    "importances = forest.feature_importances_\n",
    "std = np.std([tree.feature_importances_ for tree in forest.estimators_],\n",
    "             axis=0)\n",
    "indices = np.argsort(importances)[::-1]\n",
    "\n",
    "# Print the feature ranking\n",
    "print(\"Feature ranking:\")\n",
    "\n",
    "for f in range(X.shape[1]):\n",
    "    print(\"%d. feature %d - %s (%f) \" % (f + 1, indices[f], features[indices[f]], importances[indices[f]]))\n",
    "\n",
    "# Plot the feature importances of the forest\n",
    "plt.figure(num=None, figsize=(14, 10), dpi=80, facecolor='w', edgecolor='k')\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(X.shape[1]), importances[indices],\n",
    "       color=\"r\", yerr=std[indices], align=\"center\")\n",
    "plt.xticks(range(X.shape[1]), indices)\n",
    "plt.xlim([-1, X.shape[1]])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0.06571997,  0.06512489,  0.06389193,  0.0596978 ,  0.05881837])"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "importances[indices[:5]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1. feature 24 - e1 (0.065720)\n",
      "2. feature 18 - d1 (0.065125)\n",
      "3. feature 19 - d2 (0.063892)\n",
      "4. feature 30 - f1 (0.059698)\n",
      "5. feature 6 - b1 (0.058818)\n"
     ]
    }
   ],
   "source": [
    "for f in range(5):\n",
    "    print(\"%d. feature %d - %s (%f)\" % (f + 1, indices[f], features[indices[f]] ,importances[indices[f]]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "best_features = []\n",
    "for i in indices[:5]:\n",
    "    best_features.append(features[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqQAAAINCAYAAADoVW2PAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npGJDSkRERERSsSElIiIiIql+AksRzyPfKF0CAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xb951128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the top 5 feature importances of the forest\n",
    "plt.figure(num=None, figsize=(8, 6), dpi=80, facecolor='w', edgecolor='k')\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(5), importances[indices][:5], \n",
    "       color=\"r\",  yerr=std[indices][:5], align=\"center\")\n",
    "plt.xticks(range(5), best_features)\n",
    "plt.xlim([-1, 5])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Decision Tree accuracy and time elapsed caculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTree\n",
      "Acurracy:  0.904340746771\n",
      "time elapsed:  0.25101423263549805\n"
     ]
    }
   ],
   "source": [
    "t0=time()\n",
    "print (\"DecisionTree\")\n",
    "\n",
    "dt = DecisionTreeClassifier(min_samples_split=20,random_state=99)\n",
    "# dt = DecisionTreeClassifier(min_samples_split=20,max_depth=5,random_state=99)\n",
    "\n",
    "clf_dt=dt.fit(X_train,y_train)\n",
    "\n",
    "print (\"Acurracy: \", clf_dt.score(X_test,y_test))\n",
    "t1=time()\n",
    "print (\"time elapsed: \", t1-t0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## cross validation for DT"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.90341918  0.90282712  0.89661042  0.88535268  0.59970392]\n",
      "0.837582663714\n",
      "time elapsed:  1.5280873775482178\n"
     ]
    }
   ],
   "source": [
    "tt0=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(dt, X, y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt1=time()\n",
    "print (\"time elapsed: \", tt1-tt0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### Tuning our hyperparameters using GridSearch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 27 candidates, totalling 81 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:    8.4s\n",
      "[Parallel(n_jobs=-1)]: Done  81 out of  81 | elapsed:   11.3s finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.016\n",
      "Best parameters set:\n",
      "\tclf__max_depth: 25\n",
      "\tclf__min_samples_leaf: 1\n",
      "\tclf__min_samples_split: 2\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.91      1.00      0.95     24464\n",
      "          1       0.10      0.00      0.00      2559\n",
      "\n",
      "avg / total       0.83      0.90      0.86     27023\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "pipeline = Pipeline([\n",
    "    ('clf', DecisionTreeClassifier(criterion='entropy'))\n",
    "])\n",
    "\n",
    "parameters = {\n",
    "    'clf__max_depth': (5, 25 , 50),\n",
    "    'clf__min_samples_split': (2, 5, 10),\n",
    "    'clf__min_samples_leaf': (1, 2, 3)\n",
    "}\n",
    "\n",
    "grid_search = GridSearchCV(pipeline, parameters, n_jobs=-1, verbose=1, scoring='f1')\n",
    "grid_search.fit(X_train, y_train)\n",
    "\n",
    "print ('Best score: %0.3f' % grid_search.best_score_)\n",
    "print ('Best parameters set:')\n",
    "\n",
    "best_parameters = grid_search.best_estimator_.get_params()\n",
    "for param_name in sorted(parameters.keys()):\n",
    "    print ('\\t%s: %r' % (param_name, best_parameters[param_name]))\n",
    "\n",
    "predictions = grid_search.predict(X_test)\n",
    "\n",
    "\n",
    "print (classification_report(y_test, predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exporting Decision Tree to an Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DecisionTree\n",
      "Acurracy:  0.903600636495\n",
      "time elapsed:  0.23401331901550293\n"
     ]
    }
   ],
   "source": [
    "t0=time()\n",
    "print (\"DecisionTree\")\n",
    "\n",
    "dt1 = DecisionTreeClassifier(min_samples_split=2,max_depth=25,min_samples_leaf=1, random_state=99)\n",
    "\n",
    "clf_dt1=dt1.fit(X_train,y_train)\n",
    "\n",
    "print (\"Acurracy: \", clf_dt1.score(X_test,y_test))\n",
    "t1=time()\n",
    "print (\"time elapsed: \", t1-t0)\n",
    "\n",
    "export_graphviz(clf_dt1, \n",
    "                out_file='tree.dot', \n",
    "                feature_names=features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After we have installed GraphViz on our computer, we can convert the tree.dot file into a PNG file by executing the following command from the command line in the location where we saved the tree.dot file:\n",
    "\n",
    "> dot -Tpng tree.dot -o tree.png"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Random Forest accuracy and time elapsed caculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RandomForest\n",
      "Acurracy:  0.902712504163\n",
      "time elapsed:  2.1101207733154297\n"
     ]
    }
   ],
   "source": [
    "t2=time()\n",
    "print (\"RandomForest\")\n",
    "rf = RandomForestClassifier(n_estimators=100,n_jobs=-1)\n",
    "clf_rf = rf.fit(X_train,y_train)\n",
    "print (\"Acurracy: \", clf_rf.score(X_test,y_test))\n",
    "t3=time()\n",
    "print (\"time elapsed: \", t3-t2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## cross validation for RF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.90216104  0.90223505  0.89779455  0.88675894  0.70288675]\n",
      "0.858367266614\n",
      "time elapsed:  11.505658149719238\n"
     ]
    }
   ],
   "source": [
    "tt2=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(rf, X, y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt3=time()\n",
    "print (\"time elapsed: \", tt3-tt2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Tuning Models using GridSearch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 108 candidates, totalling 324 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done  34 tasks      | elapsed:   16.2s\n",
      "[Parallel(n_jobs=-1)]: Done 184 tasks      | elapsed:  2.3min\n",
      "[Parallel(n_jobs=-1)]: Done 324 out of 324 | elapsed:  6.8min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.904\n",
      "Best parameters set:\n",
      "\tclf__max_depth: 5\n",
      "\tclf__min_samples_leaf: 1\n",
      "\tclf__min_samples_split: 2\n",
      "\tclf__n_estimators: 5\n",
      "Accuracy: 0.905302890131\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.91      1.00      0.95     24464\n",
      "          1       0.00      0.00      0.00      2559\n",
      "\n",
      "avg / total       0.82      0.91      0.86     27023\n",
      "\n"
     ]
    }
   ],
   "source": [
    "pipeline2 = Pipeline([\n",
    "('clf', RandomForestClassifier(criterion='entropy'))\n",
    "])\n",
    "\n",
    "parameters = {\n",
    "    'clf__n_estimators': (5, 25, 50, 100),\n",
    "    'clf__max_depth': (5, 25 , 50),\n",
    "    'clf__min_samples_split': (2, 5, 10),\n",
    "    'clf__min_samples_leaf': (1, 2, 3)\n",
    "}\n",
    "\n",
    "grid_search = GridSearchCV(pipeline2, parameters, n_jobs=-1, verbose=1, scoring='accuracy', cv=3)\n",
    "\n",
    "grid_search.fit(X_train, y_train)\n",
    "\n",
    "print ('Best score: %0.3f' % grid_search.best_score_)\n",
    "\n",
    "print ('Best parameters set:')\n",
    "best_parameters = grid_search.best_estimator_.get_params()\n",
    "\n",
    "for param_name in sorted(parameters.keys()):\n",
    "    print ('\\t%s: %r' % (param_name, best_parameters[param_name]))\n",
    "\n",
    "predictions = grid_search.predict(X_test)\n",
    "print ('Accuracy:', accuracy_score(y_test, predictions))\n",
    "print (classification_report(y_test, predictions))\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Naive Bayes accuracy and time elapsed caculation\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NaiveBayes\n",
      "Acurracy:  0.903637642009\n",
      "time elapsed:  0.32901859283447266\n"
     ]
    }
   ],
   "source": [
    "t4=time()\n",
    "print (\"NaiveBayes\")\n",
    "nb = BernoulliNB()\n",
    "clf_nb=nb.fit(X_train,y_train)\n",
    "print (\"Acurracy: \", clf_nb.score(X_test,y_test))\n",
    "t5=time()\n",
    "print (\"time elapsed: \", t5-t4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## cross-validation for NB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.90031083  0.90257116  0.89150902]\n",
      "0.898130337318\n",
      "time elapsed:  2.033116340637207\n"
     ]
    }
   ],
   "source": [
    "tt4=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(nb, X,y, cv=3)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt5=time()\n",
    "print (\"time elapsed: \", tt5-tt4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# KNN accuracy and time elapsed caculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KNN\n",
      "Acurracy:  0.885319912667\n",
      "time elapsed:  45.90462565422058\n"
     ]
    }
   ],
   "source": [
    "t6=time()\n",
    "print( \"KNN\")\n",
    "# knn = KNeighborsClassifier(n_neighbors=3)\n",
    "knn = KNeighborsClassifier(n_neighbors=3)\n",
    "clf_knn=knn.fit(X_train, y_train)\n",
    "print (\"Acurracy: \", clf_knn.score(X_test,y_test) )\n",
    "t7=time()\n",
    "print (\"time elapsed: \", t7-t6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## cross validation for KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.88521314  0.86486086  0.8633807   0.81822219  0.81828275]\n",
      "0.849991929954\n",
      "time elapsed:  137.34885597229004\n"
     ]
    }
   ],
   "source": [
    "tt6=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(knn, X,y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt7=time()\n",
    "print (\"time elapsed: \", tt7-tt6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fine tuning the model using GridSearch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 2 candidates, totalling 6 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Done   6 out of   6 | elapsed:  1.7min finished\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best score: 0.903\n",
      "Best parameters set:\n",
      "\tn_neighbors: 10\n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "          0       0.91      1.00      0.95     24464\n",
      "          1       0.40      0.00      0.00      2559\n",
      "\n",
      "avg / total       0.86      0.91      0.86     27023\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn import grid_search\n",
    "\n",
    "knn = KNeighborsClassifier()\n",
    "\n",
    "parameters = {'n_neighbors':[1,10]}\n",
    "\n",
    "grid = grid_search.GridSearchCV(knn, parameters, n_jobs=-1, verbose=1, scoring='accuracy')\n",
    "\n",
    "\n",
    "grid.fit(X_train, y_train)\n",
    "\n",
    "print ('Best score: %0.3f' % grid.best_score_)\n",
    "\n",
    "print ('Best parameters set:')\n",
    "best_parameters = grid.best_estimator_.get_params()\n",
    "\n",
    "for param_name in sorted(parameters.keys()):\n",
    "    print ('\\t%s: %r' % (param_name, best_parameters[param_name]))\n",
    "    \n",
    "predictions = grid.predict(X_test)\n",
    "print (classification_report(y_test, predictions))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# SVM accuracy and time elapsed caculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM\n",
      "Acurracy:  0.905302890131\n",
      "time elapsed:  74.45425868034363\n"
     ]
    }
   ],
   "source": [
    "t7=time()\n",
    "print (\"SVM\")\n",
    "\n",
    "svc = SVC(kernel='rbf')\n",
    "clf_svc=svc.fit(X_train, y_train)\n",
    "print (\"Acurracy: \", clf_svc.score(X_test,y_test) )\n",
    "t8=time()\n",
    "print (\"time elapsed: \", t8-t7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## cross validation for SVM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cross result========\n",
      "[ 0.90452931  0.90452931  0.90452931  0.90452224  0.90458919]\n",
      "0.904539871235\n",
      "time elapsed:  2417.2702600955963\n"
     ]
    }
   ],
   "source": [
    "tt7=time()\n",
    "print (\"cross result========\")\n",
    "scores = cross_validation.cross_val_score(svc,X,y, cv=5)\n",
    "print (scores)\n",
    "print (scores.mean())\n",
    "tt8=time()\n",
    "print (\"time elapsed: \", tt7-tt6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "____"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}