XGBOOST_RETAIL.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## XGBOOST SIN UNIRED"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "path1 = '/gpfs/fs01/user/s57f-73edb3521eee1d-79e3f294add4/notebook/data2/bloqueo2balanceado.csv'\n",
    "path2 = '/gpfs/fs01/user/s57f-73edb3521eee1d-79e3f294add4/notebook/data2/unicard2balanceado.csv'\n",
    "path3 = '/gpfs/fs01/user/s57f-73edb3521eee1d-79e3f294add4/notebook/data2/muestra2desbalanceada.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "bloq2bal = pd.read_csv(path1, index_col=False, header=0, sep='\\t')\n",
    "unic2bal = pd.read_csv(path2, index_col=False, header=0, sep='\\t')\n",
    "mues2bal = pd.read_csv(path3, index_col=False, header=0, sep='\\t')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# DETERMINACIÓN DE PAGO BLOQUEADOS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "pd_boost = bloq2bal"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CUSTOMER_ID</th>\n",
       "      <th>std_pago_unired</th>\n",
       "      <th>max_pago_unired</th>\n",
       "      <th>avg_pago_unired</th>\n",
       "      <th>pago_unired_mensual</th>\n",
       "      <th>avg_dia_pago_unired</th>\n",
       "      <th>max_dia_pago_unired</th>\n",
       "      <th>min_dia_pago_unired</th>\n",
       "      <th>avg_uso_tarjeta</th>\n",
       "      <th>PAYMENT_MEAN_ID</th>\n",
       "      <th>...</th>\n",
       "      <th>std_pago_retail_mensual</th>\n",
       "      <th>SPEND_AMT</th>\n",
       "      <th>DISCOUNT_AMT</th>\n",
       "      <th>GENDER</th>\n",
       "      <th>EDAD</th>\n",
       "      <th>MOBILE_CONTACTABILITY</th>\n",
       "      <th>EMAIL_CONTACTABILITY</th>\n",
       "      <th>GSE</th>\n",
       "      <th>CIVIL_STATUS</th>\n",
       "      <th>bloqueo</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>14295828</td>\n",
       "      <td>34786.587006</td>\n",
       "      <td>135466</td>\n",
       "      <td>54942.416667</td>\n",
       "      <td>659309</td>\n",
       "      <td>5.089003</td>\n",
       "      <td>10.153817</td>\n",
       "      <td>0.707107</td>\n",
       "      <td>0.250000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>19159.363619</td>\n",
       "      <td>260589</td>\n",
       "      <td>23785</td>\n",
       "      <td>FEMENINO</td>\n",
       "      <td>60</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>C3</td>\n",
       "      <td>CASADO</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>30174068</td>\n",
       "      <td>22804.835996</td>\n",
       "      <td>97050</td>\n",
       "      <td>50127.857143</td>\n",
       "      <td>350895</td>\n",
       "      <td>5.595878</td>\n",
       "      <td>6.363961</td>\n",
       "      <td>4.949747</td>\n",
       "      <td>0.142857</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>5602.529458</td>\n",
       "      <td>29387</td>\n",
       "      <td>1633</td>\n",
       "      <td>FEMENINO</td>\n",
       "      <td>59</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>E</td>\n",
       "      <td>DIVORCIADO</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>54612656</td>\n",
       "      <td>4587.755442</td>\n",
       "      <td>43850</td>\n",
       "      <td>38700.000000</td>\n",
       "      <td>116100</td>\n",
       "      <td>2.819995</td>\n",
       "      <td>4.932883</td>\n",
       "      <td>0.707107</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>23359.003782</td>\n",
       "      <td>218009</td>\n",
       "      <td>22955</td>\n",
       "      <td>FEMENINO</td>\n",
       "      <td>42</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>C3</td>\n",
       "      <td>DIVORCIADO</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>99841158</td>\n",
       "      <td>24316.463764</td>\n",
       "      <td>77890</td>\n",
       "      <td>39529.000000</td>\n",
       "      <td>395290</td>\n",
       "      <td>10.563605</td>\n",
       "      <td>10.563605</td>\n",
       "      <td>10.563605</td>\n",
       "      <td>0.005102</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>56429.884905</td>\n",
       "      <td>1037580</td>\n",
       "      <td>247719</td>\n",
       "      <td>FEMENINO</td>\n",
       "      <td>51</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>E</td>\n",
       "      <td>CASADO</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>70026320</td>\n",
       "      <td>28030.667045</td>\n",
       "      <td>104700</td>\n",
       "      <td>63497.000000</td>\n",
       "      <td>317485</td>\n",
       "      <td>9.364604</td>\n",
       "      <td>11.150486</td>\n",
       "      <td>7.778175</td>\n",
       "      <td>0.477157</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>38881.129212</td>\n",
       "      <td>1100191</td>\n",
       "      <td>128563</td>\n",
       "      <td>MASCULINO</td>\n",
       "      <td>48</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>C3</td>\n",
       "      <td>SOLTERO</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   CUSTOMER_ID  std_pago_unired  max_pago_unired  avg_pago_unired  \\\n",
       "0     14295828     34786.587006           135466     54942.416667   \n",
       "1     30174068     22804.835996            97050     50127.857143   \n",
       "2     54612656      4587.755442            43850     38700.000000   \n",
       "3     99841158     24316.463764            77890     39529.000000   \n",
       "4     70026320     28030.667045           104700     63497.000000   \n",
       "\n",
       "   pago_unired_mensual  avg_dia_pago_unired  max_dia_pago_unired  \\\n",
       "0               659309             5.089003            10.153817   \n",
       "1               350895             5.595878             6.363961   \n",
       "2               116100             2.819995             4.932883   \n",
       "3               395290            10.563605            10.563605   \n",
       "4               317485             9.364604            11.150486   \n",
       "\n",
       "   min_dia_pago_unired  avg_uso_tarjeta  PAYMENT_MEAN_ID   ...    \\\n",
       "0             0.707107         0.250000                0   ...     \n",
       "1             4.949747         0.142857                0   ...     \n",
       "2             0.707107         0.000000                0   ...     \n",
       "3            10.563605         0.005102                0   ...     \n",
       "4             7.778175         0.477157                1   ...     \n",
       "\n",
       "   std_pago_retail_mensual  SPEND_AMT  DISCOUNT_AMT     GENDER  EDAD  \\\n",
       "0             19159.363619     260589         23785   FEMENINO    60   \n",
       "1              5602.529458      29387          1633   FEMENINO    59   \n",
       "2             23359.003782     218009         22955   FEMENINO    42   \n",
       "3             56429.884905    1037580        247719   FEMENINO    51   \n",
       "4             38881.129212    1100191        128563  MASCULINO    48   \n",
       "\n",
       "  MOBILE_CONTACTABILITY  EMAIL_CONTACTABILITY  GSE  CIVIL_STATUS bloqueo  \n",
       "0                     1                     1   C3        CASADO       0  \n",
       "1                     1                     0    E    DIVORCIADO       1  \n",
       "2                     1                     0   C3    DIVORCIADO       0  \n",
       "3                     1                     0    E        CASADO       0  \n",
       "4                     1                     0   C3       SOLTERO       0  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd_boost.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "## One Hot Encoder para GSE\n",
    "pd_boost = pd.get_dummies(pd_boost, columns=[\"GSE\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "## One Hot Encoder para CIVIL_STATUS\n",
    "pd_boost = pd.get_dummies(pd_boost, columns=[\"CIVIL_STATUS\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Reemplazamos por diccionario el Genero\n",
    "dict_GEN = {\"GENDER\":{\"FEMENINO\":0, \"MASCULINO\":1}}\n",
    "pd_boost.replace(dict_GEN, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Reemplazamos todos los nan por ceros\n",
    "name_col  = list(pd_boost)\n",
    "\n",
    "for name in name_col:\n",
    "    if name != 'CUSTOMER_ID':\n",
    "        error = pd_boost[pd_boost[name].isnull() == True]\n",
    "        if len(error) > 0:\n",
    "            pd_boost[name].fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['CUSTOMER_ID',\n",
       " 'std_pago_unired',\n",
       " 'max_pago_unired',\n",
       " 'avg_pago_unired',\n",
       " 'pago_unired_mensual',\n",
       " 'avg_dia_pago_unired',\n",
       " 'max_dia_pago_unired',\n",
       " 'min_dia_pago_unired',\n",
       " 'avg_uso_tarjeta',\n",
       " 'PAYMENT_MEAN_ID',\n",
       " 'avg_pago_retail_mensual',\n",
       " 'max_pago_retail_mensual',\n",
       " 'std_pago_retail_mensual',\n",
       " 'SPEND_AMT',\n",
       " 'DISCOUNT_AMT',\n",
       " 'GENDER',\n",
       " 'EDAD',\n",
       " 'MOBILE_CONTACTABILITY',\n",
       " 'EMAIL_CONTACTABILITY',\n",
       " 'bloqueo',\n",
       " 'GSE_ABC1',\n",
       " 'GSE_C2',\n",
       " 'GSE_C3',\n",
       " 'GSE_D',\n",
       " 'GSE_E',\n",
       " 'CIVIL_STATUS_CASADO',\n",
       " 'CIVIL_STATUS_DIVORCIADO',\n",
       " 'CIVIL_STATUS_SOLTERO',\n",
       " 'CIVIL_STATUS_VIUDO']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "name_col"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MODELO XGBOOST"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## XGBOOST CON CROSSVALIDATION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Definimos vectores X e Y \n",
    "Y = pd_boost['bloqueo'].map(lambda d: 1 if d == 1 else 0)\n",
    "X = X = pd_boost.drop(['PAYMENT_MEAN_ID', 'CUSTOMER_ID', 'bloqueo', 'std_pago_unired', 'max_pago_unired', 'avg_pago_unired',\n",
    "       'pago_unired_mensual', 'avg_dia_pago_unired', 'max_dia_pago_unired', 'min_dia_pago_unired'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/gpfs/fs01/user/s57f-73edb3521eee1d-79e3f294add4/.local/lib/python3.5/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import xgboost\n",
    "from xgboost.sklearn import XGBClassifier\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "from xgboost import plot_importance\n",
    "from matplotlib import pyplot\n",
    "import pprint\n",
    "\n",
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.75 (+/- 0.02)\n",
      "Duración 0:10:48.758504\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f78348549b0>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiAAAAGMCAYAAADjvxkUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f78dcb23ac8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import datetime as dt\n",
    "\n",
    "clf = make_pipeline(StandardScaler(), XGBClassifier(nthread=-1))\n",
    "start_time = dt.datetime.now()\n",
    "scores = cross_val_score(clf, X.values, Y.values, cv = 10, scoring = 'accuracy', n_jobs = 1)\n",
    "\n",
    "print(\"Accuracy: %0.2f (+/- %0.2f)\" % (scores.mean(), scores.std()*2))\n",
    "print('Duración', dt.datetime.now()-start_time)\n",
    "\n",
    "clf.fit(X.values, Y.values)\n",
    "xgboost.plot_importance(clf.steps[1][1], ax=plt.gca())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### XGBOOST CON MATRIZ DE CONFUSIÓN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Definimos vectores de entrenamiento y test\n",
    "from sklearn.cross_validation import train_test_split \n",
    "\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.3, random_state=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training records: 282245\n",
      "Number of testing records : 120973\n"
     ]
    }
   ],
   "source": [
    "print(\"Number of training records: \" + str(X_train.size))\n",
    "print(\"Number of testing records : \" + str(X_test.size))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f783172abe0>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhsAAAGMCAYAAABzrGDbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f783484f7b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pipeline = Pipeline([('scaler', StandardScaler()), ('classifier', XGBClassifier())])\n",
    "pipeline.fit(X_train.values, Y_train.values)\n",
    "xgboost.plot_importance(pipeline.steps[1][1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2288 889 718 2472\n"
     ]
    }
   ],
   "source": [
    "Y_pred = pipeline.predict(X_test.values)\n",
    "\n",
    "tn, fp, fn, tp = confusion_matrix(Y_test, Y_pred).ravel()\n",
    "\n",
    "print(tn, fp, fn, tp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### XGBOOST CON OPTIMIZACIÓN DE AUC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/gpfs/fs01/user/s57f-73edb3521eee1d-79e3f294add4/.local/lib/python3.5/site-packages/sklearn/grid_search.py:42: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. This module will be removed in 0.20.\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np \n",
    "import pandas as pd\n",
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from operator import itemgetter\n",
    "import xgboost as xgb\n",
    "from xgboost import plot_tree\n",
    "import random\n",
    "import time\n",
    "from sklearn.grid_search import GridSearchCV\n",
    "from sklearn.metrics import average_precision_score\n",
    "import matplotlib.pyplot as plt\n",
    "from numpy import genfromtxt\n",
    "import seaborn as sns\n",
    "from sklearn import preprocessing\n",
    "from sklearn.metrics import roc_curve, auc,recall_score,precision_score\n",
    "import datetime as dt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_feature_map(features):\n",
    "    outfile = open('xgb.fmap', 'w')\n",
    "    for i, feat in enumerate(features):\n",
    "        outfile.write('{0}\\t{1}\\tq\\n'.format(i, feat))\n",
    "    outfile.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_importance(gbm, features):\n",
    "    #create_feature_map(features)\n",
    "    importance = gbm.get_fscore()\n",
    "    importance = sorted(importance.items(), key=itemgetter(1), reverse=True)\n",
    "    return importance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_features(train, test):\n",
    "    trainval = list(train.columns.values)\n",
    "    output = trainval\n",
    "    return sorted(output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def run_single(X_train, X_valid, y_train, y_valid, features, random_state=0):\n",
    "    eta = 1.0\n",
    "    max_depth= 6 \n",
    "    subsample = 1\n",
    "    colsample_bytree = 1\n",
    "    min_chil_weight=1\n",
    "    start_time = time.time()\n",
    "\n",
    "    print('XGBoost params. ETA: {}, MAX_DEPTH: {}, SUBSAMPLE: {}, COLSAMPLE_BY_TREE: {}'.format(eta, max_depth, subsample, colsample_bytree))\n",
    "    params = {\n",
    "        \"objective\": \"binary:logistic\",\n",
    "        \"booster\" : \"gbtree\",\n",
    "        \"eval_metric\": \"auc\",\n",
    "        \"eta\": eta,\n",
    "        \"tree_method\": 'exact',\n",
    "        \"max_depth\": max_depth,\n",
    "        \"subsample\": subsample,\n",
    "        \"colsample_bytree\": colsample_bytree,\n",
    "        \"silent\": 1,\n",
    "        \"min_chil_weight\":min_chil_weight,\n",
    "        \"seed\": random_state,\n",
    "        #\"num_class\" : 22,\n",
    "    }\n",
    "    num_boost_round = 500\n",
    "    early_stopping_rounds = 20\n",
    "    test_size = 0.1\n",
    "\n",
    "   \n",
    "    \n",
    "    #X_train, X_valid = train_test_split(train, test_size=test_size, random_state=random_state)\n",
    "    print('Length train:', X_train.shape[0])\n",
    "    print('Length valid:', X_valid.shape[0])\n",
    "    #y_train = X_train[target]\n",
    "    #y_valid = X_valid[target]\n",
    "    dtrain = xgb.DMatrix(X_train[features], y_train, missing=-99, feature_names=features)\n",
    "    dvalid = xgb.DMatrix(X_valid[features], y_valid, missing =-99, feature_names=features)\n",
    "\n",
    "    watchlist = [(dtrain, 'train'), (dvalid, 'eval')]\n",
    "    gbm = xgb.train(params, dtrain, num_boost_round, evals=watchlist, early_stopping_rounds=early_stopping_rounds, verbose_eval=True)\n",
    "\n",
    "    print(\"Validating...\")\n",
    "    check = gbm.predict(xgb.DMatrix(X_valid[features]), ntree_limit=gbm.best_iteration+1)\n",
    "    \n",
    "    #area under the precision-recall curve\n",
    "    score = average_precision_score(y_valid.values, check)\n",
    "    print('area under the precision-recall curve: {:.6f}'.format(score))\n",
    "\n",
    "    \n",
    "    check2=check.round()\n",
    "    score = precision_score(y_valid.values, check2)\n",
    "    print('precision score: {:.6f}'.format(score))\n",
    "\n",
    "    score = recall_score(y_valid.values, check2)\n",
    "    print('recall score: {:.6f}'.format(score))\n",
    "    \n",
    "    imp = get_importance(gbm, features)\n",
    "    xgboost.plot_importance(gbm)\n",
    "    print('Importance array: ', imp)\n",
    "\n",
    "    print(\"Predict test set... \")\n",
    "    #fea = ['f0', 'f1', 'f2', 'f3', 'f4', 'f5', 'f6', 'f7', 'f8', 'f9', 'f10', 'f11', 'f12', 'f13', 'f14']\n",
    "    test_prediction = gbm.predict(xgb.DMatrix(X_valid[features],missing = -99), ntree_limit=gbm.best_iteration+1)\n",
    "    score = average_precision_score(y_valid.values, test_prediction)\n",
    "\n",
    "    print('area under the precision-recall curve test set: {:.6f}'.format(score))\n",
    "    \n",
    "    ############################################ ROC Curve\n",
    "     \n",
    "    # Compute micro-average ROC curve and ROC area\n",
    "    fpr, tpr, _ = roc_curve(y_valid.values, check)\n",
    "    roc_auc = auc(fpr, tpr)\n",
    "    #xgb.plot_importance(gbm)\n",
    "    #plt.show()\n",
    "    plt.figure()\n",
    "    lw = 2\n",
    "    plt.plot(fpr, tpr, color='darkorange',\n",
    "             lw=lw, label='ROC curve (area = %0.2f)' % roc_auc)\n",
    "    plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')\n",
    "    plt.xlim([-0.02, 1.0])\n",
    "    plt.ylim([0.0, 1.05])\n",
    "    plt.xlabel('False Positive Rate')\n",
    "    plt.ylabel('True Positive Rate')\n",
    "    plt.title('ROC curve')\n",
    "    plt.legend(loc=\"lower right\")\n",
    "    plt.show()\n",
    "    ##################################################\n",
    "    \n",
    "    # Plot tree\n",
    "    plt.figure()\n",
    "    plot = plot_tree(gbm, num_trees=2, rankdir='LR')\n",
    "    fig = plot.get_figure()\n",
    "    fig.savefig('xgboost_tree.eps', format='eps', dpi=2000)\n",
    "    plt.show()\n",
    "\n",
    "    print('Training time: {} minutes'.format(round((time.time() - start_time)/60, 2)))\n",
    "    return test_prediction, imp, gbm.best_iteration+1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Start time:  2017-12-15 13:47:00.302057\n"
     ]
    }
   ],
   "source": [
    "# Any results you write to the current directory are saved as output.\n",
    "start_time = dt.datetime.now()\n",
    "print(\"Start time: \",start_time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['avg_uso_tarjeta', 'avg_pago_retail_mensual', 'max_pago_retail_mensual', 'std_pago_retail_mensual', 'SPEND_AMT', 'DISCOUNT_AMT', 'GENDER', 'EDAD', 'MOBILE_CONTACTABILITY', 'EMAIL_CONTACTABILITY', 'GSE_ABC1', 'GSE_C2', 'GSE_C3', 'GSE_D', 'GSE_E', 'CIVIL_STATUS_CASADO', 'CIVIL_STATUS_DIVORCIADO', 'CIVIL_STATUS_SOLTERO', 'CIVIL_STATUS_VIUDO']\n",
      "Building model..  0:00:00.011784\n",
      "XGBoost params. ETA: 1.0, MAX_DEPTH: 6, SUBSAMPLE: 1, COLSAMPLE_BY_TREE: 1\n",
      "Length train: 14855\n",
      "Length valid: 6367\n",
      "[0]\ttrain-auc:0.757463\teval-auc:0.749027\n",
      "Multiple eval metrics have been passed: 'eval-auc' will be used for early stopping.\n",
      "\n",
      "Will train until eval-auc hasn't improved in 20 rounds.\n",
      "[1]\ttrain-auc:0.807888\teval-auc:0.793832\n",
      "[2]\ttrain-auc:0.837175\teval-auc:0.819701\n",
      "[3]\ttrain-auc:0.86979\teval-auc:0.850615\n",
      "[4]\ttrain-auc:0.894883\teval-auc:0.877624\n",
      "[5]\ttrain-auc:0.91415\teval-auc:0.896831\n",
      "[6]\ttrain-auc:0.923875\teval-auc:0.907919\n",
      "[7]\ttrain-auc:0.939856\teval-auc:0.927347\n",
      "[8]\ttrain-auc:0.953382\teval-auc:0.939905\n",
      "[9]\ttrain-auc:0.960329\teval-auc:0.947321\n",
      "[10]\ttrain-auc:0.966613\teval-auc:0.952963\n",
      "[11]\ttrain-auc:0.970297\teval-auc:0.956237\n",
      "[12]\ttrain-auc:0.975085\teval-auc:0.96058\n",
      "[13]\ttrain-auc:0.978212\teval-auc:0.963945\n",
      "[14]\ttrain-auc:0.980237\teval-auc:0.96746\n",
      "[15]\ttrain-auc:0.984714\teval-auc:0.971576\n",
      "[16]\ttrain-auc:0.987727\teval-auc:0.974504\n",
      "[17]\ttrain-auc:0.989634\teval-auc:0.977755\n",
      "[18]\ttrain-auc:0.991087\teval-auc:0.980648\n",
      "[19]\ttrain-auc:0.992252\teval-auc:0.982123\n",
      "[20]\ttrain-auc:0.993121\teval-auc:0.983135\n",
      "[21]\ttrain-auc:0.994559\teval-auc:0.985108\n",
      "[22]\ttrain-auc:0.995742\teval-auc:0.986408\n",
      "[23]\ttrain-auc:0.996143\teval-auc:0.987498\n",
      "[24]\ttrain-auc:0.996544\teval-auc:0.98831\n",
      "[25]\ttrain-auc:0.996636\teval-auc:0.988541\n",
      "[26]\ttrain-auc:0.99709\teval-auc:0.988913\n",
      "[27]\ttrain-auc:0.997854\teval-auc:0.990246\n",
      "[28]\ttrain-auc:0.998484\teval-auc:0.991086\n",
      "[29]\ttrain-auc:0.998723\teval-auc:0.991707\n",
      "[30]\ttrain-auc:0.998909\teval-auc:0.992368\n",
      "[31]\ttrain-auc:0.998995\teval-auc:0.992804\n",
      "[32]\ttrain-auc:0.999098\teval-auc:0.993185\n",
      "[33]\ttrain-auc:0.999208\teval-auc:0.993454\n",
      "[34]\ttrain-auc:0.999284\teval-auc:0.993867\n",
      "[35]\ttrain-auc:0.999344\teval-auc:0.994077\n",
      "[36]\ttrain-auc:0.999511\teval-auc:0.994298\n",
      "[37]\ttrain-auc:0.999535\teval-auc:0.994467\n",
      "[38]\ttrain-auc:0.999661\teval-auc:0.994506\n",
      "[39]\ttrain-auc:0.999712\teval-auc:0.99458\n",
      "[40]\ttrain-auc:0.999761\teval-auc:0.994836\n",
      "[41]\ttrain-auc:0.999827\teval-auc:0.995116\n",
      "[42]\ttrain-auc:0.999846\teval-auc:0.995134\n",
      "[43]\ttrain-auc:0.999864\teval-auc:0.995448\n",
      "[44]\ttrain-auc:0.999896\teval-auc:0.995654\n",
      "[45]\ttrain-auc:0.999901\teval-auc:0.995767\n",
      "[46]\ttrain-auc:0.999909\teval-auc:0.995853\n",
      "[47]\ttrain-auc:0.999922\teval-auc:0.996116\n",
      "[48]\ttrain-auc:0.99993\teval-auc:0.996526\n",
      "[49]\ttrain-auc:0.999935\teval-auc:0.996693\n",
      "[50]\ttrain-auc:0.999941\teval-auc:0.99664\n",
      "[51]\ttrain-auc:0.999967\teval-auc:0.996724\n",
      "[52]\ttrain-auc:0.999971\teval-auc:0.996697\n",
      "[53]\ttrain-auc:0.999972\teval-auc:0.996725\n",
      "[54]\ttrain-auc:0.999973\teval-auc:0.996769\n",
      "[55]\ttrain-auc:0.999973\teval-auc:0.996793\n",
      "[56]\ttrain-auc:0.999984\teval-auc:0.996887\n",
      "[57]\ttrain-auc:0.999996\teval-auc:0.996961\n",
      "[58]\ttrain-auc:0.999997\teval-auc:0.996993\n",
      "[59]\ttrain-auc:0.999998\teval-auc:0.99693\n",
      "[60]\ttrain-auc:0.999998\teval-auc:0.997026\n",
      "[61]\ttrain-auc:0.999999\teval-auc:0.996924\n",
      "[62]\ttrain-auc:0.999999\teval-auc:0.99694\n",
      "[63]\ttrain-auc:0.999999\teval-auc:0.996939\n",
      "[64]\ttrain-auc:1\teval-auc:0.997057\n",
      "[65]\ttrain-auc:1\teval-auc:0.997395\n",
      "[66]\ttrain-auc:1\teval-auc:0.997471\n",
      "[67]\ttrain-auc:1\teval-auc:0.997534\n",
      "[68]\ttrain-auc:1\teval-auc:0.997554\n",
      "[69]\ttrain-auc:1\teval-auc:0.997621\n",
      "[70]\ttrain-auc:1\teval-auc:0.997612\n",
      "[71]\ttrain-auc:1\teval-auc:0.997642\n",
      "[72]\ttrain-auc:1\teval-auc:0.997638\n",
      "[73]\ttrain-auc:1\teval-auc:0.997577\n",
      "[74]\ttrain-auc:1\teval-auc:0.997672\n",
      "[75]\ttrain-auc:1\teval-auc:0.997645\n",
      "[76]\ttrain-auc:1\teval-auc:0.997644\n",
      "[77]\ttrain-auc:1\teval-auc:0.997686\n",
      "[78]\ttrain-auc:1\teval-auc:0.99765\n",
      "[79]\ttrain-auc:1\teval-auc:0.997711\n",
      "[80]\ttrain-auc:1\teval-auc:0.997708\n",
      "[81]\ttrain-auc:1\teval-auc:0.997669\n",
      "[82]\ttrain-auc:1\teval-auc:0.99766\n",
      "[83]\ttrain-auc:1\teval-auc:0.997688\n",
      "[84]\ttrain-auc:1\teval-auc:0.997711\n",
      "[85]\ttrain-auc:1\teval-auc:0.997684\n",
      "[86]\ttrain-auc:1\teval-auc:0.997712\n",
      "[87]\ttrain-auc:1\teval-auc:0.997804\n",
      "[88]\ttrain-auc:1\teval-auc:0.99783\n",
      "[89]\ttrain-auc:1\teval-auc:0.997849\n",
      "[90]\ttrain-auc:1\teval-auc:0.997887\n",
      "[91]\ttrain-auc:1\teval-auc:0.997833\n",
      "[92]\ttrain-auc:1\teval-auc:0.997859\n",
      "[93]\ttrain-auc:1\teval-auc:0.997852\n",
      "[94]\ttrain-auc:1\teval-auc:0.997847\n",
      "[95]\ttrain-auc:1\teval-auc:0.997801\n",
      "[96]\ttrain-auc:1\teval-auc:0.997893\n",
      "[97]\ttrain-auc:1\teval-auc:0.99789\n",
      "[98]\ttrain-auc:1\teval-auc:0.99793\n",
      "[99]\ttrain-auc:1\teval-auc:0.997936\n",
      "[100]\ttrain-auc:1\teval-auc:0.997981\n",
      "[101]\ttrain-auc:1\teval-auc:0.997905\n",
      "[102]\ttrain-auc:1\teval-auc:0.997901\n",
      "[103]\ttrain-auc:1\teval-auc:0.997954\n",
      "[104]\ttrain-auc:1\teval-auc:0.997907\n",
      "[105]\ttrain-auc:1\teval-auc:0.997921\n",
      "[106]\ttrain-auc:1\teval-auc:0.997932\n",
      "[107]\ttrain-auc:1\teval-auc:0.997876\n",
      "[108]\ttrain-auc:1\teval-auc:0.998022\n",
      "[109]\ttrain-auc:1\teval-auc:0.998028\n",
      "[110]\ttrain-auc:1\teval-auc:0.998041\n",
      "[111]\ttrain-auc:1\teval-auc:0.998037\n",
      "[112]\ttrain-auc:1\teval-auc:0.998066\n",
      "[113]\ttrain-auc:1\teval-auc:0.998095\n",
      "[114]\ttrain-auc:1\teval-auc:0.998078\n",
      "[115]\ttrain-auc:1\teval-auc:0.99811\n",
      "[116]\ttrain-auc:1\teval-auc:0.998105\n",
      "[117]\ttrain-auc:1\teval-auc:0.998142\n",
      "[118]\ttrain-auc:1\teval-auc:0.998146\n",
      "[119]\ttrain-auc:1\teval-auc:0.998171\n",
      "[120]\ttrain-auc:1\teval-auc:0.998183\n",
      "[121]\ttrain-auc:1\teval-auc:0.998173\n",
      "[122]\ttrain-auc:1\teval-auc:0.998213\n",
      "[123]\ttrain-auc:1\teval-auc:0.998188\n",
      "[124]\ttrain-auc:1\teval-auc:0.998196\n",
      "[125]\ttrain-auc:1\teval-auc:0.998215\n",
      "[126]\ttrain-auc:1\teval-auc:0.998214\n",
      "[127]\ttrain-auc:1\teval-auc:0.998154\n",
      "[128]\ttrain-auc:1\teval-auc:0.998197\n",
      "[129]\ttrain-auc:1\teval-auc:0.998195\n",
      "[130]\ttrain-auc:1\teval-auc:0.998189\n",
      "[131]\ttrain-auc:1\teval-auc:0.998153\n",
      "[132]\ttrain-auc:1\teval-auc:0.998115\n",
      "[133]\ttrain-auc:1\teval-auc:0.998193\n",
      "[134]\ttrain-auc:1\teval-auc:0.998205\n",
      "[135]\ttrain-auc:1\teval-auc:0.998228\n",
      "[136]\ttrain-auc:1\teval-auc:0.998199\n",
      "[137]\ttrain-auc:1\teval-auc:0.998219\n",
      "[138]\ttrain-auc:1\teval-auc:0.998235\n",
      "[139]\ttrain-auc:1\teval-auc:0.998221\n",
      "[140]\ttrain-auc:1\teval-auc:0.99821\n",
      "[141]\ttrain-auc:1\teval-auc:0.998228\n",
      "[142]\ttrain-auc:1\teval-auc:0.998229\n",
      "[143]\ttrain-auc:1\teval-auc:0.998249\n",
      "[144]\ttrain-auc:1\teval-auc:0.998269\n",
      "[145]\ttrain-auc:1\teval-auc:0.998307\n",
      "[146]\ttrain-auc:1\teval-auc:0.998333\n",
      "[147]\ttrain-auc:1\teval-auc:0.998338\n",
      "[148]\ttrain-auc:1\teval-auc:0.998338\n",
      "[149]\ttrain-auc:1\teval-auc:0.99836\n",
      "[150]\ttrain-auc:1\teval-auc:0.998346\n",
      "[151]\ttrain-auc:1\teval-auc:0.998357\n",
      "[152]\ttrain-auc:1\teval-auc:0.998365\n",
      "[153]\ttrain-auc:1\teval-auc:0.998392\n",
      "[154]\ttrain-auc:1\teval-auc:0.998365\n",
      "[155]\ttrain-auc:1\teval-auc:0.998355\n",
      "[156]\ttrain-auc:1\teval-auc:0.998341\n",
      "[157]\ttrain-auc:1\teval-auc:0.998296\n",
      "[158]\ttrain-auc:1\teval-auc:0.99832\n",
      "[159]\ttrain-auc:1\teval-auc:0.998406\n",
      "[160]\ttrain-auc:1\teval-auc:0.998436\n",
      "[161]\ttrain-auc:1\teval-auc:0.998443\n",
      "[162]\ttrain-auc:1\teval-auc:0.998479\n",
      "[163]\ttrain-auc:1\teval-auc:0.998483\n",
      "[164]\ttrain-auc:1\teval-auc:0.99849\n",
      "[165]\ttrain-auc:1\teval-auc:0.998467\n",
      "[166]\ttrain-auc:1\teval-auc:0.99845\n",
      "[167]\ttrain-auc:1\teval-auc:0.998413\n",
      "[168]\ttrain-auc:1\teval-auc:0.998416\n",
      "[169]\ttrain-auc:1\teval-auc:0.998417\n",
      "[170]\ttrain-auc:1\teval-auc:0.998442\n",
      "[171]\ttrain-auc:1\teval-auc:0.998402\n",
      "[172]\ttrain-auc:1\teval-auc:0.998434\n",
      "[173]\ttrain-auc:1\teval-auc:0.998455\n",
      "[174]\ttrain-auc:1\teval-auc:0.998482\n",
      "[175]\ttrain-auc:1\teval-auc:0.998557\n",
      "[176]\ttrain-auc:1\teval-auc:0.998555\n",
      "[177]\ttrain-auc:1\teval-auc:0.998558\n",
      "[178]\ttrain-auc:1\teval-auc:0.998547\n",
      "[179]\ttrain-auc:1\teval-auc:0.998519\n",
      "[180]\ttrain-auc:1\teval-auc:0.998539\n",
      "[181]\ttrain-auc:1\teval-auc:0.998538\n",
      "[182]\ttrain-auc:1\teval-auc:0.998562\n",
      "[183]\ttrain-auc:1\teval-auc:0.998552\n",
      "[184]\ttrain-auc:1\teval-auc:0.998601\n",
      "[185]\ttrain-auc:1\teval-auc:0.998627\n",
      "[186]\ttrain-auc:1\teval-auc:0.998557\n",
      "[187]\ttrain-auc:1\teval-auc:0.998592\n",
      "[188]\ttrain-auc:1\teval-auc:0.998572\n",
      "[189]\ttrain-auc:1\teval-auc:0.998559\n",
      "[190]\ttrain-auc:1\teval-auc:0.998552\n",
      "[191]\ttrain-auc:1\teval-auc:0.998596\n",
      "[192]\ttrain-auc:1\teval-auc:0.998582\n",
      "[193]\ttrain-auc:1\teval-auc:0.998569\n",
      "[194]\ttrain-auc:1\teval-auc:0.998599\n",
      "[195]\ttrain-auc:1\teval-auc:0.998638\n",
      "[196]\ttrain-auc:1\teval-auc:0.998622\n",
      "[197]\ttrain-auc:1\teval-auc:0.998638\n",
      "[198]\ttrain-auc:1\teval-auc:0.998646\n",
      "[199]\ttrain-auc:1\teval-auc:0.998627\n",
      "[200]\ttrain-auc:1\teval-auc:0.998635\n",
      "[201]\ttrain-auc:1\teval-auc:0.998623\n",
      "[202]\ttrain-auc:1\teval-auc:0.998664\n",
      "[203]\ttrain-auc:1\teval-auc:0.99868\n",
      "[204]\ttrain-auc:1\teval-auc:0.998724\n",
      "[205]\ttrain-auc:1\teval-auc:0.998684\n",
      "[206]\ttrain-auc:1\teval-auc:0.998688\n",
      "[207]\ttrain-auc:1\teval-auc:0.998688\n",
      "[208]\ttrain-auc:1\teval-auc:0.998694\n",
      "[209]\ttrain-auc:1\teval-auc:0.998656\n",
      "[210]\ttrain-auc:1\teval-auc:0.998661\n",
      "[211]\ttrain-auc:1\teval-auc:0.998642\n",
      "[212]\ttrain-auc:1\teval-auc:0.998623\n",
      "[213]\ttrain-auc:1\teval-auc:0.998626\n",
      "[214]\ttrain-auc:1\teval-auc:0.998645\n",
      "[215]\ttrain-auc:1\teval-auc:0.998655\n",
      "[216]\ttrain-auc:1\teval-auc:0.998604\n",
      "[217]\ttrain-auc:1\teval-auc:0.998621\n",
      "[218]\ttrain-auc:1\teval-auc:0.998641\n",
      "[219]\ttrain-auc:1\teval-auc:0.998614\n",
      "[220]\ttrain-auc:1\teval-auc:0.998628\n",
      "[221]\ttrain-auc:1\teval-auc:0.998654\n",
      "[222]\ttrain-auc:1\teval-auc:0.998615\n",
      "[223]\ttrain-auc:1\teval-auc:0.998641\n",
      "[224]\ttrain-auc:1\teval-auc:0.998651\n",
      "Stopping. Best iteration:\n",
      "[204]\ttrain-auc:1\teval-auc:0.998724\n",
      "\n",
      "Validating...\n",
      "area under the precision-recall curve: 0.997705\n",
      "precision score: 0.983601\n",
      "recall score: 0.996552\n",
      "Importance array:  [('DISCOUNT_AMT', 758), ('std_pago_retail_mensual', 745), ('avg_pago_retail_mensual', 705), ('avg_uso_tarjeta', 701), ('SPEND_AMT', 692), ('EDAD', 653), ('max_pago_retail_mensual', 638), ('GSE_C3', 117), ('GENDER', 89), ('EMAIL_CONTACTABILITY', 82), ('CIVIL_STATUS_CASADO', 75), ('CIVIL_STATUS_SOLTERO', 56), ('GSE_D', 53), ('GSE_E', 43), ('CIVIL_STATUS_DIVORCIADO', 29), ('GSE_C2', 17), ('MOBILE_CONTACTABILITY', 15), ('GSE_ABC1', 11), ('CIVIL_STATUS_VIUDO', 3)]\n",
      "Predict test set... \n",
      "area under the precision-recall curve test set: 0.997705\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsUAAAGMCAYAAAAydN8hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f78cab1c2e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiUAAAGMCAYAAAAFltJQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nzixXbrlyhDcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7830a6e240>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f78001d3780>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAPUAAAFoCAYAAACPPr1MAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n5v8Bltc+fojhGOgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f7830a6bb38>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time: 6.25 minutes\n",
      "0:06:15.131833\n"
     ]
    }
   ],
   "source": [
    "features = X_train.columns.values.tolist()\n",
    "print(features)\n",
    "\n",
    "print(\"Building model.. \",dt.datetime.now()-start_time)\n",
    "#X_train, X_valid, y_train, y_valid, features, random_state=0\n",
    "preds, imp, num_boost_rounds = run_single(X_train, X_test, Y_train, Y_test, features, 100)\n",
    "\n",
    " \n",
    "print(dt.datetime.now()-start_time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.5 (Experimental) with Spark 2.1",
   "language": "python",
   "name": "python3-spark21"
  },
  "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": 1
}