uplift_model_draft.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "#import time\n",
    "#from datetime import datetime, timedelta\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "sns.set_style(\"whitegrid\")\n",
    "color = sns.color_palette('Set3')\n",
    "\n",
    "from sklift.models import SoloModel, ClassTransformation, TwoModels\n",
    "from catboost import CatBoostClassifier\n",
    "import lightgbm as lgb\n",
    "from sklearn.model_selection import train_test_split, StratifiedKFold\n",
    "from sklearn.metrics import roc_auc_score, precision_score, recall_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load data\n",
    "df = pd.read_csv('info_train.csv')\n",
    "\n",
    "# remove unwanted columns\n",
    "unwantedColumns = ['UNIQUE_ID', 'Unnamed: 0']\n",
    "df.drop(unwantedColumns, axis=1, inplace=True)\n",
    "\n",
    "# set up variables\n",
    "treatmentVariable = 'TREATMENT'\n",
    "targetVariable = 'PURCHASE'\n",
    "predictors = [col for col in df.columns if col not in [treatmentVariable, targetVariable]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>TREATMENT</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PURCHASE</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.4045</td>\n",
       "      <td>0.3959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0983</td>\n",
       "      <td>0.1013</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "TREATMENT       0       1\n",
       "PURCHASE                 \n",
       "0          0.4045  0.3959\n",
       "1          0.0983  0.1013"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check the distribution of both treatment/control and purchase/non-purchase\n",
    "pd.crosstab(df[targetVariable], df[treatmentVariable])/len(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split the dataset into a training and testing set. We have to do a stratified split on both the treatment and target variables because we want to keep the same ratios of treatment/control and purchase/non-purchase."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "SEED = 2\n",
    "x_train, x_test, y_train, y_test = train_test_split(df, df[targetVariable], test_size=1/3, \n",
    "                                                    stratify=df[[treatmentVariable, targetVariable]], \n",
    "                                                    random_state=SEED)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Check the ratio's treatment/control and responder/non-responder remain correct. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>TREATMENT</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PURCHASE</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.40459</td>\n",
       "      <td>0.39589</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.09826</td>\n",
       "      <td>0.10126</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "TREATMENT        0        1\n",
       "PURCHASE                   \n",
       "0          0.40459  0.39589\n",
       "1          0.09826  0.10126"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(y_train, x_train[treatmentVariable])/len(x_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>TREATMENT</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PURCHASE</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.404319</td>\n",
       "      <td>0.395921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.098380</td>\n",
       "      <td>0.101380</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "TREATMENT         0         1\n",
       "PURCHASE                     \n",
       "0          0.404319  0.395921\n",
       "1          0.098380  0.101380"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(y_test, x_test[treatmentVariable])/len(x_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Modeling Techniques    \n",
    "- Two Model Approach (also known as the Naive Approach)\n",
    "- Dummy Treatment Approach\n",
    "- Treatment Interaction Approach    \n",
    "\n",
    "Reference:    \n",
    "https://readthedocs.org/projects/scikit-uplift/downloads/pdf/latest/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Two Model Approach    \n",
    "The idea is to built two separate models, one for the treatment group and one for the control group. Then calculating the uplift as the difference of the predicted probabilities of the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define models\n",
    "treatment_model = CatBoostClassifier(iterations=200, thread_count=3,\n",
    "                                     random_state=SEED, silent=True)\n",
    "control_model = CatBoostClassifier(iterations=200, thread_count=3,\n",
    "                                   random_state=SEED, silent=True)\n",
    "\n",
    "# define approach\n",
    "tm = TwoModels(treatment_model, control_model, method='vanilla')\n",
    "# fit model\n",
    "tm = tm.fit(x_train[predictors], y_train, x_train[treatmentVariable])\n",
    "# predict uplift\n",
    "uplift_preds = tm.predict(x_test[predictors])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklift.metrics import uplift_at_k, uplift_auc_score, qini_auc_score, weighted_average_uplift, uplift_by_percentile\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "qini_auc:  0.09337401278881416\n",
      "uplift_auc:  0.06273811574190728\n",
      "weighted average uplift:  0.005017842745900688\n"
     ]
    }
   ],
   "source": [
    "# Area Under Qini Curve\n",
    "tm_qini_auc = qini_auc_score(y_true=y_test, uplift=uplift_preds, treatment=x_test[treatmentVariable])\n",
    "\n",
    "# Area Under Uplift Curve\n",
    "tm_uplift_auc = uplift_auc_score(y_true=y_test, uplift=uplift_preds, treatment=x_test[treatmentVariable])\n",
    "\n",
    "# Weighted average uplift\n",
    "tm_wau = weighted_average_uplift(y_true=y_test, uplift=uplift_preds,  treatment=x_test[treatmentVariable])\n",
    "\n",
    "print('qini_auc: ', tm_qini_auc)\n",
    "print('uplift_auc: ', tm_uplift_auc)\n",
    "print('weighted average uplift: ', tm_wau)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([<matplotlib.axes._subplots.AxesSubplot object at 0x120f190b8>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x11df492b0>],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualize result\n",
    "from sklift.viz import plot_qini_curve, plot_uplift_curve, plot_uplift_by_percentile\n",
    "\n",
    "plot_qini_curve(y_true=y_test, uplift=uplift_preds, treatment=x_test[treatmentVariable])\n",
    "plot_uplift_curve(y_true=y_test, uplift=uplift_preds, treatment=x_test[treatmentVariable])\n",
    "plot_uplift_by_percentile(y_true=y_test, uplift=uplift_preds, treatment=x_test[treatmentVariable], \n",
    "                          kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "uplift by decile: \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_treatment</th>\n",
       "      <th>n_control</th>\n",
       "      <th>response_rate_treatment</th>\n",
       "      <th>response_rate_control</th>\n",
       "      <th>uplift</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>percentile</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10.0</th>\n",
       "      <td>156</td>\n",
       "      <td>178</td>\n",
       "      <td>0.423077</td>\n",
       "      <td>0.303371</td>\n",
       "      <td>0.119706</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20.0</th>\n",
       "      <td>168</td>\n",
       "      <td>166</td>\n",
       "      <td>0.303571</td>\n",
       "      <td>0.253012</td>\n",
       "      <td>0.050559</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30.0</th>\n",
       "      <td>182</td>\n",
       "      <td>152</td>\n",
       "      <td>0.247253</td>\n",
       "      <td>0.197368</td>\n",
       "      <td>0.049884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.0</th>\n",
       "      <td>156</td>\n",
       "      <td>178</td>\n",
       "      <td>0.173077</td>\n",
       "      <td>0.117978</td>\n",
       "      <td>0.055099</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.0</th>\n",
       "      <td>165</td>\n",
       "      <td>168</td>\n",
       "      <td>0.078788</td>\n",
       "      <td>0.065476</td>\n",
       "      <td>0.013312</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60.0</th>\n",
       "      <td>163</td>\n",
       "      <td>170</td>\n",
       "      <td>0.012270</td>\n",
       "      <td>0.076471</td>\n",
       "      <td>-0.064201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70.0</th>\n",
       "      <td>173</td>\n",
       "      <td>160</td>\n",
       "      <td>0.069364</td>\n",
       "      <td>0.050000</td>\n",
       "      <td>0.019364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80.0</th>\n",
       "      <td>153</td>\n",
       "      <td>180</td>\n",
       "      <td>0.183007</td>\n",
       "      <td>0.127778</td>\n",
       "      <td>0.055229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90.0</th>\n",
       "      <td>169</td>\n",
       "      <td>164</td>\n",
       "      <td>0.242604</td>\n",
       "      <td>0.262195</td>\n",
       "      <td>-0.019592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.0</th>\n",
       "      <td>173</td>\n",
       "      <td>160</td>\n",
       "      <td>0.306358</td>\n",
       "      <td>0.518750</td>\n",
       "      <td>-0.212392</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            n_treatment  n_control  response_rate_treatment  \\\n",
       "percentile                                                    \n",
       "10.0                156        178                 0.423077   \n",
       "20.0                168        166                 0.303571   \n",
       "30.0                182        152                 0.247253   \n",
       "40.0                156        178                 0.173077   \n",
       "50.0                165        168                 0.078788   \n",
       "60.0                163        170                 0.012270   \n",
       "70.0                173        160                 0.069364   \n",
       "80.0                153        180                 0.183007   \n",
       "90.0                169        164                 0.242604   \n",
       "100.0               173        160                 0.306358   \n",
       "\n",
       "            response_rate_control    uplift  \n",
       "percentile                                   \n",
       "10.0                     0.303371  0.119706  \n",
       "20.0                     0.253012  0.050559  \n",
       "30.0                     0.197368  0.049884  \n",
       "40.0                     0.117978  0.055099  \n",
       "50.0                     0.065476  0.013312  \n",
       "60.0                     0.076471 -0.064201  \n",
       "70.0                     0.050000  0.019364  \n",
       "80.0                     0.127778  0.055229  \n",
       "90.0                     0.262195 -0.019592  \n",
       "100.0                    0.518750 -0.212392  "
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tm_uplift_decile = uplift_by_percentile(y_true=y_test, uplift=uplift_preds, \n",
    "                                        treatment=x_test[treatmentVariable], \n",
    "                                        strategy='overall', bins=10, std=False, total=False)\n",
    "print('uplift by decile: ') \n",
    "tm_uplift_decile"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dummy Treatment Approach    \n",
    "Single model with treatment as a binary feature.    \n",
    "After the model is trained, at the scoring time it is going to be applied twice: with the treatment flag equals 1 and with the treatment flag equals 0. Subtracting these model’s outcomes for each test sample, we will get an estimate of the uplift."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "ct = CatBoostClassifier(iterations=200, random_state=SEED, silent=True)\n",
    "# define approach\n",
    "sm = SoloModel(ct, method='dummy')\n",
    "# fit the model\n",
    "sm = sm.fit(x_train[predictors], y_train, \n",
    "            x_train[treatmentVariable])\n",
    "# predict uplift\n",
    "uplift_sm = sm.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "qini_auc:  0.11069689844566122\n",
      "uplift_auc:  0.07367317217102896\n",
      "weighted average uplift:  0.0086107653095905\n"
     ]
    }
   ],
   "source": [
    "# Area Under Qini Curve\n",
    "qini_auc = qini_auc_score(y_true=y_test, uplift=uplift_sm, treatment=x_test[treatmentVariable])\n",
    "\n",
    "# Area Under Uplift Curve\n",
    "uplift_auc = uplift_auc_score(y_true=y_test, uplift=uplift_sm, treatment=x_test[treatmentVariable])\n",
    "\n",
    "# Weighted average uplift\n",
    "wau = weighted_average_uplift(y_true=y_test, uplift=uplift_sm,  treatment=x_test[treatmentVariable])\n",
    "\n",
    "print('qini_auc: ', qini_auc)\n",
    "print('uplift_auc: ', uplift_auc)\n",
    "print('weighted average uplift: ', wau)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([<matplotlib.axes._subplots.AxesSubplot object at 0x11d5742e8>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x11db993c8>],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_qini_curve(y_true=y_test, uplift=uplift_sm, treatment=x_test[treatmentVariable])\n",
    "plot_uplift_curve(y_true=y_test, uplift=uplift_sm, treatment=x_test[treatmentVariable])\n",
    "plot_uplift_by_percentile(y_true=y_test, uplift=uplift_sm, treatment=x_test[treatmentVariable], \n",
    "                          kind='bar')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "uplift by decile: \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_treatment</th>\n",
       "      <th>n_control</th>\n",
       "      <th>response_rate_treatment</th>\n",
       "      <th>response_rate_control</th>\n",
       "      <th>uplift</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>percentile</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10.0</th>\n",
       "      <td>167</td>\n",
       "      <td>167</td>\n",
       "      <td>0.401198</td>\n",
       "      <td>0.281437</td>\n",
       "      <td>0.119760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20.0</th>\n",
       "      <td>157</td>\n",
       "      <td>177</td>\n",
       "      <td>0.318471</td>\n",
       "      <td>0.186441</td>\n",
       "      <td>0.132031</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30.0</th>\n",
       "      <td>179</td>\n",
       "      <td>155</td>\n",
       "      <td>0.212291</td>\n",
       "      <td>0.238710</td>\n",
       "      <td>-0.026419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.0</th>\n",
       "      <td>171</td>\n",
       "      <td>163</td>\n",
       "      <td>0.163743</td>\n",
       "      <td>0.122699</td>\n",
       "      <td>0.041043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.0</th>\n",
       "      <td>163</td>\n",
       "      <td>170</td>\n",
       "      <td>0.128834</td>\n",
       "      <td>0.135294</td>\n",
       "      <td>-0.006460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60.0</th>\n",
       "      <td>162</td>\n",
       "      <td>171</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>0.058480</td>\n",
       "      <td>0.052632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70.0</th>\n",
       "      <td>167</td>\n",
       "      <td>166</td>\n",
       "      <td>0.029940</td>\n",
       "      <td>0.054217</td>\n",
       "      <td>-0.024277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80.0</th>\n",
       "      <td>167</td>\n",
       "      <td>166</td>\n",
       "      <td>0.053892</td>\n",
       "      <td>0.060241</td>\n",
       "      <td>-0.006349</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90.0</th>\n",
       "      <td>161</td>\n",
       "      <td>172</td>\n",
       "      <td>0.329193</td>\n",
       "      <td>0.302326</td>\n",
       "      <td>0.026867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.0</th>\n",
       "      <td>164</td>\n",
       "      <td>169</td>\n",
       "      <td>0.298780</td>\n",
       "      <td>0.514793</td>\n",
       "      <td>-0.216012</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            n_treatment  n_control  response_rate_treatment  \\\n",
       "percentile                                                    \n",
       "10.0                167        167                 0.401198   \n",
       "20.0                157        177                 0.318471   \n",
       "30.0                179        155                 0.212291   \n",
       "40.0                171        163                 0.163743   \n",
       "50.0                163        170                 0.128834   \n",
       "60.0                162        171                 0.111111   \n",
       "70.0                167        166                 0.029940   \n",
       "80.0                167        166                 0.053892   \n",
       "90.0                161        172                 0.329193   \n",
       "100.0               164        169                 0.298780   \n",
       "\n",
       "            response_rate_control    uplift  \n",
       "percentile                                   \n",
       "10.0                     0.281437  0.119760  \n",
       "20.0                     0.186441  0.132031  \n",
       "30.0                     0.238710 -0.026419  \n",
       "40.0                     0.122699  0.041043  \n",
       "50.0                     0.135294 -0.006460  \n",
       "60.0                     0.058480  0.052632  \n",
       "70.0                     0.054217 -0.024277  \n",
       "80.0                     0.060241 -0.006349  \n",
       "90.0                     0.302326  0.026867  \n",
       "100.0                    0.514793 -0.216012  "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uplift_decile = uplift_by_percentile(y_true=y_test, uplift=uplift_sm, \n",
    "                                        treatment=x_test[treatmentVariable], \n",
    "                                        strategy='overall', bins=10, std=False, total=False)\n",
    "print('uplift by decile: ') \n",
    "uplift_decile"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Treatment Interaction Approach\n",
    "Single model including treatment interactions.    \n",
    "Add the product of each feature and the treatment flag as new features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "ct2 = CatBoostClassifier(iterations=200, random_state=SEED, silent=True)\n",
    "# define approach\n",
    "sm2 = SoloModel(ct2, method='treatment_interaction')\n",
    "# fit the model\n",
    "sm2 = sm2.fit(x_train[predictors], y_train, \n",
    "            x_train[treatmentVariable])\n",
    "# predict uplift\n",
    "uplift_sm2 = sm2.predict(x_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "qini_auc:  0.08718699711821329\n",
      "uplift_auc:  0.058172254027845746\n",
      "weighted average uplift:  0.006847364365518806\n"
     ]
    }
   ],
   "source": [
    "# Area Under Qini Curve\n",
    "qini_auc = qini_auc_score(y_true=y_test, uplift=uplift_sm2, treatment=x_test[treatmentVariable])\n",
    "\n",
    "# Area Under Uplift Curve\n",
    "uplift_auc = uplift_auc_score(y_true=y_test, uplift=uplift_sm2, treatment=x_test[treatmentVariable])\n",
    "\n",
    "# Weighted average uplift\n",
    "wau = weighted_average_uplift(y_true=y_test, uplift=uplift_sm2,  treatment=x_test[treatmentVariable])\n",
    "\n",
    "#print('uplift_at_k: ', uplift_at_k)\n",
    "print('qini_auc: ', qini_auc)\n",
    "print('uplift_auc: ', uplift_auc)\n",
    "print('weighted average uplift: ', wau)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([<matplotlib.axes._subplots.AxesSubplot object at 0x11dc2f160>,\n",
       "       <matplotlib.axes._subplots.AxesSubplot object at 0x11dc67710>],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_qini_curve(y_true=y_test, uplift=uplift_sm2, treatment=x_test[treatmentVariable])\n",
    "plot_uplift_curve(y_true=y_test, uplift=uplift_sm2, treatment=x_test[treatmentVariable])\n",
    "plot_uplift_by_percentile(y_true=y_test, uplift=uplift_sm2, treatment=x_test[treatmentVariable], \n",
    "                          kind='bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "uplift by decile: \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>n_treatment</th>\n",
       "      <th>n_control</th>\n",
       "      <th>response_rate_treatment</th>\n",
       "      <th>response_rate_control</th>\n",
       "      <th>uplift</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>percentile</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10.0</th>\n",
       "      <td>168</td>\n",
       "      <td>166</td>\n",
       "      <td>0.392857</td>\n",
       "      <td>0.355422</td>\n",
       "      <td>0.037435</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20.0</th>\n",
       "      <td>167</td>\n",
       "      <td>167</td>\n",
       "      <td>0.389222</td>\n",
       "      <td>0.287425</td>\n",
       "      <td>0.101796</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30.0</th>\n",
       "      <td>166</td>\n",
       "      <td>168</td>\n",
       "      <td>0.265060</td>\n",
       "      <td>0.226190</td>\n",
       "      <td>0.038870</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40.0</th>\n",
       "      <td>168</td>\n",
       "      <td>166</td>\n",
       "      <td>0.214286</td>\n",
       "      <td>0.132530</td>\n",
       "      <td>0.081756</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50.0</th>\n",
       "      <td>152</td>\n",
       "      <td>181</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>0.060773</td>\n",
       "      <td>-0.008142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60.0</th>\n",
       "      <td>170</td>\n",
       "      <td>163</td>\n",
       "      <td>0.011765</td>\n",
       "      <td>0.030675</td>\n",
       "      <td>-0.018910</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70.0</th>\n",
       "      <td>170</td>\n",
       "      <td>163</td>\n",
       "      <td>0.023529</td>\n",
       "      <td>0.079755</td>\n",
       "      <td>-0.056225</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80.0</th>\n",
       "      <td>159</td>\n",
       "      <td>174</td>\n",
       "      <td>0.150943</td>\n",
       "      <td>0.068966</td>\n",
       "      <td>0.081978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>90.0</th>\n",
       "      <td>178</td>\n",
       "      <td>155</td>\n",
       "      <td>0.219101</td>\n",
       "      <td>0.264516</td>\n",
       "      <td>-0.045415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100.0</th>\n",
       "      <td>160</td>\n",
       "      <td>173</td>\n",
       "      <td>0.312500</td>\n",
       "      <td>0.456647</td>\n",
       "      <td>-0.144147</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            n_treatment  n_control  response_rate_treatment  \\\n",
       "percentile                                                    \n",
       "10.0                168        166                 0.392857   \n",
       "20.0                167        167                 0.389222   \n",
       "30.0                166        168                 0.265060   \n",
       "40.0                168        166                 0.214286   \n",
       "50.0                152        181                 0.052632   \n",
       "60.0                170        163                 0.011765   \n",
       "70.0                170        163                 0.023529   \n",
       "80.0                159        174                 0.150943   \n",
       "90.0                178        155                 0.219101   \n",
       "100.0               160        173                 0.312500   \n",
       "\n",
       "            response_rate_control    uplift  \n",
       "percentile                                   \n",
       "10.0                     0.355422  0.037435  \n",
       "20.0                     0.287425  0.101796  \n",
       "30.0                     0.226190  0.038870  \n",
       "40.0                     0.132530  0.081756  \n",
       "50.0                     0.060773 -0.008142  \n",
       "60.0                     0.030675 -0.018910  \n",
       "70.0                     0.079755 -0.056225  \n",
       "80.0                     0.068966  0.081978  \n",
       "90.0                     0.264516 -0.045415  \n",
       "100.0                    0.456647 -0.144147  "
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "uplift_decile = uplift_by_percentile(y_true=y_test, uplift=uplift_sm2, \n",
    "                                        treatment=x_test[treatmentVariable], \n",
    "                                        strategy='overall', bins=10, std=False, total=False)\n",
    "print('uplift by decile: ') \n",
    "uplift_decile"
   ]
  }
 ],
 "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": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}