Random_Forest.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.metrics import accuracy_score, recall_score, precision_score, precision_recall_curve, confusion_matrix, classification_report \n",
    "from sklearn.metrics import make_scorer, fbeta_score, auc\n",
    "from sklearn.metrics import roc_curve, roc_auc_score, f1_score, plot_confusion_matrix\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn .model_selection import RandomizedSearchCV\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "import pickle\n",
    "from imblearn.combine import SMOTETomek\n",
    "from collections import Counter  #for count the 1/0\n",
    "from imblearn.over_sampling import SMOTE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('Telecom.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>SeniorCitizen</th>\n",
       "      <th>Partner</th>\n",
       "      <th>Dependents</th>\n",
       "      <th>PaperlessBilling</th>\n",
       "      <th>MonthlyCharges</th>\n",
       "      <th>Churn</th>\n",
       "      <th>tenure_group</th>\n",
       "      <th>InternetService_Fiber optic</th>\n",
       "      <th>InternetService_No</th>\n",
       "      <th>...</th>\n",
       "      <th>OnlineBackup_Yes</th>\n",
       "      <th>DeviceProtection_Yes</th>\n",
       "      <th>TechSupport_Yes</th>\n",
       "      <th>StreamingTV_Yes</th>\n",
       "      <th>StreamingMovies_Yes</th>\n",
       "      <th>Contract_One year</th>\n",
       "      <th>Contract_Two year</th>\n",
       "      <th>PaymentMethod_Credit card (automatic)</th>\n",
       "      <th>PaymentMethod_Electronic check</th>\n",
       "      <th>PaymentMethod_Mailed check</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>29.85</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>56.95</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>53.85</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>42.30</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>70.70</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0  SeniorCitizen  Partner  Dependents  PaperlessBilling  \\\n",
       "0           0              0        1           0                 1   \n",
       "1           1              0        0           0                 0   \n",
       "2           2              0        0           0                 1   \n",
       "3           3              0        0           0                 0   \n",
       "4           4              0        0           0                 1   \n",
       "\n",
       "   MonthlyCharges  Churn  tenure_group  InternetService_Fiber optic  \\\n",
       "0           29.85      0             1                            0   \n",
       "1           56.95      0             3                            0   \n",
       "2           53.85      1             1                            0   \n",
       "3           42.30      0             4                            0   \n",
       "4           70.70      1             1                            1   \n",
       "\n",
       "   InternetService_No  ...  OnlineBackup_Yes  DeviceProtection_Yes  \\\n",
       "0                   0  ...                 1                     0   \n",
       "1                   0  ...                 0                     1   \n",
       "2                   0  ...                 1                     0   \n",
       "3                   0  ...                 0                     1   \n",
       "4                   0  ...                 0                     0   \n",
       "\n",
       "   TechSupport_Yes  StreamingTV_Yes  StreamingMovies_Yes  Contract_One year  \\\n",
       "0                0                0                    0                  0   \n",
       "1                0                0                    0                  1   \n",
       "2                0                0                    0                  0   \n",
       "3                1                0                    0                  1   \n",
       "4                0                0                    0                  0   \n",
       "\n",
       "   Contract_Two year  PaymentMethod_Credit card (automatic)  \\\n",
       "0                  0                                      0   \n",
       "1                  0                                      0   \n",
       "2                  0                                      0   \n",
       "3                  0                                      0   \n",
       "4                  0                                      0   \n",
       "\n",
       "   PaymentMethod_Electronic check  PaymentMethod_Mailed check  \n",
       "0                               1                           0  \n",
       "1                               0                           1  \n",
       "2                               0                           1  \n",
       "3                               0                           0  \n",
       "4                               1                           0  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#variables for training\n",
    "X = data.drop(['Churn', 'Unnamed: 0'], axis = 1)\n",
    "y = data['Churn']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train,X_test,y_train,y_test=train_test_split(X,y,train_size=0.8,random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from imblearn.combine import SMOTETomek\n",
    "from collections import Counter  #for count the 1/0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The num of classes before fit Counter({0: 4125, 1: 1500})\n",
      "the num of classes after fit Counter({0: 3976, 1: 3151})\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Soumit\\anaconda3\\envs\\C_churn\\lib\\site-packages\\imblearn\\utils\\_validation.py:635: FutureWarning: Pass sampling_strategy=0.8 as keyword args. From version 0.9 passing these as positional arguments will result in an error\n",
      "  warnings.warn(\"Pass {} as keyword args. From version 0.9 \"\n"
     ]
    }
   ],
   "source": [
    "os=SMOTETomek(0.8)\n",
    "X_train_Ov,y_train_Ov=os.fit_sample(X_train,y_train)\n",
    "print(\"The num of classes before fit {}\".format(Counter(y_train)))\n",
    "print(\"the num of classes after fit {}\".format(Counter(y_train_Ov)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 10 folds for each of 180 candidates, totalling 1800 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 8 concurrent workers.\n",
      "[Parallel(n_jobs=-1)]: Done  25 tasks      | elapsed:    7.2s\n",
      "[Parallel(n_jobs=-1)]: Done 146 tasks      | elapsed:   31.3s\n",
      "[Parallel(n_jobs=-1)]: Done 349 tasks      | elapsed:  1.1min\n",
      "[Parallel(n_jobs=-1)]: Done 632 tasks      | elapsed:  2.0min\n",
      "[Parallel(n_jobs=-1)]: Done 997 tasks      | elapsed:  3.1min\n",
      "[Parallel(n_jobs=-1)]: Done 1442 tasks      | elapsed:  4.3min\n",
      "[Parallel(n_jobs=-1)]: Done 1800 out of 1800 | elapsed:  5.3min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=10, estimator=RandomForestClassifier(), n_jobs=-1,\n",
       "             param_grid={'bootstrap': [True], 'max_depth': [10, 15, 20, 30],\n",
       "                         'max_features': ['auto'],\n",
       "                         'min_samples_leaf': [3, 4, 5],\n",
       "                         'min_samples_split': [8, 10, 12],\n",
       "                         'n_estimators': [20, 50, 100, 200, 300]},\n",
       "             verbose=2)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create the parameter grid based on the results of random search \n",
    "param_grid = {\n",
    "    'bootstrap': [True],\n",
    "    'max_depth': [10, 15, 20, 30],\n",
    "    'max_features': ['auto'],\n",
    "    'min_samples_leaf': [3, 4, 5],\n",
    "    'min_samples_split': [8, 10, 12],\n",
    "    'n_estimators': [20,50,100, 200, 300]\n",
    "}\n",
    "# Create a based model\n",
    "rf_C = RandomForestClassifier()\n",
    "# Instantiate the grid search model\n",
    "RF_model = GridSearchCV(estimator = rf_C, param_grid = param_grid, \n",
    "                          cv = 10, n_jobs = -1, verbose = 2)\n",
    "RF_model .fit(X_train_Ov,y_train_Ov)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bootstrap': True,\n",
       " 'max_depth': 30,\n",
       " 'max_features': 'auto',\n",
       " 'min_samples_leaf': 3,\n",
       " 'min_samples_split': 8,\n",
       " 'n_estimators': 300}"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "RF_model.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "rf_model=RandomForestClassifier(class_weight={0:1, 1:3}, max_depth = 10,random_state=1,\n",
    "                                  n_estimators=200, oob_score=True, min_samples_leaf=10, min_samples_split=8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(class_weight={0: 1, 1: 3}, max_depth=10,\n",
       "                       min_samples_leaf=10, min_samples_split=8,\n",
       "                       n_estimators=200, oob_score=True, random_state=1)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_model .fit(X_train_Ov,y_train_Ov)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[644 394]\n",
      " [ 43 326]]\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.94      0.62      0.75      1038\n",
      "           1       0.45      0.88      0.60       369\n",
      "\n",
      "    accuracy                           0.69      1407\n",
      "   macro avg       0.70      0.75      0.67      1407\n",
      "weighted avg       0.81      0.69      0.71      1407\n",
      "\n",
      "0.7519463633942699\n"
     ]
    }
   ],
   "source": [
    "y_pred=rf_model.predict(X_test)\n",
    "print(confusion_matrix(y_test,y_pred))\n",
    "print(classification_report(y_test,y_pred))\n",
    "print(roc_auc_score(y_test,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RF test roc-auc: 0.8371137950300506\n"
     ]
    }
   ],
   "source": [
    "y_prob = rf_model.predict_proba(X_test)\n",
    "print('RF test roc-auc: {}'.format(roc_auc_score(y_test, y_prob[:,1])))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, '2-class Precision-Recall curve')"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "from sklearn.metrics import plot_precision_recall_curve\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "disp = plot_precision_recall_curve(rf_model, X_test, y_test)\n",
    "disp.ax_.set_title('2-class Precision-Recall curve')\n",
    "                   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# calculate the fpr and tpr for all thresholds of the classification\n",
    "fpr, tpr, threshold = roc_curve(y_test, y_prob[:,1])\n",
    "roc_auc = auc(fpr, tpr)\n",
    "\n",
    "# method I: plt\n",
    "import matplotlib.pyplot as plt\n",
    "plt.title('Receiver Operating Characteristic')\n",
    "plt.plot(fpr, tpr, 'b', label = 'AUC = %0.2f' % roc_auc)\n",
    "plt.legend(loc = 'lower right')\n",
    "plt.plot([0, 1], [0, 1],'r--')\n",
    "plt.xlim([0, 1])\n",
    "plt.ylim([0, 1])\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Save model\n",
    "pickle.dump(rf_model, open('model.pkl','wb'))"
   ]
  }
 ],
 "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}