X-Company.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/neelima08/anaconda2/lib/python2.7/site-packages/matplotlib/font_manager.py:273: UserWarning: Matplotlib is building the font cache using fc-list. This may take a moment.\n",
      "  warnings.warn('Matplotlib is building the font cache using fc-list. This may take a moment.')\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "#import xgboost as xgb\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn.cross_validation import KFold\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import RandomForestClassifier as RF\n",
    "from sklearn.neighbors import KNeighborsClassifier as KNN\n",
    "from sklearn.linear_model import LogisticRegression as LR\n",
    "from sklearn.ensemble import GradientBoostingClassifier as GB\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.linear_model import RidgeClassifier\n",
    "from sklearn import metrics\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "churn = pd.read_csv(\"data/churn.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "churn.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "churn.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#creating churn column\n",
    "churn['churn'] = churn['last_trip_date'] < '2014-06-01'\n",
    "churn = churn.drop(['phone',], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "#date treatment\n",
    "churn['last_trip_date'] = pd.to_datetime(churn['last_trip_date'])\n",
    "churn['signup_date'] = pd.to_datetime(churn['signup_date'])\n",
    "\n",
    "def to_integer(dt_time):\n",
    "    return 10000*dt_time.year + 100*dt_time.month + dt_time.day\n",
    "\n",
    "churn['last_trip_date']= churn['last_trip_date'].apply(to_integer)\n",
    "churn['signup_date']= churn['signup_date'].apply(to_integer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#changing true/false to booleans\n",
    "cols = ['luxury_car_user','churn']\n",
    "churn[cols] = churn[cols].astype(int)\n",
    "#creating city dummies\n",
    "churn = pd.concat([churn, pd.get_dummies(churn.city, prefix='city', drop_first=True)], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#filling missing values\n",
    "churn['avg_rating_by_driver'] = churn['avg_rating_by_driver'].fillna(churn['avg_rating_by_driver'].mean())\n",
    "churn['avg_rating_of_driver'] = churn['avg_rating_of_driver'].fillna(churn['avg_rating_of_driver'].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "columns = list(churn.columns.values)\n",
    "scaler = StandardScaler()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "y = churn[\"churn\"].values\n",
    "X = churn.drop([\"city\",\"churn\", \"last_trip_date\"], axis = 1).values\n",
    "X = scaler.fit_transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "X_intermed, X_main_test, y_intermed, y_main_test = train_test_split(X, y, random_state=0)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_intermed, y_intermed, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def accuracy(y_true,y_pred):\n",
    "    return np.mean(y_true == y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def cv(X,y,clf_class,**kwargs):\n",
    "    kf = KFold(len(y),n_folds=5,shuffle=True)\n",
    "    y_pred = y.copy()\n",
    "    \n",
    "    for train_index, test_index in kf:\n",
    "        X_train, X_test = X[train_index], X[test_index]\n",
    "        y_train = y[train_index]\n",
    "        clf = clf_class(**kwargs)\n",
    "        clf.fit(X_train,y_train)\n",
    "        y_pred[test_index] = clf.predict(X_test)\n",
    "    return y_pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "GB = GradientBoostingClassifier(n_estimators = 2000, max_features = \"sqrt\", learning_rate=.03, max_depth = 4)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Random_Forest 0.7421\n"
     ]
    }
   ],
   "source": [
    "#print \"SVC:\", accuracy(y, cv(X,y,SVC))\n",
    "print \"Random_Forest\", accuracy(y, cv(X,y, RF))\n",
    "#print \"K-Neighbours\", accuracy(y, cv(X,y, KNN))\n",
    "#print \"Logistic\", accuracy(y, cv(X, y, LR))\n",
    "#print \"GradientBoostingClassifier\", accuracy(y, cv(X,y, GB))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "rf = RandomForestRegressor()\n",
    "rf.fit(X, y)\n",
    "print sorted(zip(map(lambda x: round(x, 4), rf.feature_importances_), columns), \n",
    "             reverse=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "rf1 = RF(n_estimators=13, criterion='entropy', max_depth=10)\n",
    "accuracy(y, cv(X,y, rf1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "param_grid = {\"max_features\":[\"sqrt\"], 'learning_rate': np.linspace(.7, 1.2, 10), \"n_estimators\": range(200, 600, 100)\n",
    "   , \"max_depth\": [2,3,4,5] }\n",
    "g = GridSearchCV(GradientBoostingClassifier(), param_grid , cv=5, n_jobs = -1).fit(X_train, y_train)\n",
    "g.best_score_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA6EAAAIVCAYAAAAgd/n9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nkHMDeC0eOgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0xaf371d8c>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "grad_ens_conf_matrix = metrics.confusion_matrix(y, cv(X, y, GB))\n",
    "ridge_clf_conf_matrix = metrics.confusion_matrix(y, cv(X, y, RidgeClassifier))\n",
    "svm_svc_conf_matrix = metrics.confusion_matrix(y, cv(X, y, SVC))\n",
    "random_forest_conf_matrix = metrics.confusion_matrix(y, cv(X, y, RF))\n",
    "k_neighbors_conf_matrix = metrics.confusion_matrix(y,cv(X, y, KNN))\n",
    "logistic_reg_conf_matrix = metrics.confusion_matrix(y, cv(X, y, LR))\n",
    "conf_matrix = {\n",
    "                \n",
    "                1: {\n",
    "                    'matrix': grad_ens_conf_matrix,\n",
    "                    'title': 'Gradient Boosting',\n",
    "                   },\n",
    "                \n",
    "                2: {\n",
    "                    'matrix': ridge_clf_conf_matrix,\n",
    "                    'title': 'Ridge',\n",
    "                   },\n",
    "                3: {\n",
    "                    'matrix': svm_svc_conf_matrix,\n",
    "                    'title': 'Support Vector Machine',\n",
    "                   },\n",
    "                4: {\n",
    "                    'matrix': random_forest_conf_matrix,\n",
    "                    'title': 'Random Forest',\n",
    "                   },\n",
    "                5: {\n",
    "                    'matrix': k_neighbors_conf_matrix,\n",
    "                    'title': 'K Nearest Neighbors',\n",
    "                   },\n",
    "                6: {\n",
    "                    'matrix': logistic_reg_conf_matrix,\n",
    "                    'title': 'Logistic Regression',\n",
    "                   }}\n",
    "\n",
    "\n",
    "fix, ax = plt.subplots(figsize=(16, 12))\n",
    "plt.suptitle('Confusion Matrix of Various Classifiers')\n",
    "for ii, values in conf_matrix.items():\n",
    "    matrix = values['matrix']\n",
    "    title = values['title']\n",
    "    plt.subplot(3, 3, ii) # starts from 1\n",
    "    plt.title(title);\n",
    "    sns.heatmap(matrix, annot=True,  fmt='');\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "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": "2.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}