Exploration and benchmark.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 266,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "import numpy as np\n",
    "import datetime\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import precision_score, recall_score\n",
    "% matplotlib inline\n",
    "plt.style.use('ggplot')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 267,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_customer = pd.read_csv('data/customer.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([u'customerId2', u'churnlabel', u'gender', u'shippingCountry',\n",
       "       u'dateCreated', u'yearOfBirth', u'premier'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 268,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_customer.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_receipts = pd.read_csv('data/receipts.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([u'customerId2', u'productId', u'divisionId', u'sourceId', u'itemQty',\n",
       "       u'signalDate', u'receiptId', u'price'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_receipts.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_returns = pd.read_csv('data/returns.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([u'customerId2', u'productId', u'divisionId', u'sourceId', u'itemQty',\n",
       "       u'signalDate', u'receiptId', u'returnId', u'returnAction',\n",
       "       u'returnReason'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 272,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_returns.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 273,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_sessionsummary = pd.read_csv('data/sessionsummary.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 274,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([u'customerId2', u'country', u'startTime', u'site', u'pageViewCount',\n",
       "       u'nonPageViewEventsCount', u'userAgent', u'screenResolution',\n",
       "       u'browserSize', u'productViewCount', u'productViewsDistinctCount',\n",
       "       u'productsAddedToBagCount',\n",
       "       u'productsSavedForLaterFromProductPageCount',\n",
       "       u'productsSavedForLaterFromCategoryPageCount',\n",
       "       u'productsPurchasedDistinctCount', u'productsPurchasedTotalCount'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 274,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sessionsummary.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Customer data exploration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(470169, 7)"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_customer.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "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>customerId2</th>\n",
       "      <th>churnlabel</th>\n",
       "      <th>gender</th>\n",
       "      <th>shippingCountry</th>\n",
       "      <th>dateCreated</th>\n",
       "      <th>yearOfBirth</th>\n",
       "      <th>premier</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1852043</td>\n",
       "      <td>2</td>\n",
       "      <td>F</td>\n",
       "      <td>UK</td>\n",
       "      <td>2005-03-02T19:35:50.000Z</td>\n",
       "      <td>1986</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1825138</td>\n",
       "      <td>1</td>\n",
       "      <td>F</td>\n",
       "      <td>UK</td>\n",
       "      <td>2005-11-25T20:32:39.000Z</td>\n",
       "      <td>1981</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>863960</td>\n",
       "      <td>1</td>\n",
       "      <td>M</td>\n",
       "      <td>UK</td>\n",
       "      <td>2005-05-12T13:04:19.000Z</td>\n",
       "      <td>1983</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>585725</td>\n",
       "      <td>1</td>\n",
       "      <td>F</td>\n",
       "      <td>UK</td>\n",
       "      <td>2006-03-23T16:59:31.000Z</td>\n",
       "      <td>1976</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1773666</td>\n",
       "      <td>2</td>\n",
       "      <td>F</td>\n",
       "      <td>UK</td>\n",
       "      <td>2006-04-20T18:42:44.000Z</td>\n",
       "      <td>1990</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   customerId2  churnlabel gender shippingCountry               dateCreated  \\\n",
       "0      1852043           2      F              UK  2005-03-02T19:35:50.000Z   \n",
       "1      1825138           1      F              UK  2005-11-25T20:32:39.000Z   \n",
       "2       863960           1      M              UK  2005-05-12T13:04:19.000Z   \n",
       "3       585725           1      F              UK  2006-03-23T16:59:31.000Z   \n",
       "4      1773666           2      F              UK  2006-04-20T18:42:44.000Z   \n",
       "\n",
       "   yearOfBirth  premier  \n",
       "0         1986        0  \n",
       "1         1981        0  \n",
       "2         1983        0  \n",
       "3         1976        6  \n",
       "4         1990        0  "
      ]
     },
     "execution_count": 276,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_customer.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('customerId2', 470169, 0)\n",
      "('churnlabel', 2, 0)\n",
      "('gender', 2, 0)\n",
      "('shippingCountry', 190, 11)\n",
      "('dateCreated', 470109, 0)\n",
      "('yearOfBirth', 125, 0)\n",
      "('premier', 7, 0)\n"
     ]
    }
   ],
   "source": [
    "for k in df_customer.keys():\n",
    "    print(k, df_customer[k].nunique(), df_customer[k].isnull().sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_customer['Age'] = 2017-df_customer['yearOfBirth']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_customer['subscription_date'] = pd.to_datetime(df_customer['dateCreated']).dt.date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "def barchart(df, column, title='', labels=-1, vertical=False):\n",
    "    if labels==-1:\n",
    "        labels = df[column].unique()\n",
    "    values = df[column].value_counts()/float(len(df))*100\n",
    "    \n",
    "    #for v in df[column].unique():\n",
    "    #    values.append((df[column]==v).sum()/float(len(df))*100)\n",
    "    if vertical:\n",
    "        fig, ax = plt.subplots(figsize=(8,10))\n",
    "        ax.barh(range(len(values)), values[::-1])\n",
    "        ax.set_xlim(0, 100) \n",
    "        ax.set_yticks(np.arange(0.4, len(labels)+0.4, 1))\n",
    "        ax.set_yticklabels(labels[::-1])\n",
    "    else:\n",
    "        fig, ax = plt.subplots(figsize=(5,3))\n",
    "        ax.bar(range(len(values)), values)\n",
    "        ax.set_ylim(0, 100) \n",
    "        ax.set_xticks(np.arange(0.4, len(labels)+0.4, 1))\n",
    "        ax.set_xticklabels(labels)\n",
    "    \n",
    "    ax.set_title(title)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([355708, 114461])"
      ]
     },
     "execution_count": 281,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_customer.gender.value_counts().values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb0AAAErCAYAAABKEV0aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n6AEArEHoAQCs8f8Bj4ZziWBjBKYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1399d0ad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "barchart(df_customer, 'churnlabel', title='Churn distribution')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb0AAAErCAYAAABKEV0aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAwBYg9ADAFiD0AMAWIPQAwBY4/8B75ydef7z8MMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11c9a8790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "barchart(df_customer, 'gender', title='Gender distribution')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x11e0730d0>"
      ]
     },
     "execution_count": 284,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjIAAAF5CAYAAACFu8BrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x115e72150>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_customer.Age.hist()\n",
    "plt.title('Age distribution')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1434"
      ]
     },
     "execution_count": 285,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# removing outliers\n",
    "sum(df_customer.Age>100)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "big = df_customer.groupby('shippingCountry').customerId2.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_customer_big = df_customer[df_customer.shippingCountry.isin(big[big>2000].keys())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Discarding 8.14% of data points\n"
     ]
    }
   ],
   "source": [
    "df_customer_big = df_customer_big[(df_customer_big.Age>=15) & (df_customer_big.Age<80)]\n",
    "print('Discarding {:.2f}% of data points'.format((1-len(df_customer_big)/float(len(df_customer)))*100))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {
    "collapsed": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAx8AAANKCAYAAADr5/PhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11642f650>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "barchart(df_customer_big, 'shippingCountry', title='Country distribution', vertical=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "idx = np.random.permutation(df_customer_big.shape[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Can we say something with this really basic info?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df_features = df_customer_big.copy()\n",
    "\n",
    "df_features.loc[df_features.gender=='F', 'gender'] = 0\n",
    "df_features.loc[df_features.gender=='M', 'gender'] = 1\n",
    "\n",
    "df_features.drop('yearOfBirth', axis=1, inplace=True)\n",
    "\n",
    "df_features.drop('dateCreated', axis=1, inplace=True)\n",
    "\n",
    "#df_features['n_days'] = (datetime.date.today()-df_features['subscription_date']).dt.days\n",
    "\n",
    "#df_features.n_days.max()\n",
    "\n",
    "#df_features=df_features[df_features.n_days<10000]\n",
    "\n",
    "df_features.drop('subscription_date', axis=1, inplace=True)\n",
    "df_features.drop('customerId2', axis=1, inplace=True)\n",
    "\n",
    "countrydum = pd.get_dummies(df_features['shippingCountry'])\n",
    "df_features = pd.concat([df_features, countrydum],axis=1)\n",
    "\n",
    "df_features.drop('shippingCountry', axis=1, inplace=True)\n",
    "\n",
    "premierdum = pd.get_dummies(df_features.premier)\n",
    "\n",
    "df_features = pd.concat([df_features, premierdum],axis=1)\n",
    "\n",
    "df_features.drop('premier', axis=1, inplace=True)\n",
    "df_features.fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "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>churnlabel</th>\n",
       "      <th>gender</th>\n",
       "      <th>Age</th>\n",
       "      <th>Australia</th>\n",
       "      <th>Belgium</th>\n",
       "      <th>Canada</th>\n",
       "      <th>Denmark</th>\n",
       "      <th>France</th>\n",
       "      <th>Germany</th>\n",
       "      <th>Hong Kong</th>\n",
       "      <th>...</th>\n",
       "      <th>Sweden</th>\n",
       "      <th>UK</th>\n",
       "      <th>United States</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>41</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>27</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   churnlabel  gender  Age  Australia  Belgium  Canada  Denmark  France  \\\n",
       "0           2       0   31        0.0      0.0     0.0      0.0     0.0   \n",
       "1           1       0   36        0.0      0.0     0.0      0.0     0.0   \n",
       "2           1       1   34        0.0      0.0     0.0      0.0     0.0   \n",
       "3           1       0   41        0.0      0.0     0.0      0.0     0.0   \n",
       "4           2       0   27        0.0      0.0     0.0      0.0     0.0   \n",
       "\n",
       "   Germany  Hong Kong ...   Sweden   UK  United States    0    1    2    3  \\\n",
       "0      0.0        0.0 ...      0.0  1.0            0.0  1.0  0.0  0.0  0.0   \n",
       "1      0.0        0.0 ...      0.0  1.0            0.0  1.0  0.0  0.0  0.0   \n",
       "2      0.0        0.0 ...      0.0  1.0            0.0  1.0  0.0  0.0  0.0   \n",
       "3      0.0        0.0 ...      0.0  1.0            0.0  0.0  0.0  0.0  0.0   \n",
       "4      0.0        0.0 ...      0.0  1.0            0.0  1.0  0.0  0.0  0.0   \n",
       "\n",
       "     4    5    6  \n",
       "0  0.0  0.0  0.0  \n",
       "1  0.0  0.0  0.0  \n",
       "2  0.0  0.0  0.0  \n",
       "3  0.0  0.0  1.0  \n",
       "4  0.0  0.0  0.0  \n",
       "\n",
       "[5 rows x 27 columns]"
      ]
     },
     "execution_count": 292,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "Y = np.array(df_features['churnlabel'])\n",
    "X = np.array(df_features.drop('churnlabel', axis=1)/df_features.drop('churnlabel', axis=1).max())\n",
    "\n",
    "\n",
    "\n",
    "ntrain = int(2*len(idx)/3)\n",
    "xtrain = X[idx[:ntrain], :]\n",
    "ytrain = Y[idx[:ntrain]]\n",
    "xtest = X[idx[ntrain:], :]\n",
    "ytest = Y[idx[ntrain:]]\n",
    "\n",
    "clf = RandomForestClassifier()\n",
    "\n",
    "clf.fit(xtrain, ytrain)\n",
    "\n",
    "pred = clf.predict(xtest)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[50853, 24796],\n",
       "       [34063, 34248]])"
      ]
     },
     "execution_count": 294,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "confusion_matrix(ytest, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 298,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.59114337315921084"
      ]
     },
     "execution_count": 298,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "1-sum(np.abs(ytest-pred))/float(len(pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.59886240520043332"
      ]
     },
     "execution_count": 299,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision_score(ytest, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 300,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.67222303004666284"
      ]
     },
     "execution_count": 300,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall_score(ytest, pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x11ebed810>"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7wAAAKvCAYAAABJdsO2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11ed1fa50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(10, 8))\n",
    "ax.barh(range(len(clf.feature_importances_)), clf.feature_importances_)\n",
    "ax.set_yticks(range(len(df_features.columns[1:])))\n",
    "ax.set_yticklabels(df_features.columns[1:])\n",
    "ax.set_title('Feature importance')"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}