{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Customer retention - Thinkful final exam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.metrics import confusion_matrix\n",
    "sns.set_style('darkgrid')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Let's import the data and have a first look.\n",
    "\n",
    "We use the id column as our index, and try to answer some initial questions.\n",
    "\n",
    "- Are there class imbalances?\n",
    "- Are some features irrelevant?\n",
    "- Are there immediately obvious trends?\n",
    "- Are there problems with missing or NaN values?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>purch_amt</th>\n",
       "      <th>gender</th>\n",
       "      <th>card_on_file</th>\n",
       "      <th>age</th>\n",
       "      <th>days_since_last_purch</th>\n",
       "      <th>loyalty</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></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",
       "      <td>0</td>\n",
       "      <td>19.58</td>\n",
       "      <td>male</td>\n",
       "      <td>no</td>\n",
       "      <td>31.0</td>\n",
       "      <td>35.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>65.16</td>\n",
       "      <td>male</td>\n",
       "      <td>yes</td>\n",
       "      <td>23.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>40.60</td>\n",
       "      <td>female</td>\n",
       "      <td>no</td>\n",
       "      <td>36.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>38.01</td>\n",
       "      <td>male</td>\n",
       "      <td>yes</td>\n",
       "      <td>47.0</td>\n",
       "      <td>57.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>22.32</td>\n",
       "      <td>female</td>\n",
       "      <td>yes</td>\n",
       "      <td>5.0</td>\n",
       "      <td>39.0</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    purch_amt  gender card_on_file   age  days_since_last_purch  loyalty\n",
       "id                                                                      \n",
       "0       19.58    male           no  31.0                   35.0    False\n",
       "1       65.16    male          yes  23.0                   61.0    False\n",
       "2       40.60  female           no  36.0                   49.0    False\n",
       "3       38.01    male          yes  47.0                   57.0    False\n",
       "4       22.32  female          yes   5.0                   39.0    False"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('test.csv', index_col='id')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "yes    60087\n",
       "no     59913\n",
       "Name: card_on_file, dtype: int64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.card_on_file.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "male      60181\n",
       "female    59819\n",
       "Name: gender, dtype: int64"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.gender.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    100000\n",
       "True      20000\n",
       "Name: loyalty, dtype: int64"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loyalty.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### There is a clear class imbalance problem.\n",
    "\n",
    "Customer loyalty in this sample leans strongly in the false direction. We can hope to address this and evalute true model performance via a confusion matrix."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 120000 entries, 0 to 119999\n",
      "Data columns (total 6 columns):\n",
      "purch_amt                120000 non-null float64\n",
      "gender                   120000 non-null object\n",
      "card_on_file             120000 non-null object\n",
      "age                      120000 non-null float64\n",
      "days_since_last_purch    120000 non-null float64\n",
      "loyalty                  120000 non-null bool\n",
      "dtypes: bool(1), float64(3), object(2)\n",
      "memory usage: 5.6+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.join(pd.get_dummies(df.gender))\n",
    "df = df.join(pd.get_dummies(df.card_on_file, drop_first=True), rsuffix='card')\n",
    "df.drop(columns=['gender', 'card_on_file'], inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x1ece1aa6b88>"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 540x540 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df.drop(columns=['female', 'male', 'yes', 'loyalty']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Considering synthetic features.\n",
    "\n",
    "These distributions look healthy, and I can't justify turning them into synthetic features. We could do things like age above 35, or 50 days since last purchase - we might go down that road when there is a clear separation in the distribution."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1ece162c908>"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(df.corr())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### It appears that only a few features are affecting loyalty outcomes.\n",
    "\n",
    "The correlation map above is showing total neutrality between some features. Before putting something like this into production, we could run several models verifying that.\n",
    "\n",
    "#### Preliminary questions have been answered.\n",
    "\n",
    "Now we can try and set up some models. We will use a few classification algorithms, as the question we are trying to answer is a positive or negative loyalty outcome. I have selected a few of the standard models - logistic regression, random forests, naive bayes, and even the gradient boosting classifier later on."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(df.drop(columns='loyalty'), df.loyalty, test_size=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8694166666666666"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_log = LogisticRegression()\n",
    "clf_log.fit(X_train, y_train)\n",
    "clf_log.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\ensemble\\forest.py:245: FutureWarning: The default value of n_estimators will change from 10 in version 0.20 to 100 in 0.22.\n",
      "  \"10 in version 0.20 to 100 in 0.22.\", FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8452222222222222"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_rf = RandomForestClassifier()\n",
    "clf_rf.fit(X_train, y_train)\n",
    "clf_rf.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8716944444444444"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_gnb = GaussianNB()\n",
    "clf_gnb.fit(X_train, y_train)\n",
    "clf_gnb.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Maybe days since last purchase indicates leaving customers?\n",
    "df2 = df\n",
    "df2.days_since_last_purch = df.days_since_last_purch * - 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train2, X_test2, y_train2, y_test2 = train_test_split(df.drop(columns='loyalty'), df.loyalty, test_size=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8683055555555556"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_log = LogisticRegression()\n",
    "clf_log.fit(X_train2, y_train2)\n",
    "clf_log.score(X_test2, y_test2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### No significant difference between models.\n",
    "\n",
    "I chose the logistic regression model as this is tried and true - plus it performed well and *very fast.* We can run a grid search on it to see if any parameters help with performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "params = {'penalty': ['l1', 'l2'], 'C': [1, .1, .01, .001, .0001]}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "CV = GridSearchCV(clf_log, params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n",
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv='warn', error_score='raise-deprecating',\n",
       "             estimator=LogisticRegression(C=1.0, class_weight=None, dual=False,\n",
       "                                          fit_intercept=True,\n",
       "                                          intercept_scaling=1, l1_ratio=None,\n",
       "                                          max_iter=100, multi_class='warn',\n",
       "                                          n_jobs=None, penalty='l2',\n",
       "                                          random_state=None, solver='warn',\n",
       "                                          tol=0.0001, verbose=0,\n",
       "                                          warm_start=False),\n",
       "             iid='warn', n_jobs=None,\n",
       "             param_grid={'C': [1, 0.1, 0.01, 0.001, 0.0001],\n",
       "                         'penalty': ['l1', 'l2']},\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring=None, verbose=0)"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CV.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 0.001, 'penalty': 'l2'}"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CV.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf_log = LogisticRegression(penalty='l2', C=0.001)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=0.001, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='warn', n_jobs=None, penalty='l2',\n",
       "                   random_state=None, solver='warn', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_log.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8698055555555556"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_log.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = clf_log.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.96309643, 0.18341625],\n",
       "       [0.11948615, 0.40613599]])"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = confusion_matrix(y_test, y_pred)\n",
    "results / results.sum(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Results didn't change much.\n",
    "\n",
    "There are no immediately seen benefits when using the tuned hyper parameters.\n",
    "\n",
    "The normalized matrix indicates how well classifications are occurring as a percentage - how well true negatives perform are identified in the top left, and the bottom right indicates how well true positive labels are identified. This could be expected as we have pointed out the class imbalance issue, but we are seeing much better than random selection performance. True negatives are identified corrently more than 96 percent of the time, and true positives are correctly identified more than twice the percentage in which they are present.\n",
    "\n",
    "#### One more model - the GBC."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf_gb = GradientBoostingClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GradientBoostingClassifier(criterion='friedman_mse', init=None,\n",
       "                           learning_rate=0.1, loss='deviance', max_depth=3,\n",
       "                           max_features=None, max_leaf_nodes=None,\n",
       "                           min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                           min_samples_leaf=1, min_samples_split=2,\n",
       "                           min_weight_fraction_leaf=0.0, n_estimators=100,\n",
       "                           n_iter_no_change=None, presort='auto',\n",
       "                           random_state=None, subsample=1.0, tol=0.0001,\n",
       "                           validation_fraction=0.1, verbose=0,\n",
       "                           warm_start=False)"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_gb.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8703055555555556"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_gb.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\model_selection\\_split.py:1978: FutureWarning: The default value of cv will change from 3 to 5 in version 0.22. Specify it explicitly to silence this warning.\n",
      "  warnings.warn(CV_WARNING, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'learning_rate': 0.1, 'min_samples_split': 4, 'n_estimators': 100}"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "params2 = {'learning_rate': [0.1, 0.01, 0.001], 'n_estimators': [100, 200, 500], 'min_samples_split': [2, 3, 4]}\n",
    "CV2 = GridSearchCV(clf_gb, params2)\n",
    "CV2.fit(X_train, y_train)\n",
    "CV2.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf_gb = GradientBoostingClassifier(learning_rate=0.01, min_samples_split=3, n_estimators=500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.9627294 , 0.18524046],\n",
       "       [0.11828495, 0.41210614]])"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_gb.fit(X_train, y_train)\n",
    "y_pred = clf_gb.predict(X_test)\n",
    "results = confusion_matrix(y_test, y_pred)\n",
    "results / results.sum(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8705"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf_gb.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Significantly higher computation time at no apparent benefit.\n",
    "\n",
    "I held on to this until the end as it is a computation hog. I would not take the GB route for this problem, even though it is known to perform better in many cases. The grid search took several minutes, as opposed to logistic regressions ten seconds or less. The scoring and confusion matrices are very similar. \n",
    "\n",
    "#### Dropping features.\n",
    "\n",
    "Let's check our selected model - logistic regression - against the aforementioned data: the more highly correlated features of purchase amount, age, and days since last purchase."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\clayp\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8695277777777778"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3 = df.drop(columns=['female', 'male', 'yes'])\n",
    "X_train, X_test, y_train, y_test = train_test_split(df3.drop(columns='loyalty'), df3.loyalty, test_size=0.3)\n",
    "clf_log.fit(X_train, y_train)\n",
    "clf_log.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.96226856, 0.18771784],\n",
       "       [0.1189658 , 0.40813278]])"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results = confusion_matrix(y_test, clf_log.predict(X_test))\n",
    "results / results.sum(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Very similar results with the uncorrelated features removed.\n",
    "\n",
    "As we suspected, it looks like the model performance did not at all rely on the uncorrelated features. This would be my final approach to this problem, and I would work to improve it from there."
   ]
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}