notebook859b19da0e.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Prediction of Credit Card Default\n",
    "\n",
    "### Tutorial for beginners in classification analysis using scikit-learn."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This dataset contains information on default payments, demographic factors, credit data, history of payment, and bill statements of credit card clients in Taiwan from April 2005 to September 2005.\n",
    "\n",
    "\n",
    "# Table of Content\n",
    "\n",
    "* [Objectives](#obj)\n",
    "* [Importing packages and loading data](#imp)\n",
    "* [Feature Engineering](#fe)\n",
    "* [Exploratory Data Analysis (EDA)](#eda)\n",
    "    * [Mapping the target: categorizing](#map)\n",
    "    * [Descriptive Statistics](#stat)\n",
    "    * [Standardizing and plotting features](#std)\n",
    "    * [Correlation](#corr)\n",
    "* [Machine Learning: Classification models](#ml)\n",
    "    * [Feature Selection](#fs)\n",
    "    * [Spliting the data: train and test](#sp)\n",
    "    * [Logistic Regression (original data)](#lr1)\n",
    "    * [Logistic Regression (standardized features)](#lr2)\n",
    "    * [Logistic Regression (most important features)](#lr3)\n",
    "    * [ExtraTree-decision](#tree)\n",
    "    * [Random-Forest Classifier](#rf)\n",
    "* [Comparison of model performance](#sum)\n",
    "    * [Receiver operating characteristic (ROC) Curve](#roc)\n",
    "    * [Mean Accuracy (coss-validation)](#ac)\n",
    "    * [Precision, Recall, F1-score](#m)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='obj'></a>\n",
    "## Objectives:<br>\n",
    "-       Identify the key drivers that determine the likelihood of credit card default.\n",
    "-       Predict the likelihood of credit card default for customers of the Bank."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='imp'></a>\n",
    "## Importing packages and loading data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [],
   "source": [
    "# here we will import the libraries used for machine learning\n",
    "import numpy as np # linear algebra\n",
    "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
    "from scipy.stats import randint\n",
    "import pandas as pd # data processing, CSV file I/O, data manipulation \n",
    "import matplotlib.pyplot as plt # this is used for the plot the graph \n",
    "import seaborn as sns # used for plot interactive graph. \n",
    "from pandas import set_option\n",
    "plt.style.use('ggplot') # nice plots\n",
    "\n",
    "from sklearn.model_selection import train_test_split # to split the data into two parts\n",
    "from sklearn.linear_model import LogisticRegression # to apply the Logistic regression\n",
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.model_selection import KFold # for cross validation\n",
    "from sklearn.model_selection import GridSearchCV # for tuning parameter\n",
    "from sklearn.model_selection import RandomizedSearchCV  # Randomized search on hyper parameters.\n",
    "from sklearn.preprocessing import StandardScaler # for normalization\n",
    "from sklearn.pipeline import Pipeline \n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "# import xgboost as xgb\n",
    "# from xgboost import XGBClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.neighbors import KNeighborsClassifier #KNN\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "from sklearn import metrics # for the check the error and accuracy of the model\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\", category=FutureWarning)\n",
    "\n",
    "import os\n",
    "#print(os.listdir(\"../input\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "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>ID</th>\n",
       "      <th>LIMIT_BAL</th>\n",
       "      <th>SEX</th>\n",
       "      <th>EDUCATION</th>\n",
       "      <th>MARRIAGE</th>\n",
       "      <th>AGE</th>\n",
       "      <th>PAY_0</th>\n",
       "      <th>PAY_2</th>\n",
       "      <th>PAY_3</th>\n",
       "      <th>PAY_4</th>\n",
       "      <th>...</th>\n",
       "      <th>BILL_AMT4</th>\n",
       "      <th>BILL_AMT5</th>\n",
       "      <th>BILL_AMT6</th>\n",
       "      <th>PAY_AMT1</th>\n",
       "      <th>PAY_AMT2</th>\n",
       "      <th>PAY_AMT3</th>\n",
       "      <th>PAY_AMT4</th>\n",
       "      <th>PAY_AMT5</th>\n",
       "      <th>PAY_AMT6</th>\n",
       "      <th>default.payment.next.month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>28052</th>\n",
       "      <td>28053</td>\n",
       "      <td>10000.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>32</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>9493.0</td>\n",
       "      <td>9678.0</td>\n",
       "      <td>9634.0</td>\n",
       "      <td>7417.0</td>\n",
       "      <td>1166.0</td>\n",
       "      <td>1200.0</td>\n",
       "      <td>300.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>500.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18603</th>\n",
       "      <td>18604</td>\n",
       "      <td>20000.0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>5689.0</td>\n",
       "      <td>4587.0</td>\n",
       "      <td>2293.0</td>\n",
       "      <td>1249.0</td>\n",
       "      <td>1125.0</td>\n",
       "      <td>1000.0</td>\n",
       "      <td>92.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20505.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22433</th>\n",
       "      <td>22434</td>\n",
       "      <td>100000.0</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>36</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>...</td>\n",
       "      <td>112016.0</td>\n",
       "      <td>112906.0</td>\n",
       "      <td>120101.0</td>\n",
       "      <td>5220.0</td>\n",
       "      <td>16608.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4233.0</td>\n",
       "      <td>9200.0</td>\n",
       "      <td>21000.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18866</th>\n",
       "      <td>18867</td>\n",
       "      <td>170000.0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>164606.0</td>\n",
       "      <td>121547.0</td>\n",
       "      <td>121118.0</td>\n",
       "      <td>7400.0</td>\n",
       "      <td>6328.0</td>\n",
       "      <td>5280.0</td>\n",
       "      <td>4300.0</td>\n",
       "      <td>4500.0</td>\n",
       "      <td>4300.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14719</th>\n",
       "      <td>14720</td>\n",
       "      <td>240000.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>33</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>-1</td>\n",
       "      <td>...</td>\n",
       "      <td>6997.0</td>\n",
       "      <td>15607.0</td>\n",
       "      <td>44412.0</td>\n",
       "      <td>58243.0</td>\n",
       "      <td>20469.0</td>\n",
       "      <td>7032.0</td>\n",
       "      <td>15685.0</td>\n",
       "      <td>44513.0</td>\n",
       "      <td>20159.0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          ID  LIMIT_BAL  SEX  EDUCATION  MARRIAGE  AGE  PAY_0  PAY_2  PAY_3  PAY_4  ...  \\\n",
       "28052  28053    10000.0    2          2         2   32     -1     -1      0      0  ...   \n",
       "18603  18604    20000.0    2          1         2   25      0      0      0      0  ...   \n",
       "22433  22434   100000.0    2          3         1   36      2      0      0      2  ...   \n",
       "18866  18867   170000.0    2          2         1   30      0      0      0      0  ...   \n",
       "14719  14720   240000.0    1          1         2   33     -1     -1     -1     -1  ...   \n",
       "\n",
       "       BILL_AMT4  BILL_AMT5  BILL_AMT6  PAY_AMT1  PAY_AMT2  PAY_AMT3  PAY_AMT4  PAY_AMT5  \\\n",
       "28052     9493.0     9678.0     9634.0    7417.0    1166.0    1200.0     300.0    1000.0   \n",
       "18603     5689.0     4587.0     2293.0    1249.0    1125.0    1000.0      92.0       0.0   \n",
       "22433   112016.0   112906.0   120101.0    5220.0   16608.0       0.0    4233.0    9200.0   \n",
       "18866   164606.0   121547.0   121118.0    7400.0    6328.0    5280.0    4300.0    4500.0   \n",
       "14719     6997.0    15607.0    44412.0   58243.0   20469.0    7032.0   15685.0   44513.0   \n",
       "\n",
       "       PAY_AMT6  default.payment.next.month  \n",
       "28052     500.0                           1  \n",
       "18603   20505.0                           0  \n",
       "22433   21000.0                           1  \n",
       "18866    4300.0                           0  \n",
       "14719   20159.0                           0  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('/Users/nic/Downloads/学校/论文/UCI_Credit_Card.csv')\n",
    "data.sample(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are variables that need to be converted to categories:\n",
    "* __SEX:__ Gender   \n",
    "                    1 = male \n",
    "                    2 = female\n",
    "* __EDUCATION:__     \n",
    "                     1 = graduate school \n",
    "                     2 = university \n",
    "                     3 = high school \n",
    "                     4 = others \n",
    "                     5 = unknown \n",
    "                     6 = unknown\n",
    "* __MARRIAGE:__ Marital status \n",
    "                    1 = married\n",
    "                    2 = single\n",
    "                    3 = others\n",
    "* __PAY_0,2,3,4,5,6:__ Repayment status in September 2005, August 2005, July 2005, June 2005, May 2005, April 2005 (respectivey)\n",
    "                    -2= no consumption\n",
    "                    -1= pay duly\n",
    "                    1 = payment delay for one month\n",
    "                    2 = payment delay for two months\n",
    "                    ... \n",
    "                    8 = payment delay for eight months\n",
    "                    9 = payment delay for nine months and above"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='fe'></a>\n",
    "## Feature engineering\n",
    "\n",
    "The data has been already encoded and cleaned. However, some categorical data have repeated categories. For instance, the variable ‘education’ has three categories with similar information:<br>\n",
    "4: others, 5: unknown, and 6: unknown<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 30000 entries, 0 to 29999\n",
      "Data columns (total 24 columns):\n",
      " #   Column     Non-Null Count  Dtype  \n",
      "---  ------     --------------  -----  \n",
      " 0   LIMIT_BAL  30000 non-null  float64\n",
      " 1   SEX        30000 non-null  int64  \n",
      " 2   EDUCATION  30000 non-null  int64  \n",
      " 3   MARRIAGE   30000 non-null  int64  \n",
      " 4   AGE        30000 non-null  int64  \n",
      " 5   PAY_0      30000 non-null  int64  \n",
      " 6   PAY_2      30000 non-null  int64  \n",
      " 7   PAY_3      30000 non-null  int64  \n",
      " 8   PAY_4      30000 non-null  int64  \n",
      " 9   PAY_5      30000 non-null  int64  \n",
      " 10  PAY_6      30000 non-null  int64  \n",
      " 11  BILL_AMT1  30000 non-null  float64\n",
      " 12  BILL_AMT2  30000 non-null  float64\n",
      " 13  BILL_AMT3  30000 non-null  float64\n",
      " 14  BILL_AMT4  30000 non-null  float64\n",
      " 15  BILL_AMT5  30000 non-null  float64\n",
      " 16  BILL_AMT6  30000 non-null  float64\n",
      " 17  PAY_AMT1   30000 non-null  float64\n",
      " 18  PAY_AMT2   30000 non-null  float64\n",
      " 19  PAY_AMT3   30000 non-null  float64\n",
      " 20  PAY_AMT4   30000 non-null  float64\n",
      " 21  PAY_AMT5   30000 non-null  float64\n",
      " 22  PAY_AMT6   30000 non-null  float64\n",
      " 23  Default    30000 non-null  int64  \n",
      "dtypes: float64(13), int64(11)\n",
      "memory usage: 5.5 MB\n"
     ]
    }
   ],
   "source": [
    "data.rename(columns={\"default.payment.next.month\": \"Default\"}, inplace=True)\n",
    "data.drop('ID', axis = 1, inplace =True) # drop column \"ID\"\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Separating features and target\n",
    "y = data.Default     # target default=1 or non-default=0\n",
    "features = data.drop('Default', axis = 1, inplace = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1, 3, 5, 4, 6, 0])"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['EDUCATION'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The categories 4:others, 5:unknown, and 6:unknown can be grouped into a single class '4'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['EDUCATION']=np.where(data['EDUCATION'] == 5, 4, data['EDUCATION'])\n",
    "data['EDUCATION']=np.where(data['EDUCATION'] == 6, 4, data['EDUCATION'])\n",
    "data['EDUCATION']=np.where(data['EDUCATION'] == 0, 4, data['EDUCATION'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After grouping, the education column has the following categories:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1, 3, 4])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['EDUCATION'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly, the column 'marriage' should have three categories: 1 = married, 2 = single, 3 = others but it contains a category '0' which will be joined to the category '3'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 2, 3, 0])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['MARRIAGE'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 2, 3])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['MARRIAGE']=np.where(data['MARRIAGE'] == 0, 3, data['MARRIAGE'])\n",
    "data['MARRIAGE'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='eda'></a>\n",
    "## Exploratory Data Analysis (EDA)\n",
    "\n",
    "<a id='map'></a>\n",
    "### Mapping the target: categorizing\n",
    "\n",
    "From this sample of 30,000 credit card holders, there were 6,636 default credit cards; that is, the proportion of default in the data is 22,1%."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['font.sans-serif'] = ['Arial Unicode MS']\n",
    "\n",
    "# The frequency of defaults\n",
    "yes = data.Default.sum()\n",
    "no = len(data)-yes\n",
    "\n",
    "# Percentage\n",
    "yes_perc = round(yes/len(data)*100, 1)\n",
    "no_perc = round(no/len(data)*100, 1)\n",
    "\n",
    "import sys \n",
    "plt.figure(figsize=(7,4))\n",
    "sns.set_context('notebook', font_scale=1.2)\n",
    "sns.countplot('Default',data=data, palette=\"Blues\")\n",
    "plt.annotate('未违约还款: {}'.format(no), xy=(-0.3, 15000), xytext=(-0.3, 3000), size=12)\n",
    "plt.annotate('违约还款: {}'.format(yes), xy=(0.7, 15000), xytext=(0.7, 3000), size=12)\n",
    "plt.annotate(str(no_perc)+\" %\", xy=(-0.3, 15000), xytext=(-0.1, 8000), size=12)\n",
    "plt.annotate(str(yes_perc)+\" %\", xy=(0.7, 15000), xytext=(0.9, 8000), size=12)\n",
    "plt.xlabel('下一月是否违约还款')\n",
    "plt.ylabel('计数')\n",
    "plt.title('标签（Default）统计', size=14)\n",
    "#Removing the frame\n",
    "plt.box(False);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='stat'></a>\n",
    "### Descriptive Statistics\n",
    "The table below shows the descriptive statistics of the variables of this dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SUMMARY STATISTICS OF NUMERIC COLUMNS\n",
      "\n",
      "             count       mean        std       min       25%       50%        75%       max\n",
      "LIMIT_BAL  30000.0  167484.32  129747.66   10000.0  50000.00  140000.0  240000.00  1.00e+06\n",
      "SEX        30000.0       1.60       0.49       1.0      1.00       2.0       2.00  2.00e+00\n",
      "EDUCATION  30000.0       1.84       0.74       1.0      1.00       2.0       2.00  4.00e+00\n",
      "MARRIAGE   30000.0       1.56       0.52       1.0      1.00       2.0       2.00  3.00e+00\n",
      "AGE        30000.0      35.49       9.22      21.0     28.00      34.0      41.00  7.90e+01\n",
      "PAY_0      30000.0      -0.02       1.12      -2.0     -1.00       0.0       0.00  8.00e+00\n",
      "PAY_2      30000.0      -0.13       1.20      -2.0     -1.00       0.0       0.00  8.00e+00\n",
      "PAY_3      30000.0      -0.17       1.20      -2.0     -1.00       0.0       0.00  8.00e+00\n",
      "PAY_4      30000.0      -0.22       1.17      -2.0     -1.00       0.0       0.00  8.00e+00\n",
      "PAY_5      30000.0      -0.27       1.13      -2.0     -1.00       0.0       0.00  8.00e+00\n",
      "PAY_6      30000.0      -0.29       1.15      -2.0     -1.00       0.0       0.00  8.00e+00\n",
      "BILL_AMT1  30000.0   51223.33   73635.86 -165580.0   3558.75   22381.5   67091.00  9.65e+05\n",
      "BILL_AMT2  30000.0   49179.08   71173.77  -69777.0   2984.75   21200.0   64006.25  9.84e+05\n",
      "BILL_AMT3  30000.0   47013.15   69349.39 -157264.0   2666.25   20088.5   60164.75  1.66e+06\n",
      "BILL_AMT4  30000.0   43262.95   64332.86 -170000.0   2326.75   19052.0   54506.00  8.92e+05\n",
      "BILL_AMT5  30000.0   40311.40   60797.16  -81334.0   1763.00   18104.5   50190.50  9.27e+05\n",
      "BILL_AMT6  30000.0   38871.76   59554.11 -339603.0   1256.00   17071.0   49198.25  9.62e+05\n",
      "PAY_AMT1   30000.0    5663.58   16563.28       0.0   1000.00    2100.0    5006.00  8.74e+05\n",
      "PAY_AMT2   30000.0    5921.16   23040.87       0.0    833.00    2009.0    5000.00  1.68e+06\n",
      "PAY_AMT3   30000.0    5225.68   17606.96       0.0    390.00    1800.0    4505.00  8.96e+05\n",
      "PAY_AMT4   30000.0    4826.08   15666.16       0.0    296.00    1500.0    4013.25  6.21e+05\n",
      "PAY_AMT5   30000.0    4799.39   15278.31       0.0    252.50    1500.0    4031.50  4.27e+05\n",
      "PAY_AMT6   30000.0    5215.50   17777.47       0.0    117.75    1500.0    4000.00  5.29e+05\n",
      "Default    30000.0       0.22       0.42       0.0      0.00       0.0       0.00  1.00e+00\n"
     ]
    }
   ],
   "source": [
    "set_option('display.width', 100)\n",
    "set_option('precision', 2)\n",
    "\n",
    "print(\"SUMMARY STATISTICS OF NUMERIC COLUMNS\")\n",
    "print()\n",
    "print(data.describe().T)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The average value for the amount of credit card limit is 167,484 NT dollars. The standard deviation is 129,747 NT dollars, ranging from 10,000 to 1M NT dollars.<br>\n",
    "Education level is mostly graduate school (1) and university (2). Most of the clients are either marrined or single (less frequent the other status). Average age is 35.5 years, with a standard deviation of 9.2 years."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Frequency of explanatory variables by defaulted and non-defaulted cards"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['LIMIT_BAL' 'AGE' 'BILL_AMT1' 'BILL_AMT2' 'BILL_AMT3' 'BILL_AMT4'\n",
      " 'BILL_AMT5' 'BILL_AMT6' 'PAY_AMT1' 'PAY_AMT2' 'PAY_AMT3' 'PAY_AMT4'\n",
      " 'PAY_AMT5' 'PAY_AMT6' 'Default']\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1080x216 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## data are distributed in a wide range (below), need to be normalizded.\n",
    "plt.figure(figsize=(15,3))\n",
    "data_tmp = data.drop(['SEX', 'EDUCATION', 'MARRIAGE', 'PAY_0', 'PAY_2', 'PAY_3', 'PAY_4', \n",
    "               'PAY_5', 'PAY_6'], axis=1)\n",
    "data_tmp_col = data_tmp.columns.values\n",
    "print(data_tmp_col)\n",
    "f, axes = plt.subplots(ncols=2, figsize=(15, 3), facecolor='white')\n",
    "\n",
    "data_bill_amount_col = ['BILL_AMT1','BILL_AMT2','BILL_AMT3','BILL_AMT4','BILL_AMT5','BILL_AMT6']\n",
    "data_bill_amount = data_tmp[data_bill_amount_col]\n",
    "ax1= data_bill_amount.boxplot(data_bill_amount.columns.name, rot=90, ax=axes[0])\n",
    "outliers = dict(markerfacecolor='b', marker='p')\n",
    "ax1= data_bill_amount.boxplot(data_bill_amount.columns.name, rot=90, flierprops=outliers, ax=axes[0])\n",
    "ax1.set_ylim([-100000,170000])\n",
    "\n",
    "data_pay_amount_col = ['PAY_AMT1','PAY_AMT2','PAY_AMT3','PAY_AMT4','PAY_AMT5','PAY_AMT6']\n",
    "data_pay_amount = data_tmp[data_pay_amount_col]\n",
    "ax2= data_pay_amount.boxplot(data_pay_amount.columns.name, rot=90, ax=axes[1])\n",
    "outliers = dict(markerfacecolor='b', marker='p')\n",
    "ax2= data_pay_amount.boxplot(data_pay_amount.columns.name, rot=90, flierprops=outliers, ax=axes[1])\n",
    "ax2.set_ylim([-5000,13000])\n",
    "\n",
    "plt.xticks(size=12)\n",
    "plt.box(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1080x216 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, axes = plt.subplots(ncols=2, figsize=(15, 3), facecolor='white')\n",
    "data_bill_amount_col = ['BILL_AMT1','BILL_AMT2','BILL_AMT3','BILL_AMT4','BILL_AMT5','BILL_AMT6']\n",
    "data_bill_amount = data_tmp[data_bill_amount_col]\n",
    "ax1 = data_bill_amount.plot.box(title=\"Consumer spending in each country\", ax=axes[0])\n",
    "\n",
    "ax1.set_ylim([-100000,170000])\n",
    "# data_bill_amount.describe().T\n",
    "# data_pay_amount_col = ['PAY_AMT1','PAY_AMT2','PAY_AMT3','PAY_AMT4','PAY_AMT5','PAY_AMT6']\n",
    "# data_pay_amount = data_tmp[data_pay_amount_col]\n",
    "# ax2= data_pay_amount.boxplot(data_pay_amount.columns.name, rot=90, ax=axes[1])\n",
    "# outliers = dict(markerfacecolor='b', marker='p')\n",
    "# ax2= data_pay_amount.boxplot(data_pay_amount.columns.name, rot=90, flierprops=outliers, ax=axes[1])\n",
    "# ax2.set_ylim([-5000,13000])\n",
    "\n",
    "# plt.xticks(size=12)\n",
    "plt.box(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 9 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "plt.rcParams['font.sans-serif'] = ['Arial Unicode MS']\n",
    "sns.color_palette(\"Set2\")\n",
    "\n",
    "subset = data[['SEX', 'EDUCATION', 'MARRIAGE', 'PAY_0', 'PAY_2', 'PAY_3', 'PAY_4', \n",
    "               'PAY_5', 'PAY_6', 'Default']]\n",
    "\n",
    "f, axes = plt.subplots(3, 3, figsize=(20, 15), facecolor='white')\n",
    "# f.suptitle(u'特征变量的统计信息')\n",
    "ax1 = sns.countplot(x=\"SEX\", hue=\"Default\", data=subset, palette=\"Blues\", ax=axes[0,0])\n",
    "ax2 = sns.countplot(x=\"EDUCATION\", hue=\"Default\", data=subset, palette=\"Blues\",ax=axes[0,1])\n",
    "ax3 = sns.countplot(x=\"MARRIAGE\", hue=\"Default\", data=subset, palette=\"Blues\",ax=axes[0,2])\n",
    "ax4 = sns.countplot(x=\"PAY_0\", hue=\"Default\", data=subset, palette=\"Blues\", ax=axes[1,0])\n",
    "ax5 = sns.countplot(x=\"PAY_2\", hue=\"Default\", data=subset, palette=\"Blues\", ax=axes[1,1])\n",
    "ax6 = sns.countplot(x=\"PAY_3\", hue=\"Default\", data=subset, palette=\"Blues\", ax=axes[1,2])\n",
    "ax7 = sns.countplot(x=\"PAY_4\", hue=\"Default\", data=subset, palette=\"Blues\", ax=axes[2,0])\n",
    "ax8 = sns.countplot(x=\"PAY_5\", hue=\"Default\", data=subset, palette=\"Blues\", ax=axes[2,1])\n",
    "ax9 = sns.countplot(x=\"PAY_6\", hue=\"Default\", data=subset, palette=\"Blues\", ax=axes[2,2]);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/nic/Library/Python/3.7/lib/python/site-packages/numpy/core/_asarray.py:83: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray\n",
      "  return array(a, dtype, copy=False, order=order)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x1 = list(data[data['Default'] == 1]['LIMIT_BAL'])\n",
    "x2 = list(data[data['Default'] == 0]['LIMIT_BAL'])\n",
    "\n",
    "plt.figure(figsize=(12,4))\n",
    "sns.set_context('notebook', font_scale=1.2)\n",
    "#sns.set_color_codes(\"pastel\")\n",
    "plt.hist([x1, x2], bins = 40, density=False, stacked=False, color=['steelblue', 'lightblue']) # , rwidth=0.8\n",
    "plt.xlim([0,600000])\n",
    "plt.legend(['1', '0'], title = 'Default', loc='upper right', facecolor='white')\n",
    "plt.xlabel('信贷金额(新台币)')\n",
    "plt.ylabel('计数')\n",
    "# plt.title('LIMIT BALANCE HISTOGRAM BY TYPE OF CREDIT CARD', SIZE=15)\n",
    "plt.box(False)\n",
    "plt.savefig('ImageName', format='png', dpi=400, transparent=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 30,000 credit card clients.\n",
    "\n",
    "The average value for the amount of credit card limit is 167,484 NT dollars. The standard deviation is 129,747 NT dollars, ranging from 10,000 to 1M NT dollars.\n",
    "\n",
    "Education level is mostly graduate school and university.\n",
    "\n",
    "Most of the clients are either marrined or single (less frequent the other status).\n",
    "\n",
    "Average age is 35.5 years, with a standard deviation of 9.2.\n",
    "\n",
    "As the value 0 for default payment means 'not default' and value 1 means 'default', the mean of 0.221 means that there are 22.1% of credit card contracts that will default next month (will verify this in the next sections of this analysis)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x1 = list(data[data['Default'] == 1]['AGE'])\n",
    "x2 = list(data[data['Default'] == 0]['AGE'])\n",
    "\n",
    "plt.figure(figsize=(12,4))\n",
    "sns.set_context('notebook', font_scale=1.2)\n",
    "#sns.set_color_codes(\"pastel\")\n",
    "plt.hist([x1, x2], bins = 40, density=False, stacked=False, color=['steelblue', 'lightblue']) # , rwidth=0.8\n",
    "plt.xlim([20,80])\n",
    "plt.legend(['1', '0'], title = 'Default', loc='upper right', facecolor='white')\n",
    "plt.xlabel('年龄')\n",
    "plt.ylabel('计数')\n",
    "# plt.title('LIMIT BALANCE HISTOGRAM BY TYPE OF CREDIT CARD', SIZE=15)\n",
    "plt.box(False)\n",
    "plt.savefig('ImageName', format='png', dpi=400, transparent=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAAFWCAYAAAAsfDqEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAgAElEQVR4nOzdd1gUd/c28HuXKkuzK6IiJip2YgtWbFgQUYPYe+ya+DMoGLvRCGjU2IOaoLEiiRWN5bH3EomP2KLYEQERERBp8/7hu/u4Lijssjuzcn+uay+d2Zk5Z84Wzn5ndlYmCIIAIiIiIpIEudgJEBEREdH/sDkjIiIikhA2Z0REREQSwuaMiIiISELYnBERERFJCJszIiIiIglhc0ZkAE5OTpDJZKqbXC6HnZ0dOnfujJiYGNVy7u7uasu9e1uyZAkAYNasWWrzTUxMULx4cfTo0UNtW0pjx46FTCbD5cuX1eZPnDgRtra2iI2N1Vhn1KhRqFSpElJSUlTxhg4dqrGcIAhwcHCATCZDVlZWgfbB3NwcUVFRudYqNDQUx44dy3M7MpkMgwcPzrXWiYmJGD16NCpXrgyFQoHatWtj8eLFqvwGDx78we0eO3ZMbXu1a9dGyZIlkZGRoZHnh7ajjNW/f/9c83w/1unTp9G6dWsUL14cpUqVgoeHB06dOpXruvn1/mNhZWUFd3d3/PPPPxrLHj9+HB4eHrCzs4NCoUCzZs2wY8cOtWV69+6NKlWqIDU1VW3+zp07YWlpiWvXrqn27d2bhYUFqlevjrVr16rlNm3atA/mf+/ePQwcOBDlypWDpaUlatasieDgYGRmZqqW+dBj4OTkVNCSEUkCmzMiA1m8eDHi4+MRHx+Pp0+fYufOnbh9+zaGDBmittzEiRNVy717GzVqlGqZJk2aqG3ryJEjuHfvHnr37q22raysLGzfvh21atXC+vXr1e6bPXs2bGxs4O/vrzb/8uXLWLNmDZYvXw5ra2sAgJmZGSIiIpCTk6O27Pnz5xEXF6exr/nZh8zMTIwZMybPejVr1kxt3YoVK2LZsmWq6eXLl2usIwgCunXrhmfPnmHHjh24efMmZs+ejR9//BETJkwAACxfvly1jZ07dwKAWpxmzZqptvfPP/8gJiYGlpaW2Lt3r1qsK1euqNaZNGkSmjZtqradgrh27Ro6deqE7t274++//8bp06dRt25dtG7dGufOnSvQtt6nfCzi4uJw9epVODk5oUePHmrN5qZNm+Dh4QE3NzdcuHABUVFR8PX1Rf/+/REUFKRabtmyZUhJScHUqVNV8+Li4jBixAjMmTMHtWvXVs3/448/VLW4c+cORo8ejeHDh+PIkSP5yjsqKgqNGjXCmzdvsHfvXty9exfz5s3D6tWr0a1bN9Vz8d2a9+7dG71791ZNX7lyRafaEYlGICK9q1y5srBmzRqN+Rs3bhQACImJiYIgCEKrVq2EqVOnfnBbM2fOFJo1a6Yxf9u2bWrbEgRB2Ldvn2BtbS2sWrVKKFWqlJCRkaG2ztatWwWZTCacOXNGEARByMnJEdzc3IQePXqoxWvUqJFgY2OjWk7J399fcHd3FwAImZmZBdoHBwcHwdTUVFi/fr3afZUrVxZ+++03jXXymv+uW7duCQCE+Ph4tfnLly8XFAqFxvJHjx4VPvQ2OHnyZKFTp05C3759ha5du35wf1q1aqUxf9CgQUK/fv1yXQeAcPToUUEQBGHKlCmCp6enxjINGzYUxo4dm2fcj8ntsYiLixMACBcvXhQEQRCePn0qWFlZCcuXL9dYf8uWLYJcLhciIyNV8zZt2iTI5XLh7NmzgiAIQrdu3YSmTZsK2dnZavt26NAhje25uLgI33zzTZ65veuLL74QevbsqTH/3r17goWFhbBs2TKN+wYNGiQMGjQoz20SGQuOnBGJyNLSEsDbkSldKQ+XWlhYqOZt3rwZTZs2hbe3N54/f64x+tOrVy+0bdsW48aNgyAI2LhxI6KiorBs2TK15czNzdGpUyfs2bNHbf7OnTvRrVs3rfKtUqUKJkyYgEmTJiEpKUmrbbwvOzsbALB//361+YMGDUJkZGSBtiUIArZu3YoWLVrAy8sL+/fvL/CIWH5lZ2fj6tWrePLkidr8P//8E7NmzSrUWMWKFQMAFC9eHACwceNGKBQKjBw5UmPZXr16wcXFBSEhIap5ffv2RefOnTF8+HBs2LABBw8exPr16yGXf/zPiUwmU8X/kEuXLuHvv/9WG6FTcnJywtChQ7Fq1aqPbofIWLE5IxLJkydPEBwcjE6dOqkOH2rrzp07CAwMhIeHB6ysrAAAr1+/xs6dO9G1a1eUL18eTZo00Ti0CQArVqzAtWvXsHbtWgQEBGDevHlwcHDQWM7b2xu7d+9WTd+6dQsZGRmoV6+e1nnPnDkT5ubmmDJlitbbeFeNGjXQpk0bDBw4ELVq1cKkSZNw4MABWFhY4LPPPivQtk6dOoWHDx+ia9eu8PT0hEwmw6ZNmwolz/cNGTIECQkJcHZ2RseOHbFo0SLcuHEDFStWRKlSpQotTlZWFhYvXoyRI0eiatWqAICLFy+iYcOGMDU11VheJpOhQYMGuHjxotr81atX4+HDhxg2bBiCgoI+WtuMjAwsX74cN2/ehLe390fzvHjxIhQKBerWrZvr/Y0aNcL169c1zn0j+lSwOSMykG+++Qb29vawt7eHjY0N6tevD1dXV2zevFltuYULF6qWe/f27h+ic+fOqeZbWVmhRo0aKFu2LH799VfVMnv27EFqaqrqj2GPHj2wb98+JCQkqMWrVq0aJk2ahNGjR8PR0THP88A6d+6M27dv4969ewDejprl9Yc2P/sAANbW1liyZAlCQkJw4cKFfFYybzKZDPv27cO8efMgCAIWLlyIjh07onLlyoiIiCjQtrZs2YLPPvsMtWrVgo2NDdq1a5drc1sYatSogcjISPTq1Qvnz5/Hd999h5o1a6J9+/a5ntNXEO8+FhYWFpg1axaKFy+uGmVMTExEiRIl8ly/fPnyePHihdq8ChUqwNXVFVlZWXBzc8t1vW7duqniFitWDD/++CNWr16d5/LvSkxMhL29veqLFbnlBKDQRlyJpIbNGZGBzJw5E5GRkTh69Cg8PDxQuXJlTJ06Ffb29mrLDR8+HJGRkRo35YgYALi6uqrm37p1CykpKdi3b5/qjxbw9pBmw4YN4ejoCADo3r07MjMzNZpBAJg6dSqys7Mxbdq0PA9P2dvbo2XLlqrRsw8d0szPPih99dVX8PDwwOjRo1UNgy4sLCzw/fff4/r167h79y6WLl0KhUIBHx8fjcOGeVF+kaJ79+6qed27d0dkZCSuXr2a71zMzMwgCILGfOXJ7O8ezq5WrRo2bNiA+Ph4nDx5Et988w1OnjyJYcOG5brtWrVqwdraGtbW1qhVq1aeObz7WFy+fBmLFy/GokWLMGfOHACAra0tnj59muf6T58+VR0CVVq3bh0uXryIxo0b4+uvv1Z9E/Zdv/zyCyIjI/HPP/8gNjYWMTExGD58eJ5x3mVra4v4+Pg8nw/KfN9/7RB9KticERlIyZIl4eTkpBoty8nJgY+Pj9plAQDAzs4OTk5OGrd3RxEsLCxU8ytWrKg6d00pKSkJ+/fvx+XLl2FpaQlLS0vVN+lCQ0M1clOeB/Sx84G8vb2xZ88exMbGIjo6Gs2bN891ufzsw7uWL1+O69evY+XKlR+M/zHh4eFq5yI5Oztj/PjxOHfuHLKzs3H27Nl8befgwYNISEjA4sWLVfUbO3YsgNzrlxd7e3u8evVKY/7Lly9V9wOAn5+f6pw4U1NTNG/eHD///DMWLFiAo0eP5rrtAwcO4Nq1a7h27RoOHDiQZw7vPhb169fHN998g4EDB6ouk9GkSRNcvXoV6enpGusKgoBLly6hUaNGqnkPHz7ExIkTMWvWLISGhuL69esIDg7WWLds2bJwcnJC5cqVUbp06Tzzy02TJk2QkZGR53mCFy9eRM2aNaFQKAq0XSJjweaMSAQWFhZYu3YtLl26hIULFxb69sPDw5GVlYXjx4+rjVzNnTsXV65cwX//+1+tttu1a1ecOHECGzZsQOfOnWFiYlIo+VatWhUBAQGYPn06kpOTtd7OnTt38MMPPyAlJUVtvo2NDczNzTVGgPKyefNmVK1aFf/9739Vtfvnn3/QtWtXbNq0KdeRotzUrFkTFy9e1LhG2rlz56BQKFTnau3duxfr1q3TWL9EiRJ55uzo6KhqupSjo/llZmYGGxsbAG+vxZaeno6lS5dqLBcWFoabN2+qRrwEQcDQoUPx+eefY+LEiXBxccGUKVMwZ84c3Lp1q0A5fEjjxo3RuHFjzJgxQ+O+Bw8e4LffflO7LAvRp4bNGZFIvvjiCwwfPhzz5s1TO9z2+vVrJCQkaNxyG4HJy+bNm9GuXTs0b94cNWrUUN3Gjx8PhUKh9blTlStXRq1atTBv3rwPfktTm30ICAhA6dKlNc5vKohhw4YhJycHHTt2xKFDh/D48WNcunQJvr6+cHFxgbu7+0e3kZaWhl27dmHYsGFqtatRowa+/fZbxMXF4a+//spXPj4+PjA1NUWfPn1w9epVPH78GHv37sXo0aPx7bffqr5ZO3XqVKxYsQL+/v64evUqHj16hD///BP+/v6YNGmS1vUA1B+LBw8eYPPmzdiwYYPq+nplypTBunXrMGPGDMyZMwd37tzB/fv3sXTpUgwbNgzz5s1D/fr1Abz98siJEyfw66+/qhrz77//HlWrVsXXX3+d6yHcD4mOjsZff/2ldrt//z4AYP369bhw4QL69OmDv//+GzExMdixYwfatGmDNm3aqEYyiT5JIl7Gg6jIyOs6Z/Hx8ULx4sWFXr16CYLw9tpPAHK9ffXVV4Ig5H2dM6WYmBhBLpcLYWFhud4/YsQIoVy5cqrrkikhj2tTvR9v5syZgpWVlZCWliYIwv+uFfbudc603YcDBw4IALS+zpkgCEJ0dLTQr18/oVy5coKZmZlQpUoV4f/+7/+EpKQkjWVzu87Z1q1bBRMTEyEmJibX7VevXl21H+/WJLfrnCnz8fHxEUqWLCmYm5sLn3/+uRAcHKx2XTBBEITw8HChWbNmgrW1taBQKIQGDRoIW7du/ej+fsj7j4WFhYVQvXp14aefftJY9vjx40L79u0FGxsbwcrKSnBzcxP++OMP1f3//vuvYGVlJcycOVNj3bNnzwpyuVx17bG8nksfyk15W7BggWqZe/fuCQMGDBDKlCkjmJubCzVq1BCCgoI0rtenxOuc0adCJggF/KhDRERERHrDw5pEREREEsLmjIiIiEhC2JwRERERSQibMyIiIiIJYXNGREREJCFszoiIiIgkhM0ZERERkYSwOSMiIiKSEDZnRERERBLC5oyIiIhIQticEREREUmIqdgJGEJycjLi4uKQmZkpdiqFyszMDGXKlIGtra3YqRAREVEh+eSbs+TkZDx79gwVKlRAsWLFIJPJxE6pUAiCgNevX+PJkycAwAaNiIjoEyGZw5qRkZGYNGkSBg0ahAkTJuDMmTOFst24uDhUqFABVlZWn0xjBgAymQxWVlaoUKEC4uLixE4nT6tXr4anpyd++eUXUeLPmTMHnp6emDt3rijxdZWYmAh/f38kJiaKnYpRErt+0dHR6NmzJ6Kjo0WJrysx6yf2Y2fsWD/diP3alURzlpGRgQULFsDLywvr16/HhAkTsGbNGjx+/FjnbWdmZqJYsWKFkKU0FStWTNKHa/fs2QMA2L17tyjxz58/DwA4e/asKPF1tWXLFkRFRWHr1q1ip2KUxK7fggULkJaWhoULF4oSX1di1k/sx87YsX66Efu1K4nmTBAEmJubQxAE5OTkQC6XIycnp1CaMwCf1IjZ+6S8b6tXr1abNvTo2Zw5c9SmjW30LDExEYcPH4YgCDh06BA/AReQ2PWLjo7Gw4cPAQAPHjwwutEzMesn9mNn7Fg/3UjhtSuJ5szCwgLjxo3D2rVr0bdvX/j5+cHR0RH169fXWDY1NRVxcXEat9TUVBEypw9RjpopGXr0TDlqpmRso2dbtmxBTk4OACAnJ4efgAtI7PotWLBAbdrYRs/ErJ/Yj52xY/10I4XXriSas4SEBCxduhRDhgzBxo0bMWvWLFSsWBFZWVkay0ZERGDcuHEat4iIiALHdXJygkwmU90sLS3h4uKCVatW5Xsb8fHxaNasGSwtLdG3b98C5/C+wYMHw8fHB8DbF9WqVauQnp6u83bJ+Bw7dkz1GsjKysLRo0dFzsi4iF0/5SdvpQcPHhg0vq7ErJ/Yj52xY/10I4XXriSaswsXLsDR0RFt2rSBqakpatasCWtra4SFhWks6+npieXLl2vcPD09tYodGBiI+Ph4xMfH4+bNmxgzZgwmTZqE+fPn52v9sLAw3Lx5ExcvXsSyZcu0yiEvJ0+exJgxY3JtUunT5+7uDlPTt1+oNjU1RevWrUXOyLiIXb9KlSqpTVeuXNmg8XUlZv3EfuyMHeunGym8diXRnJmZmWnMk8vlSEhI0JivUChQpkwZjZtCodAqtkKhQKlSpVCqVCk4OTlh/PjxWLRoEebPn4+XL19+dP2EhARUqVIFderUQcmSJbXKIS+CIBTq9gzNy8tLbbpr164Gjd+kSRO1aTc3N4PG11WfPn0gl799icrlcvTu3VvkjIyL2PWbNGmS2rSfn59B4+tKzPqJ/dgZO9ZPN1J47UqiOWvQoAGePHmCI0eOICcnB9HR0fjPf/6DL774QpR8+vXrh+zsbNXIXWBgICpVqgRLS0s0btwYR44cAfD2EOSsWbNw+fJlyGQyHDt2DG/evMHEiRNRsWJFWFhYoGLFipg3b55q2+8etlRycnLC8uXL1ebdv39f9WnHxsYGoaGhetxj/Rg1apTa9MiRIw0af8aMGWrT06ZNM2h8XZUoUQLt2rWDTCZD+/btUaJECbFTMipi18/Z2Vn1Cbxy5cpwdnY2aHxdiVk/sR87Y8f66UYKr11JNGclSpTA999/jyNHjmDIkCFYvHgxunfvjnbt2omSj0KhQJUqVRAVFYVVq1ZhyZIlWLFiBW7duoWvv/4a3t7euH79OpYvX45JkyahXr16qnPP/Pz8cPLkSezatQv37t3DxIkTMW3aNFy4cKFAOVSqVAk7d+4E8LZR69Onjz52Ve+Uo2eGHjVTUo6eGduomVKfPn1Qq1YtfvLVktj1mzRpEqysrIxu1ExJzPqJ/dgZO9ZPN2K/dmWCsR87+4gbN27AxcUl1/ucnJzg5+eHcePGadzXvHlzuLi44NChQ5g9ezYGDRqkuq9Pnz5wcHDATz/9hFmzZmHv3r24dOkSAOD3339H3bp1Ua9ePdXy9vb2WLlyJfr27YvBgwcjJSUF4eHhuebx7v3Hjh1D69at8erVK1hbW2u1j0RERGRcJDFyJkXJycmws7PDgwcPMG7cONjb26tuO3bswJ07d3Jdb8CAAXj27Bn8/f3h5eWFChUq4OXLl8jIyDDwHhAREZEx+uR/W1Mb6enpuH37Nr777jsAQGhoKBo0aKC2TF6/OtC/f38cP34c/fr1Q9++fbF8+XLUqVNHdb9MJtM40Z/fxiQiIiIlNme5WL9+PUxMTODl5YXy5cvjwYMH+Oqrr1T3f//996hWrRoGDx6stl5SUhI2bdqEo0ePwt3dHQBw7949vHr1StWQWVlZISYmRrVOSkoKnj17lmseUr76PxEREelHkW/OUlNTkZCQAEEQ8PLlS/z111+YPHkyfvjhB5QoUQKTJ0/G7Nmz4eDggKZNm2LXrl1YuHAhTpw4obEthUIBa2tr7NixA1WrVsX9+/fx3XffwczMTHUhWVdXV4SGhuLQoUNwcnLCjBkzcr2UCPC2kQOAq1evol69elpfLoSIiIiMR5E/5ywgIAClS5dGmTJl4Orqis2bN2PNmjWqQ5rffPMN/Pz8MHnyZFSrVg1r165FeHg4vvzyS41tmZmZYcuWLThw4AA+//xzjB8/HpMnT0aHDh1U39YcOHAgBg4cCB8fHzRv3hyurq5o2rRprrnVq1cPbdu2RevWrbFmzRr9FYGIiIgko0h/W/NTURT2kYiIqKgo8iNnRERERFLC5oyIiIhIQticEREREUkImzMiIiIiCWFzRkRERCQhbM6IiIiIJITNGREREZGEsDkjIiIikpAi+/NNM2bNRtLLZIPEsrezxZxZMw0Si4iIiIxbkW3Okl4mw2PgOIPEOrhheYGWf/ToEb7++mucOnUKJUuWxPTp0zF8+HA9Zadf0dHR8Pf3R1BQEJydnQ0ef/369QgLC0OvXr0wcOBAg8fX1erVq7Fnzx507doVI0eONHj8adOm4cqVK2jQoAHmzJlj8Pi62rdvH1asWIFx48ahU6dOBo/fp08fJCcnw97eHps2bTJ4fF0FBQXhxIkTaNWqFSZPnmzQ2J6enqr/R0REGDR2YRCzdoDx10/s9z6x68fDmhLk6+uLatWqISEhATt27ICfnx/OnDkjdlpaWbBgAdLS0rBw4UJR4oeFhQEAtm3bJkp8Xe3ZswcAsHv3blHiX7lyBQBw+fJlUeLrauXKlQCAFStWiBI/Ofnt6HxSUpIo8XV14sQJAMDx48dFzsT4sHa6Efu9T2xsziTm9u3buHTpEubPn49ixYqhQYMG8PX1xcaNG8VOrcCio6Px8OFDAMCDBw8QHR1t0Pjr169Xm96wYYNB4+tq9erVatO//PKLQeNPmzZNbXrGjBkGja+rffv2QfnTwYIgYP/+/QaN36dPH7Xpfv36GTS+roKCgtSmg4ODDRb73VGL3KalTszaAcZfP7Hf+6RQPzZnEhMVFYXKlSvD2tpaNa9OnTq4fv26iFlpZ8GCBWrThh49U46aKRnb6Jnyk6OSoT9BKkfNlIxt9Ew5aqZk6NEz5aiZkrGNnilHfpQ4ApR/rJ1uxH7vkwI2ZxKTkpICKysrtXkKhQIpKSkiZaQ95aiZ0oMHD0TKhIoi5ahZXtNERFLF5kxiFAoFUlNT1ealpaVBoVCIlJH2KlWqpDZduXJlkTKhokgmk31wmohIqticSYyLiwsePHigdkgkKioKNWvWFDEr7UyaNElt2s/Pz6DxfX191aZ79epl0Pi68vLyUpvu2rWrQeO7urqqTTdo0MCg8XU1ZswYtemxY8caNL6tra3atL29vUHj66ply5Zq061atRIpE+PD2ulG7Pc+KWBzJjEuLi6oX78+ZsyYgaysLJw5cwZbtmwxystAODs7q0bPKleubPBLaQwaNEht2thqOGrUKLVpQ3+dfO7cuWrTxnYpjc6dO6tGy2QymcEvpbFlyxa1aWO7lIa/v7/atCEvB/H+pQuM7VIQYtYOMP76if3eJ4X6FdnrnNnb2Rb4+mO6xCqI7du3Y+jQobC1tUWpUqWwaNEiuLm56Sk7/Zo0aRL8/f0NPmqm5Ovrq7rOmTHy8vJSXetHDK6urqrrnBmjMWPGYMWKFQYfNVOytbVVXefMGLVs2VJ1rS4qGNZON2K/94lNJnziZ8neuHEDLi4uYqehV0VhH4mIiIoKHtYkIiIikhA2Z0REREQSwuaMiIiISELYnBERERFJCJszIiIiIglhc0ZEREQkIWzOiIiIiCSEzRkRERGRhBTZXwiY/P0MxCe+MEis0iWKI/hH4/rpGyIiIhJHkW3O4hNfILZiW8MEe/QfrVY7f/48vLy8EBcXV8gJERERkVTxsKZEbdu2DR06dEBGRobYqRAREZEBsTmToFmzZiEoKAizZs0SOxUiIiIyMDZnEjRixAj8/fffqF+/vtipEBERkYGxOZMgBwcHsVMgIiIikUiqOXv+/DmCgoIwcOBADB8+HFu3boUgCGKnRTrYt28fPD09sX//flHib9++HZ6enggPDxclvq6uXLkCLy8vREZGihJ/yZIl8PT0xM8//yxKfF0Ze/5iE/v5Z8wSExPh7++PxMREUeIb+2Mn9t+OiRMnwtPTE35+fqLEl0xzJggCgoODYW9vj5CQEMyfPx/Hjx/HyZMnxU6NdLBy5UoAwIoVK0SJHxoaCgD47bffRImvq8DAQOTk5GD+/PmixD906BAA4ODBg6LE15Wx5y82sZ9/xmzLli2IiorC1q1bRYlv7I+d2H87bt26BQC4ceOGKPEl05z9+++/iI+Px5AhQ2BpaYlSpUph5syZqF27ttipkZb27dunGvkUBMHgn4C2b9+uNm1so2dXrlxBSkoKACAlJcXgn4CXLFmiNm1so0/Gnr/YxH7+GbPExEQcPnwYgiDg0KFDBh89M/bHTuy/HRMnTlSbFmP0TCZI5LjhX3/9hdOnT8PFxQXHjx+HmZkZOnbsiC5duqgtl5qaitTUVI31FQoFFAqFxvwbN27AxcVFY74xXIT22LFj6NatG5KSkj64XF77KLYuXbqoHZaWyWTYu3evweJ7enpqzIuIiDBYfF316tVL9QYLANbW1ti2bZvB4ht7/QyRf0hICKKjo/O8PyYmBsCHzyN1dnbGiBEjCjWvwmCI59+H6mfMtVuxYgUOHjyIrKwsmJqaokOHDhgzZozB4ov93qEr/u2Q0EVoU1NTcefOHdSqVQvLli1DTEwMfvzxR9jb26N58+aq5SIiInIdAfHx8YGvr2++4xnDFfvd3d0/2phJ2ft9v0Q+BxiNd99cc5sm6UtPTxc7Ba2J/fwz5todO3YMWVlZAICsrCwcPXrUoM2Z2I+drvi3Q0LNmZmZGczNzdGrVy/IZDI4OTmhTZs2OH/+vFpz5unpCXd3d431cxs1I3HJZDKNTz+Uf9bW1hqffklaPjZqExAQAODt+T/GxhDPvw/Vz5hr5+7urjZy1rp1a4PGN/b3Dv7tkNA5Z+XLl0d2djays7NV87Kzs2Fqqt4/KhQKlClTRuPG5kx63v+kOHbsWIPGHzx4sNr0kCFDDBpfV8o/TkpTpkwxaPz27durTXt4eBg0vq6MPX+xif38M2Z9+vSBXP72z6tcLkfv3r0NGt/YHzux/3ZUr15dbVqM04Yk05zVq1cPCoUCGzduREZGBh48eIAjR46gWbNmYqdGWurcubPqE49MJkOnTp0MGr9nz55q0+L4e70AACAASURBVD4+PgaNrytXV1fVJ15ra2uDX5R4woQJatPffvutQePrytjzF5vYzz9jVqJECbRr1w4ymQzt27dHiRIlDBrf2B87sf92LFq0SG164cKFBo0PSKg5Mzc3x+zZsxEbG4uRI0diwYIF8PX1RcOGDcVOjXSg/ARk6E8+SsrRM2MbNVMKCAiAXC4X7ZOvcvTJWEedjD1/sYn9/DNmffr0Qa1atQw+aqZk7I+d2H87lKNnYn3ZTjLf1tSXGzduoHr16qoh5k9NTk4Obt26JclvaxIVdcZ83pTYWDsqyj7NjuUdCoUCT548QUZGxif1jQ9BEJCRkYEnT57wfDsiIqJPiGS+rakvjo6OSEhIwIMHD1Rfbf5UmJqaws7ODqVKlRI7FSIiIiokn3xzJpfLVd/oJCIiIpK6T/6wJhEREZExYXNGREREJCFszoiIiIgkhM0ZERERkYSwOSMiIiKSEDZnRERERBLC5oyIiIhIQticEREREUkImzMiIiIiCWFzRkRERCQhbM6IiIiIJITNGREREZGEsDkjvfL394enpyemTJkidipGKSgoCJ6enggODhY7FaO0ZMkSeHp64ueffxY7FaM0YsQIeHp6YtSoUWKnYnTWr18PT09PbNiwQZT4xv7esX37dnh6eiI8PFyU+GI/fmzOSK+uXbsGALh69arImRinEydOAACOHz8ucibG6dChQwCAgwcPipyJcXry5AkA4NGjRyJnYnzCwsIAANu2bRMlvrG/d4SGhgIAfvvtN1Hii/34sTkjvfH391eb5uhZwQQFBalNG+snYLEsWbJEbZqjZwUzYsQItWmOnuXf+vXr1aYNPfpi7O8d27dvV5s29OiZ2I8fAMgEQRAMHpWKBE9PT415ERERhR4nJCQE0dHRud4XExMDAHBwcMhzfWdnZ40/RFJgiPp9qHYA66ergIAAAEBgYKBB4xYGsevH2hlvfF2Jnb/Y8QHA1KDRiAwsPT1d7BSMGutHRGR4bM7I6H1o1MaYP30bwsdGvFg/IiLD4zlnpDe1a9dWm65bt65ImRinli1bqk23atVKpEyMU/v27dWmPTw8RMrEOFWoUEFtumLFiiJlYnx8fX3Vpnv16mXQ+Mb+3jF48GC16SFDhhg0vtiPH8DmjPTo/ZNS58+fL1Imxun9L1RMnjxZpEyM04QJE9Smv/32W5EyMU4hISFq06tXrxYpE+MzaNAgtemBAwcaNL6xv3f07NlTbdrHx8eg8cV+/AA2Z6RnytEzjpppR/kJ2Ng++UqFcvSMo2baUY6ecdSs4JSjL2KMugDG/96hHD0z9KiZktiPH7+tSZ80njOlG9ZPN6yf9lg7Kso4ckZEREQkIWzOiIiIiCSEzRkRERGRhLA5IyIiIpIQNmdEREREEsLmjIiIiEhC2JwRERERSQibMyIiIiIJYXNGREREJCFszoiIiIgkhM0ZERERkYSwOSMiIiKSEDZnRERERBLC5oyIiIhIQticEREREUmIqdgJvC8nJwezZ89GlSpVMHjwYLHTIR317NkTaWlpsLa2xrZt28ROx+h4enqq/h8RESFiJsbJy8sLOTk5MDExwe7du8VOx+j06tULKSkpsLW1xZYtW8ROx6h06dIFgiBALpdjz549Bo9v7O8dw4YNQ2xsLBwcHLBmzRqDx+/atSuys7NhamqKXbt2GTy+5EbO/vzzT9y+fVvsNKiQpKWlAQBSUlJEzoSKopycHABAdna2yJkYJ+XrNjk5WeRMjI8gCAD+9xykgomNjQUAxMTEiBJf+Z6RlZUlSnxJNWe3b9/GuXPn0LhxY7FToULQs2dPtelevXqJlIlxeveTb27T9GFeXl5q0127dhUpE+P0/uu1T58+ImVifLp06aI2/f5zUd+M/b1j2LBhatPDhw83aPz33yu8vb0NGh+Q0GHNtLQ0rFy5Et9++y327duX53KpqalITU3VmK9QKKBQKPSZIhWQctRMiaNnZEjvj1hw9Kxg3n+9cvQs/5SjZkocPSsY5aiZkqFHz95/rxBj9EwyzdnatWvh7u6OKlWqfHC5iIgIhIeHa8z38fGBr6+vvtIjIiIiMghJNGfHjx9HYmIixo0b99FlPT094e7urjGfo2ZERET0KZBEc3bq1ClER0dj6NChAIA3b95AJpMhNjYWAQEBasvy8KXxsLKyUju0aW1tLWI2VNTI5XK1w0kmJiYiZmN8rK2t1Q5t2traipiNcZHJZGqHNuVySZ3eLXnlypVTO7Tp4OBg0PgmJiZqhzZNTQ3fKkniGTN16lRs2LABoaGhCA0NRfPmzeHh4aHRmJFx2b59u9o0L6VRMO9//d0Yvw4vpvcvX8BLaRTM+69XXkoj//bu3as2behLaRj7e8e6devUpg19KY333yt4KQ365FhZWQHgqBmJQzliwVEz7Shftxw1KziZTAaAo2baKleuHADDj5opKd8zxBg1AyRyWPN9Y8eOFTsFKiTvj55RwRjbJ16pEePin58SjnZr7/3RM0Mz9veO90fPDE3skXa29EREREQSwuaMiIiISELYnBERERFJCJszIiIiIgnJ9xcCrl27hgMHDiAuLg4BAQE4efIkypYtiyZNmugzPyIiIqIiJV8jZ2fPnsVPP/2E8uXLIyYmBtnZ2TAxMcHy5ctx7NgxPadIREREVHTkqzn7888/MXLkSPTt21d1zQ9PT0+MGjVKlIuzEREREX2q8tWcxcTE4PPPP9eYX716dcTFxRV6UkRERERFVb6as1KlSuHGjRsa8//++2+ULl260JMiIiIiKqry9YUAb29vrFmzBrGxscjOzsbFixfx7NkzHDp0SPVj5URERESku3w1Z23atEHx4sWxe/dumJmZISwsDI6OjpgwYQIaNWqk7xyJiIiIiox8X0rD1dUVrq6u+syFiIiIqMjLV3O2cePGD97fv3//QkmGiIiIqKjLV3N29+5dtens7Gw8e/YMaWlpaNasmV4SIyIiIiqK8tWczZw5M9f5oaGhquueEeXG09NT9f+IiAgRMzFOrJ9uWD/dsH7aE7t2YsfXldj5ix1fp9/W7Ny5M/7zn/8UVi5ERERERZ5OzdmdO3c4ckZ5eveTR27T9GGsn25YP92wftoTu3Zix9eV2PmLHR/I52HNKVOmaMxLT0/H06dP0aNHj0JPioiIiKioyldz5urqCplMppqWyWQwNzdHlSpVUKdOHb0lR0RERFTU5Ks58/X11XceRERERIQPNGcfu7bZu3idMyIiIqLCkecXAu7evZvvG1Fu3v/6sTF+nVtMrJ9uWD/dsH7aE7t2YsfXldj5ix0f+MDIWV7XNiMiIiIi/cn3dTCSkpIQFxeHrKws1byMjAzcuXMHPj4+ekmOjJ+xfWKTGtZPN6yfblg/7YldO7Hj60rs/MWOn6/m7MiRIwgJCYEgCGrz5XI5HB0d2ZwRERERFZJ8XYR2586d6NKlC9atWwcbGxssWrQI06dPR7ly5TBkyBB950hERERUZOSrOXv+/Dk6d+4Ma2trVK1aFU+ePEHt2rUxaNAgbN26Vd85EhERERUZ+WrOrK2tkZ6eDgBwcHDA/fv3AQCVKlVCdHS03pIjIiIiKmry1ZzVq1cPGzZsQEJCAlxcXHD27FkkJCTg1KlTsLe313eOREREREVGvpqzQYMGQSaT4fLly2jYsCHs7OwwduxYbNmyhV8GICIiIipE+fq25vTp09G8eXO4urpCLpdjxowZePjwIWxtbVGiRAl950hERERUZOSrOWvdujXOnj2LsLAwVK1aFS1atICbmxvs7Oz0nR8RERFRkZKv5szLywteXl6Ii4vD2bNncfz4caxfvx61a9dG8+bN0apVK33nSURERFQk5PsXAgCgTJky8Pb2RocOHXDw4EH8+eefuHr1KpszIiIiokKS7+YsPT0dly5dwrlz5/DPP/+gePHi6NixI1q0aKHP/IiIiIiKlHw1Z8HBwbh69SqKFSsGNzc3TJ8+HdWqVdN3bkRERERFTr6aMysrK/j5+aFu3bqQy/N19Q0iIiIi0kK+mrNx48bpOw8iIiIiQj4vQktEREREhlGgb2vq082bN/H777/j8ePHsLa2RqdOndClSxex0yIdeXp6qv4fEREhYibGifXTDeunG9ZPe2LXTuz4uhI7f7HjS6I5e/nyJQIDAzF06FA0b94cjx8/xty5c1GiRAk0bdpU7PSIiIiIDEYShzXj4+PRoEEDtGzZEnK5HJUqVUL9+vVx8+ZNsVMjHbz7ySO3afow1k83rJ9uWD/tiV07sePrSuz8xY4PSGTk7LPPPsP48eNV02/evMH169dz/VH11NRUpKamasxXKBRQKBR6zZPEERISgujoaK3WVa4XEBCgdXxnZ2eMGDFC6/XFpEvtANZP7PoZc+0AvnaJtCWJ5uxdr1+/xqJFi1CiRIlcL3AbERGB8PBwjfk+Pj7w9fU1RIpkYNHR0bh6/RagKFnwlTMFAMDVBwnaBU99rt16EqFT7QDWT8z6GXntAL52ibQlqebs+fPnCAwMhLW1NQICAmBiYqKxjKenJ9zd3TXmc9TsE6coCaG24b8gIru21+AxC51ItQNYP118ErUDWD8iLUimOYuOjkZgYCAaNWqEIUOGwNQ099R4+JKIiIg+ZZL4QkBiYiLmzZuHzp07Y/jw4Xk2ZmRc3v/6sTF+nVtMrJ9uWD/dsH7aE7t2YsfXldj5ix0fkMjI2YEDB/Dq1Sv88ccf+OOPP1TzW7RowZM5iYiIqEiRRHPWp08f9OnTR+w0SA+M7ROb1LB+umH9dMP6aU/s2okdX1di5y92fEkc1iQiIiKit9icEREREUkImzMiIiIiCWFzRkRERCQhbM6IiIiIJITNGREREZGEsDkjIiIikhA2Z0REREQSwuaMiIiISELYnBERERFJCJszIiIiIglhc0ZEREQkIWzOiIiIiCSEzRkRERGRhLA5IyIiIpIQNmdEREREEsLmjIiIiEhC2JwRERERSYip2AmQ8QsJCUF0dHSu98XExAAAHBwc8lzf2dkZI0aMyPP+mJgYIPUVZNf26paoNlKfIyYmQ2+b/1DtgMKp36eM9dONrvUryrUj0ic2Z6RX6enpYqdg1Fg/3bB+umH9iMTB5ox09qFPzgEBAQCAwMBArbfv4OCAhMwECLW7aL0Nbcmu7YWDQym9bf9jow6FUb9PGeunG9aPSJp4zhkRERGRhLA5IyIiIpIQNmdEREREEsLmjIiIiEhC2JwRERERSQibMyIiIiIJYXNGREREJCFszoiIiIgkhM0ZERERkYSwOSMiIiKSEDZnRERERBLC5oyIiIhIQticEREREUkImzMiIiIiCWFzRkRERCQhbM6IiIiIJITNGREREZGEsDkjIiIikhA2Z0REREQSYip2Au9KSEjAL7/8gps3b8La2hpfffUV2rVrJ3ZaRERERAYjqZGzxYsXo3z58li3bh0mTZqE33//Hbdu3RI7LSIiIiKDkUxzFhMTg+joaPTt2xfm5uZwdnaGm5sbTp48KXZqRERERAYjmcOajx8/RqlSpWBpaamaV6lSJVy4cEFtudTUVKSmpmqsr1AooFAo9J5nUTR+/Hg8e/ZMq3Vfv34NAPD19dU6fnZ2NpCZA9m1vQVfOf3l238t7bQLnvocQCnt1v3/xKyf8rUiO79eq/WRk/X2X7mWbxXZmYiJydBuXehWO6CQ6ieTa/fc01Xqc51qB4hfv/T0dAAmn2T93rx5g5ycHJ22L5fLYWFhkef9mZmZMDMzEy1+2bJlsWzZMq22/bHnnr7zT0tLAwBYWVmJEh/QrX6Sac7S09M1dtLS0vL/v7j/JyIiAuHh4Rrr+/j46NQAUN5evnyJtLTXMDM3L/jKMhkAIDMrW6vYmRkZsLAwR92a1bVaPzo6GQDgXFnbBqsUnJ2dtVz3LTHrBwAyAArL3N/gP+b167fNWTEt1wfMYGenZWMMHWsHFEr9jJmY9cvMyAAgAHIT7WJLwIfqlyMIEARBp+3nCEKetVXW701GJmCSy+svJwfQMX62kIPM9Mw87syE5cuXWm/7Y889fddPEAQIAFLy2j+J108yzZmFhQXevHmjNu/NmzcaDZunpyfc3d011ueomf44ODjAwqY4vL+eYPDYu9YuQUmbYggMDNRq/YCAAADQev3CwPppT8zaAcC6H/zwxqI4hNpdDB5bdm0vHBx0G7UV+7n3Mv4pUkxsWT8tGHv9ivJrF9C9fpJpzipUqID4+HikpaWphiEfPXoER0dHteV4+JKIiIg+ZZL5QoCjoyOcnJwQFhaG7Oxs3Lp1C6dPn0arVq3ETo2IiIjIYCQzcgYAEydOxKpVqzBo0CDY2Nhg0KBBqFatmthpERERERmMpJqzMmXKYObMmWKnQURERCQayRzWJCIiIiI2Z0RERESSwuaMiIiISELYnBERERFJCJszIiIiIglhc0ZEREQkIWzOiIiIiCSEzRkRERGRhLA5IyIiIpIQNmdEREREEsLmjIiIiEhC2JwRERERSQibMyIiIiIJYXNGREREJCFszoiIiIgkhM0ZERERkYSwOSMiIiKSEFOxEyDjkBD7BLvWLinwei+fxwMA7EqW1jpuSZvPtFpXSlg/7WlbO0D3+mVmZABZzyG7tler9ZH+8u2/lnYFXzf1OYBS2sV9h1j1S4h9AjMTOZCqZf10qR0gev0K47Vr7PUrsq9dQOf6sTmjj3J2dtZ63ZfxWQCAkjbFtFq/pM1nOsWXAtZPe7rmrmv93pQoDgBwcNDuTTY6OhkA4FxZm/VL6bz/YtavpM1nSE1NhbNCoVVs3WoHiF2/wnjtGnP9ivZrF9C1fjJBEASt1yb6iICAAABAYGBgkYyvK7HzFzu+rsTOX+z4uhIzf9bOuOPrSuz8xY7Pc86IiIiIJITNGREREZGEsDkjIiIikhA2Z0REREQSwuaMiIiISELYnBERERFJCJszIiIiIglhc0ZEREQkIWzOiIiIiCSEzRkRERGRhLA5IyIiIpIQNmdEREREEsLmjIiIiEhC2JwRERERSQibMyIiIiIJYXNGREREJCFszoiIiIgkhM0ZERERkYSwOSMiIiKSEFOxE1D6z3/+g927dyMpKQlly5ZFv379UK9ePbHTIiIiIjIoSYycRUZGYvPmzfj222/x22+/oXv37li4cCHi4+PFTo2IiIjIoCTRnCUmJqJr165wdnaGXC6Hm5sb7OzscPfuXbFTIyIiIjIomSAIgiECZWdnIz09Pdf7FAqF2vSjR48wefJkLF68GOXKlVO7LzU1Fampqblu4/3tkGGEhIQgOjo61/uU852dnfNc39nZGSNGjDDa+Lr4UO6A/vMXO76uxM5f7Pi60jV/vnb53NOW2PmLHf9jDHbOWVRUFObOnZvrfWFhYar/P3nyBIGBgfDy8tJozAAgIiIC4eHhGvN9fHzg6+tbeAlTobC0tCzS8XUldv5ix9eV2PmLHV9XYubP2hl3fF2Jnb/Y8Q02cpYf169fx8KFC+Hu7o6BAwfmugxHzoiIiOhTJpnm7NixY/j1118xePBgtGnTRux0iIiIiEQhiUtpXLlyBevWrcOUKVNQs2ZNsdMhIiIiEo0kRs6mT5+O27dvw9zcXG3+iBEj0KJFC5GyIiIiIjI8STRnRERERPSWJK5zRkRERERvsTkjIiIikhA2Z0REREQSwuaMiIiISELYnBERERFJCJszIiIiIglhc0aSlpqairCwsFx/sos+jvXTDeunG9ZPe6ydboy9fmzOSNJSU1MRHh5utC8wsbF+umH9dMP6aY+1042x14/NGREREZGEsDkjIiIikhA2Z0REREQSwuaMiIiISEJMZs2aNUvsJIg+xMzMDLVq1YK5ubnYqRgl1k83rJ9uWD/tsXa6Meb6yQRBEMROgoiIiIje4mFNIiIiIglhc0ZEREQkIWzOiIiIiCTEVOwE6NPk6+sLc3NzyGQyyGQymJiYoGLFiujRowfq16+vtmxgYCD++9//YsWKFbC3twcAZGdnY/r06bCxscGUKVPUlj937hxWrFiBoKAgODg45JlDamoqfvnlF0RGRsLCwgIdOnSAj49P4e+sHkihfo8ePUJoaCiio6NhYWGBFi1aoE+fPpDLpf+ZTgr1u3XrFjZs2IDHjx/Dzs4OX331FVq1alX4O1vIpFC7dy1duhTp6emYPHly4eygnkmhfteuXcMPP/ygdiK8t7e3Ubz/SaF+qamp+PXXX3H58mXI5XJ8+eWXGDp0KExNDdgyCUR60LNnT+HevXuq6ezsbGH//v1Cv379hOTkZNX8+Ph4YejQocKSJUuEbdu2qW0jJiZG6N+/v3D48GHVvBcvXghDhw4Vjh49+tEcFi9eLAQHBwtpaWlCbGysMGrUKOHkyZM675shiF2/zMxMYdSoUcIff/whZGVlCc+ePRMmTpwo7Ny5s1D2T9/Ert+bN2+EIUOGCIcPHxZycnKEZ8+eCSNGjBBu375dKPunT2LX7l3Hjx8XevfuLQQFBWm9P4Ymhfrt2rVLWLx4sc77IgYp1C84OFiYP3++kJycLCQnJwv+/v5CeHi4zvtWENL/CEyfBLlcDnd3d2RkZCAuLk41//Dhw3B1dUX79u1x6NAhZGVlqe4rX748Bg0ahA0bNiA+Ph4AEBISgnr16sHd3f2D8dLT03H+/Hn069cPxYoVQ9myZdG+fXucPHlSL/unb4au34sXL+Do6Ihu3brBxMQEZcqUgZubG27cuKGX/dM3Q9fP3NwcS5cuRdu2bSGTyZCSkoKMjAw8evRIL/unT4aundKzZ88QHh6Otm3bFur+GJoY9bt37x6cnJwKe1dEYej6JSYm4u+//8aIESNgY2MDGxsb+Pn5oUWLFnrZv7ywOSODSEtLw44dO2Bvb4+KFSsCeDv8fPToUXh4eKBmzZqws7PDmTNn1NZr164datWqhZCQEJw4cQKPHj3C8OHDPxrv6dOnAN6+SJUqVaqEx48fF+JeGY6h61e6dGlMnTpVdQgzJycHkZGRcHZ2LvydMwBD1w8ArK2tAQDDhw9HQEAASpQogUaNGhXujhmAGLXLzs7G0qVLMXjwYNja2hb6PhmSGPWLjo7GjRs3MHbsWIwePRobN25Ua16MiaHrd//+fdjb2+PMmTOq+h08eBAlS5bUy/7lheeckd7MmDFD9cfd1NQUVatWxffff686D+LSpUuwt7dHtWrVAAAdO3bE/v370bJlS7XtjBo1Cn5+frh9+zamT5+OYsWKfTR2enq66rwFJQsLC6SnpxfW7umdmPV7V1ZWFlavXo1Xr16hS5cuhbBnhiGV+q1cuRLPnz9HeHg4Xr16BRsbm0LYO/0Su3bbtm2Dk5MTvvjiC9y5c6cQ98wwxKxfdnY2SpYsiQYNGuC7775DYmIiFi1aBADo379/Ye6m3ohZv5SUFCQlJeHhw4dYuHAhXr16hcDAQBQrVgzdu3cv5D3NG5sz0ps5c+Z8cGj90KFDuHfvHnx9fdXm37p1C9WrV1dN29raok2bNrh//z4+++yzfMW2sLBARkYGBEFQNWhv3ryBhYVFwXdEJGLWTyk1NRWLFi3Cy5cvMXPmTFhZWRVofTFJoX7A26uUlytXDi4uLli/fr3GScpSJGbtoqKicPHiRQQGBmqVuxSIWT8TExPMmDFDNV2uXDn06NEDmzZtMprmTMz6mZmZITs7GwMGDECxYsVQrFgxeHp64tChQ2zO6NMXGxuL69evY+3atWqHLVatWoX9+/ervcCAt+cdFORbgsrDmU+ePIGjoyMA4PHjx6r/Gzt91w8A4uLiMG/ePDg6OsLPz6/AI0ZSpu/6PXnyBIGBgVi0aBHMzMwAAJmZmXjx4kXh7ICI9F27U6dOISEhASNHjgTwtm45OTnw8/PDwoULC2cnRKTv+iUmJmLfvn3o3bu36tuFmZmZRvXB9EMM9bcjMzNTNS87O9uw39QEzzkjkRw+fBh16tTROJ+kVatWOH/+PBITE3XafrFixdCwYUOEhYUhIyMDT548wV9//WUUlzLID33XLz09HXPnzkWdOnU+ucYM0H/9ypcvD7lcjm3btiE7OxsPHz7Erl270KRJE522KwX6rt3IkSPx+++/IzQ0FKGhofD29oarq+sn0ZgB+q+ftbU1jh07hh07diA7OxuxsbH4888/0aZNG522KxX6rp+TkxOqVKmC0NBQvH79GvHx8YiIiEDTpk112m5BsTkjg8vKysKxY8fQrFkzjftcXFxQsmRJHDx4UOc4I0aMQFZWFoYNG4ZZs2bBw8Mj15jGxhD1O3XqFGJjY3H8+HEMHDgQAwYMwIABA/Djjz/qtF0pMET95HI5/P39ce/ePQwdOhRBQUFo3769QQ+L6IOhXrufKkPUz9zcHFOnTkVUVBSGDh2KuXPnwt3dHR07dtRpu1JgqOfflClTYGpqivHjx2P69Olo1qyZwevHHz4nIiIikhCec0ZGa8OGDTh06FCu982ZMwdVqlQxcEbGhfXTDeunPdZON6yfboyhfhw5IyIiIpIQnnNGREREJCFszoiIiIgkhM0ZERERkYSwOSMiIiKSEH5bk4gKZOzYsYiPj1ebZ25ujkqVKmHIkCH4/PPPRcpMHElJSbh69arG7/rl5fHjx9iyZQtu376NjIwMlC9fHh06dEDr1q1Vy0RHRyMtLQ21a9fO1zYLujwRSRubMyIqsL59+6pdcTwhIQGbNm3CwoULsXjxYqP6DU5dbd68Ga9fv85Xc5aUlITZs2ejVatW6Nu3L8zMzHDt2jWsXbsWL168QI8ePQAAP/30E7y8vPLdbBV0eSKSNh7WJKICs7S0hK2trerm7OyM0aNH48WLF4iKihI7PYMqyNWILl26BEtLS/Tv3x8VKlRAmTJl0KZNG3Tq1AlHjx7VY5ZEZEw4ckZEhUL5w8omJiYAgJ07d+LAgQNITk5GpUqV0K9fP9XIEUgswwAACLtJREFUzqxZs1CjRg38+++/uHnzJkqXLo2BAwfiiy++AACkpKRgw4YNiIyMREpKCkqWLIlu3bqhbdu2OHPmDFauXIm1a9fC0tISwNsfaR8/fjwWL16MHTt2wNLSEq9evcKlS5dga2uLr7/+Gi9evFD91mqrVq0wePBgVe4fy7VOnTqIjo7GP//8g1KlSqF79+5o1aoVwsLCcPz4cQCAr68vwsLCPlijnJwcvHjxAvfu3VO70GW3bt3g4eGhihcfH49ff/0V586dw6xZs3D//n1s2rQJd+7cQWZmJipWrIjBgwejevXquS7v6+sLf39/NGjQAAAQFRWF2bNnY8OGDbC0tMThw4exa9cuJCYmokyZMujZs6fBfzuQiPLGkTMi0tnr16+xadMmFC9eHDVq1MDBgwcRERGBYcOGYcmSJWjbti2Cg4Px+PFj1Tq7d+9GnTp18NNPP8HNzQ0LFy5EQkICAGD58uV4+fIlZsyYgaVLl6Jx48YICQnBixcv0LBhQ5iYmODSpUuqbZ06dQpVq1aFg4MDgLc/jlylShUsWrQINWrUwM8//4wzZ85g5syZGDZsGPbt24ebN28CQL5y3blzJxo0aIAlS5bA1dUVq1evRmJiIry9vdG0aVM0aNAAa9eu/WidvvzyS1hZWSEgIADTpk1DWFgYbt68CSsrK5QpUwYAMHnyZJQsWRJ9+/bF5MmTkZGRgblz56JKlSoIDg5GcHAwrKyssHr16lyX/5jo6GisWbMGAwYMwLJly+Dt7Y2lS5ciNjb2o+sSkWGwOSOiAtu4cSMGDx6MwYMHY9CgQRg5ciSSk5Mxc+ZMWFlZYdeuXejXrx8aNmyI0qVLo127dvjiiy9w5MgR1TZq166Nbt26oVy5cujVqxccHR1VP1pcv359DBs2DI6OjqqRKkEQEBcXB3NzczRp0gSnT59Wbev06dNo0aKFatrBwQHe3t4oU6YMWrRogdevX2PAgAFwcHBA8+bNYWNjo2q+8pNr3bp10aZNG1Uu2dnZePz4MSwsLGBmZgZTU1PY2tp+tG62trYIDg5Ghw4dEB8fj/DwcMyYMQMTJkzArVu3AABWVlaQy+WwtLSElZUVXr9+DW9vb/Tu3RulS5eGg4MDOnTooGqm3l/+YxISEiCXy2Fvb48SJUrA3d0d06ZNy1f+RGQYPKxJRAXWrVs3tGzZEhkZGTh48CAuXLiAbt26wcHBAenp6arDbKGhoap1MjMzUa9ePdV0zZo11bZZtWpVVcPk4eGBixcvYt++fYiJicHdu3cBAFlZWQCAli1b4scff0RqaiqeP3+OmJgYNGvWTLWtsmXLqv5vbm4OAKqRKeDtOXNZWVn5zrV06dKq/yubGGUuBWVvb4+hQ4di6NChePToES5fvox9+/YhMDAQK1as0Giw7Ozs0LZtWxw9ehT//vsvnj59irt37yI7O1ur+PXr14ezszOmTZuGsmXLwtXVFS1btixSX+Igkjo2Z0RUYNbW1qpmZ+jQoUhOTkZQUBCCg4NV54GNGTMGzs7OauspGyUAkMvVB+4FQYCpqSlycnIwZ84cJCUlwc3NDW3atMHAgQMxceJE1bK1atWCnZ0dLly4gGfPnqFu3bpqIz/K897e9X48AKoGp6C5KvMtqJ07d8LJyQn169cHAFSsWBEVK1aEq6srJk2ahHv37qFWrVpq6yQmJsLf3x8VK1ZE/fr14ebmhqSkJKxYseKDsd7N791GztzcHHPnzsX169cRGRmJyMhIHDhwAN999x0aN25c4H0iosLHw5pEpLOvv/4aMpkMISEhUCgUKF68OBISElCmTBnV7fDhw4iMjFStc//+fbVt3L17F5UqVcLDhw9x/fp1fP/99+jVqxe+/PJLJCUlAfhfwyGTydCsWTOcP38ely9fVjukWRD5zfVDZDJZvuNFRUVhz549GvNtbGxU+bzv7NmzsLCwwPTp09G1a1fUq1cPz549A5B3g2hhYYH09HTV9Lvnk127dg3bt29HrVq10K9fPyxYsACfffYZrl69mu/9ICL9YnNGRDqztrZG//79ceXKFVy4cAFdu3bFH3/8gdOnTyMhIQH79+/Hnj17VCfsA2+bjiNHjiA2NhabN29GfHw8PDw8YGtrC7lcjrNnz+L58+eIjIzE/2vn7kEaCYMwjv+DIFiphSAqsqQISRpjYSEY7QQbRWKTJk0gmEaLYOEX2AVBFAXBxi8QRCEgFlqEWFqIjVGCCamNiKuQbjFwVxwsCn7c5by79Xh+9ewyvNW8O7Ozvr6Oy+Xi6enJfr63t5erqyvu7u7o6uqqOvefyfU9tbW1mKaJaZofxg4PD5PL5VheXqZQKPDw8EAul2NpaYmOjg4MwwB+FFelUolyuUxDQwOPj49cXFxwf3/PyckJR0dHAPZ5PI8HMAyDTCbD7e0tl5eXHB8f2znU1NSQSqVIp9OYpkk2m+Xm5gav1/srxyYif5DamiLyKfr6+shkMmxvb7O4uIhlWezs7FAul2lpaSGRSODxeOz47u5uTk9P2djYoLW1lampKbs1GYvFSKVS7O/v43a7iUajbG5uUiwW6ezsBKC9vZ3m5mYMw7DXeFRjYGDgw1zfEwwGOT8/Z3x8nNXVVerr69+M9fv9zM7OcnBwQDKZxLIsmpqaCAaDDA4O2nH9/f3s7u5yfX1NMpkkn8+zsrJCpVIhEAgwMzPD9PQ0xWIRv9//In5+fp5YLMba2hqJRIK2tjbC4TALCwsA+Hw+RkdHOTw8ZGtri8bGRkKhED09PVWfoYh8Lte3agYnRER+w9zcHG63m0gkUvU7KpUK8XiceDxu70cTEfkf6MuZiHw5Z2dnZLNZ6urq7OF6J5icnHyxH+25oaEhRkZG/nJGIvIVqTgTkS9nb28Py7IYGxt79U/Kf2ViYuLNFRuvDfuLiLxGbU0RERERB3HOlVNEREREVJyJiIiIOImKMxEREREHUXEmIiIi4iAqzkREREQc5Ds/Sx0bE2DIFAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Repayment = data[['PAY_0', 'PAY_2', 'PAY_3', 'PAY_4', 'PAY_5', 'PAY_6']]\n",
    "\n",
    "Repayment = pd.concat([y,Repayment],axis=1)\n",
    "Repayment = pd.melt(Repayment,id_vars=\"Default\",\n",
    "                    var_name=\"Repayment_Status\",\n",
    "                    value_name='value')\n",
    "\n",
    "plt.figure(figsize=(10,5))\n",
    "sns.set_context('notebook', font_scale=1.2)\n",
    "sns.boxplot(y=\"value\", x=\"Repayment_Status\", hue=\"Default\", data=Repayment, palette='Blues')\n",
    "plt.legend(loc='best', title= 'Default', facecolor='white')\n",
    "plt.xlim([-1.5,5.5])\n",
    "plt.title('REPAYMENT STATUS - BOXPLOT', size=14)\n",
    "plt.box(False)\n",
    "plt.savefig('ImageName', format='png', dpi=200);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It seems that PAY_0 (Repayment status in September) and PAY_2 (Repayment status in August)  have more discriminatory power the repayment status in other months."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Repayment = data[['BILL_AMT1','BILL_AMT2','BILL_AMT3','BILL_AMT4','BILL_AMT5','BILL_AMT6']]\n",
    "\n",
    "Repayment = pd.concat([y,Repayment],axis=1)\n",
    "Repayment = pd.melt(Repayment,id_vars=\"Default\",\n",
    "                    var_name=\"Repayment_Status\",\n",
    "                    value_name='value')\n",
    "\n",
    "plt.figure(figsize=(10,5))\n",
    "sns.set_context('notebook', font_scale=1.2)\n",
    "sns.boxplot(y=\"value\", x=\"Repayment_Status\", hue=\"Default\", data=Repayment, palette='Blues')\n",
    "plt.legend(loc='best', title= 'Default', facecolor='white')\n",
    "plt.xlim([-1.5,5.5])\n",
    "plt.title('REPAYMENT STATUS - BOXPLOT', size=14)\n",
    "plt.box(False)\n",
    "plt.savefig('ImageName', format='png', dpi=200);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='std'></a>\n",
    "### Standardizing and plotting the data\n",
    "The boxplot below reveals that features are in different scales and units. Many models use some form of distance when making predictions, and therefore, it is recommended to normalize the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## data are distributed in a wide range (below), need to be normalizded.\n",
    "plt.figure(figsize=(15,3))\n",
    "data_tmp = data.drop(['SEX', 'EDUCATION', 'MARRIAGE', 'PAY_0', 'PAY_2', 'PAY_3', 'PAY_4', \n",
    "               'PAY_5', 'PAY_6','Default'], axis=1)\n",
    "ax= data_tmp.boxplot(data_tmp.columns.name, rot=90)\n",
    "outliers = dict(markerfacecolor='b', marker='p')\n",
    "ax= data_tmp.boxplot(data_tmp.columns.name, rot=90, flierprops=outliers)\n",
    "plt.xticks(size=12)\n",
    "ax.set_ylim([-5000,250000])\n",
    "plt.box(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Standardization of data was performed; i.e, all features are centered around zero and have variance one. Features were plotted again, using a violin plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['LIMIT_BAL' 'AGE' 'BILL_AMT1' 'BILL_AMT2' 'BILL_AMT3' 'BILL_AMT4'\n",
      " 'BILL_AMT5' 'BILL_AMT6' 'PAY_AMT1' 'PAY_AMT2' 'PAY_AMT3' 'PAY_AMT4'\n",
      " 'PAY_AMT5' 'PAY_AMT6' 'Default']\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1080x216 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## data are distributed in a wide range (below), need to be normalizded.\n",
    "plt.figure(figsize=(15,3))\n",
    "data_tmp = data.drop(['SEX', 'EDUCATION', 'MARRIAGE', 'PAY_0', 'PAY_2', 'PAY_3', 'PAY_4', \n",
    "               'PAY_5', 'PAY_6'], axis=1)\n",
    "data_tmp_col = data_tmp.columns.values\n",
    "print(data_tmp_col)\n",
    "f, axes = plt.subplots(ncols=2, figsize=(15, 3), facecolor='white')\n",
    "\n",
    "data_bill_amount_col = ['BILL_AMT1','BILL_AMT2','BILL_AMT3','BILL_AMT4','BILL_AMT5','BILL_AMT6']\n",
    "data_bill_amount = data_tmp[data_bill_amount_col]\n",
    "ax1= data_bill_amount.boxplot(data_bill_amount.columns.name, rot=90, ax=axes[0])\n",
    "outliers = dict(markerfacecolor='b', marker='p')\n",
    "ax1= data_bill_amount.boxplot(data_bill_amount.columns.name, rot=90, flierprops=outliers, ax=axes[0])\n",
    "ax1.set_ylim([-100000,170000])\n",
    "\n",
    "data_pay_amount_col = ['PAY_AMT1','PAY_AMT2','PAY_AMT3','PAY_AMT4','PAY_AMT5','PAY_AMT6']\n",
    "data_pay_amount = data_tmp[data_pay_amount_col]\n",
    "ax2= data_pay_amount.boxplot(data_pay_amount.columns.name, rot=90, ax=axes[1])\n",
    "outliers = dict(markerfacecolor='b', marker='p')\n",
    "ax2= data_pay_amount.boxplot(data_pay_amount.columns.name, rot=90, flierprops=outliers, ax=axes[1])\n",
    "ax2.set_ylim([-5000,13000])\n",
    "\n",
    "plt.xticks(size=12)\n",
    "plt.box(False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "stdX = (features - features.mean()) / (features.std())              # standardization\n",
    "data_st = pd.concat([y,stdX.iloc[:,:]],axis=1)\n",
    "data_st = pd.melt(data_st,id_vars=\"Default\",\n",
    "                    var_name=\"features\",\n",
    "                    value_name='value')\n",
    "plt.figure(figsize=(20,10))\n",
    "sns.set_context('notebook', font_scale=1)\n",
    "sns.violinplot(y=\"value\", x=\"features\", hue=\"Default\", data=data_st,split=True, \n",
    "               inner=\"quart\", palette='Blues')\n",
    "plt.legend(loc=4, title= 'Default', facecolor='white')\n",
    "plt.ylim([-3,3])\n",
    "plt.title('STANDARDIZED FEATURES - VIOLIN PLOT', size=14)\n",
    "plt.box(False)\n",
    "plt.savefig('ImageName', format='png', dpi=200, transparent=False);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='corr'></a>\n",
    "### Correlation\n",
    "A correlation matrix of all variables is shown in the heatmap below. The only feature with a notable positive correlation with the dependent variable ‘Default’ is re-payment status during the last month (September). The highest negative correlation with default occurs with Limit_Balance, indicating that customers with lower limit balance are more likely to default. It can also be observed that some variables are highly correlated to each other, that is the case of the amount of bill statement and the repayment status in different months.\n",
    "\n",
    "Looking at correlations matrix, defined via Pearson function. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#  looking at correlations matrix, defined via Pearson function  \n",
    "corr = data.corr() # .corr is used to find corelation\n",
    "f,ax = plt.subplots(figsize=(8, 7))\n",
    "sns.heatmap(corr, cbar = True,  square = True, annot = False, fmt= '.1f', \n",
    "            xticklabels= True, yticklabels= True\n",
    "            ,cmap=\"YlGnBu\", linewidths=.5, ax=ax)\n",
    "plt.title('Pearson积差相关系数 - 热力图', size=18);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The heatmat shows that features are correlated with each other (collinearity), such us like PAY_0,2,3,4,5,6 and BILL_AMT1,2,3,4,5,6. In those cases, the correlation is positive.\n",
    "\n",
    "Uncorrelated data are poentially more useful: discriminatory!\n",
    "\n",
    "**What do correlations mean?**<br>\n",
    "\n",
    "Lets separately fit correlated and uncorrelated data via linear regression: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Uncorrelated data are poentially more useful: discrimentory!\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 405.725x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 405.725x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(x='LIMIT_BAL', y= 'BILL_AMT2', data = data, hue ='Default', \n",
    "           palette='coolwarm')\n",
    "plt.title('Linear Regression: distinguishing between Default and Non-default', size=16)\n",
    "\n",
    "\n",
    "sns.lmplot(x='BILL_AMT1', y= 'BILL_AMT2', data = data, hue ='Default', \n",
    "           palette='coolwarm')\n",
    "plt.title('Linear Regression: Cannot distinguish between Default and Non-default', size=16);\n",
    "\n",
    "print('Uncorrelated data are poentially more useful: discrimentory!')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='ml'></a>\n",
    "## Machine Learning: Classification models\n",
    "\n",
    "The classification models used for this analysis are: Logistic Regression, Decision Tree and Random Forest Classifier.<br>\n",
    "\n",
    "To build machine learning models the original data was divided into features (X) and target (y) and then split into train (80%) and test (20%) sets. Thus, the algorithms would be trained on one set of data and tested out on a completely different set of data (not seen before by the algorithm).\n",
    "\n",
    "<a id='sp'></a>\n",
    "### Spliting the data into train and test sets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Original dataset\n",
    "X = data.drop('Default', axis=1)  \n",
    "y = data['Default']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,y, test_size=0.2, stratify=y, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dataset with standardized features\n",
    "Xstd_train, Xstd_test, ystd_train, ystd_test = train_test_split(stdX,y, test_size=0.2, stratify=y,\n",
    "                                                                random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='fs'></a>\n",
    "### Feature Selection\n",
    "\n",
    "#### Recursive Feature Elimination\n",
    "Recursive Feature Elimination (RFE) is based on the idea to repeatedly construct a model and choose either the best or worst performing feature, setting the feature aside and then repeating the process with the rest of the features. This process is applied until all features in the dataset are exhausted. The goal of RFE is to select features by recursively considering smaller and smaller sets of features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Std Model Feature Ranking: [ 7 15  9  8 10  1  6  4 18 11 21  1  5  2 20 13 19  3  1 14 12 16 17]\n",
      "Standardized Model Score with selected features is: 0.812600 (0.000000)\n"
     ]
    }
   ],
   "source": [
    "NUM_FEATURES = 3\n",
    "model = LogisticRegression()\n",
    "rfe_stand = RFE(model, NUM_FEATURES)\n",
    "fit_stand = rfe_stand.fit(stdX, y)\n",
    "#print(\"St Model Num Features:\", fit_stand.n_features_)\n",
    "#print(\"St Model Selected Features:\", fit_stand.support_)\n",
    "print(\"Std Model Feature Ranking:\", fit_stand.ranking_)\n",
    "# calculate the score for the selected features\n",
    "score_stand = rfe_stand.score(stdX,y)\n",
    "print(\"Standardized Model Score with selected features is: %f (%f)\" % (score_stand.mean(), score_stand.std()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Most important features (RFE): ['PAY_0' 'BILL_AMT1' 'PAY_AMT2']\n"
     ]
    }
   ],
   "source": [
    "feature_names = np.array(features.columns)\n",
    "print('Most important features (RFE): %s'% feature_names[rfe_stand.support_])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Repayment status in September (PAY_0)\n",
    "* Amount of bill statement in September (BILL_AMT1)\n",
    "* Amount of previous payments in August (PAY_AMT2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dataset with three most important features\n",
    "Ximp = stdX[['PAY_0', 'BILL_AMT1', 'PAY_AMT2']]\n",
    "X_tr, X_t, y_tr, y_t = train_test_split(Ximp,y, test_size=0.2, stratify=y, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='lr1'></a>\n",
    "### Logistic Regression (original data)\n",
    "\n",
    "Logistic Regression is one of the simplest algorithms which estimates the relationship between one dependent binary variable and independent variables, computing the probability of occurrence of an event. The regulation parameter C controls the trade-off between increasing complexity (overfitting) and keeping the model simple (underfitting). For large values of C, the power of regulation is reduced and the model increases its complexity, thus overfitting the data.<br>\n",
    "\n",
    "The parameter ‘C’ was tuned using RandomizedSearchCV( ) for the different datasets: original, standardized and with most important features. Once the parameter ‘C’ was defined for each dataset, the logistic regression model initiated and then fitted to the training data, as it was described in the methodology."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tuned Logistic Regression Parameters: {'C': 0.0007196856730011522}\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    }
   ],
   "source": [
    "# Setup the hyperparameter grid, (not scaled data)\n",
    "param_grid = {'C': np.logspace(-5, 8, 15)}\n",
    "\n",
    "# Instantiate a logistic regression classifier\n",
    "logreg = LogisticRegression()\n",
    "\n",
    "# Instantiate the RandomizedSearchCV object\n",
    "logreg_cv = RandomizedSearchCV(logreg,param_grid , cv=5, random_state=0)\n",
    "\n",
    "# Fit it to the data\n",
    "logreg_cv.fit(X_train, y_train)\n",
    "\n",
    "# Print the tuned parameters and score\n",
    "print(\"Tuned Logistic Regression Parameters: {}\".format(logreg_cv.best_params_)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.7786666666666666\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n",
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.78      1.00      0.88      4673\n",
      "           1       0.00      0.00      0.00      1327\n",
      "\n",
      "    accuracy                           0.78      6000\n",
      "   macro avg       0.39      0.50      0.44      6000\n",
      "weighted avg       0.61      0.78      0.68      6000\n",
      "\n",
      "\n",
      "Average 5-Fold CV Score: 0.7787 , Standard deviation: 0.0001\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Python/3.7/site-packages/sklearn/linear_model/_logistic.py:940: ConvergenceWarning: lbfgs failed to converge (status=1):\n",
      "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n",
      "\n",
      "Increase the number of iterations (max_iter) or scale the data as shown in:\n",
      "    https://scikit-learn.org/stable/modules/preprocessing.html\n",
      "Please also refer to the documentation for alternative solver options:\n",
      "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n",
      "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "LR = LogisticRegression(C=0.00005, random_state=0)\n",
    "LR.fit(X_train, y_train)\n",
    "y_pred = LR.predict(X_test)\n",
    "print('Accuracy:', metrics.accuracy_score(y_pred,y_test))\n",
    "\n",
    "## 5-fold cross-validation \n",
    "cv_scores =cross_val_score(LR, X, y, cv=5)\n",
    "\n",
    "# Print the 5-fold cross-validation scores\n",
    "print()\n",
    "print(classification_report(y_test, y_pred))\n",
    "print()\n",
    "print(\"Average 5-Fold CV Score: {}\".format(round(np.mean(cv_scores),4)),\n",
    "      \", Standard deviation: {}\".format(round(np.std(cv_scores),4)))\n",
    "\n",
    "plt.figure(figsize=(4,3))\n",
    "ConfMatrix = confusion_matrix(y_test,LR.predict(X_test))\n",
    "sns.heatmap(ConfMatrix,annot=True, cmap=\"Blues\", fmt=\"d\", \n",
    "            xticklabels = ['Non-default', 'Default'], \n",
    "            yticklabels = ['Non-default', 'Default'])\n",
    "plt.ylabel('True label')\n",
    "plt.xlabel('Predicted label')\n",
    "plt.title(\"Confusion Matrix - Logistic Regression\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The model has not power predicting default credit cards. However, it can be observed that the average accuracy of the model is about 78%, which demonstrates that this metrics is not appropriate for the evaluation of this problem. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='lr2'></a>\n",
    "### Logistic Regression (standardized features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tuned Logistic Regression Parameters with standardized features: {'C': 2275.845926074791}\n"
     ]
    }
   ],
   "source": [
    "# Instantiate the RandomizedSearchCV object:\n",
    "logreg_cv_std = RandomizedSearchCV(logreg,param_grid , cv=5, random_state=0)\n",
    "\n",
    "# Fit it to the standardized data\n",
    "logreg_cv_std.fit(Xstd_train, ystd_train)\n",
    "\n",
    "# Print the tuned parameters \n",
    "print(\"Tuned Logistic Regression Parameters with standardized features: {}\".format(logreg_cv_std.best_params_)) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.8078333333333333\n",
      "\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.82      0.97      0.89      4673\n",
      "           1       0.69      0.24      0.36      1327\n",
      "\n",
      "    accuracy                           0.81      6000\n",
      "   macro avg       0.75      0.60      0.62      6000\n",
      "weighted avg       0.79      0.81      0.77      6000\n",
      "\n",
      "\n",
      "Average 5-Fold CV Score: 0.8099 , Standard deviation: 0.0048\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "LRS = LogisticRegression(C=3.73, random_state=0)\n",
    "LRS.fit(Xstd_train, ystd_train)\n",
    "y_pred = LRS.predict(Xstd_test)\n",
    "print('Accuracy:', metrics.accuracy_score(y_pred,ystd_test))\n",
    "\n",
    "## 5-fold cross-validation \n",
    "cv_scores =cross_val_score(LRS, stdX, y, cv=5)\n",
    "\n",
    "# Print the 5-fold cross-validation scores\n",
    "print()\n",
    "print(classification_report(ystd_test, y_pred))\n",
    "print()\n",
    "print(\"Average 5-Fold CV Score: {}\".format(round(np.mean(cv_scores),4)),\n",
    "      \", Standard deviation: {}\".format(round(np.std(cv_scores),4)))\n",
    "\n",
    "plt.figure(figsize=(4,3))\n",
    "ConfMatrix = confusion_matrix(ystd_test,LRS.predict(Xstd_test))\n",
    "sns.heatmap(ConfMatrix,annot=True, cmap=\"Blues\", fmt=\"d\", \n",
    "            xticklabels = ['Non-default', 'Default'], \n",
    "            yticklabels = ['Non-default', 'Default'])\n",
    "plt.ylabel('True label')\n",
    "plt.xlabel('Predicted label')\n",
    "plt.title(\"Confusion Matrix - Logistic Regression with standardized data\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The performance of the logistic regression model standardized data improved compared to the model built with the original dataset. By using the standardized dataset, the model is able to predict defaults; however, with a very low recall (0.24)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='lr3'></a>\n",
    "### Logistic Regression (most important features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "LR_imp = LogisticRegression(C=3.73, random_state=0)\n",
    "LR_imp.fit(X_tr, y_tr)\n",
    "y_pred = LR_imp.predict(X_t)\n",
    "print('Accuracy:', metrics.accuracy_score(y_pred,y_t))\n",
    "\n",
    "## 5-fold cross-validation \n",
    "cv_scores =cross_val_score(LR_imp, Ximp, y, cv=5)\n",
    "\n",
    "# Print the 5-fold cross-validation scores\n",
    "print()\n",
    "print(classification_report(y_t, y_pred))\n",
    "print()\n",
    "print(\"Average 5-Fold CV Score: {}\".format(round(np.mean(cv_scores),4)),\n",
    "      \", Standard deviation: {}\".format(round(np.std(cv_scores),4)))\n",
    "\n",
    "plt.figure(figsize=(4,3))\n",
    "ConfMatrix = confusion_matrix(y_t,LR_imp.predict(X_t))\n",
    "sns.heatmap(ConfMatrix,annot=True, cmap=\"Blues\", fmt=\"d\", \n",
    "            xticklabels = ['Non-default', 'Default'], \n",
    "            yticklabels = ['Non-default', 'Default'])\n",
    "plt.ylabel('True label')\n",
    "plt.xlabel('Predicted label')\n",
    "plt.title(\"Confusion Matrix - Logistic Regression (most important features)\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When using just important features for building the model, there is a slight improvement in performance with respect to the previous model. Therefore, taking into account three features only, the model has the same predictive power than using 24 features. Thus, future strategies should be focused on: repayment status in September (PAY_0), amount of bill statement in September (BILL_AMT1), and amount of previous payments in August (PAY_AMT2)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='tree'></a>\n",
    "### Decision Tree Classifier\n",
    "Decision Tree is another very popular algorithm for classification problems because it is easy to interpret and understand. An internal node represents a feature, the branch represents a decision rule, and each leaf node represents the outcome. Some advantages of decision trees are that they require less data preprocessing, i.e., no need to normalize features. However, noisy data can be easily overfitted and results in biased results when the data set is imbalanced."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Setup the parameters and distributions to sample from: param_dist\n",
    "param_dist = {\"max_depth\": [1,2,3,4,5,6,7,8,9],\n",
    "              \"max_features\": [1,2,3,4,5,6,7,8,9],\n",
    "              \"min_samples_leaf\": [1,2,3,4,5,6,7,8,9],\n",
    "              \"criterion\": [\"gini\", \"entropy\"]}\n",
    "\n",
    "# Instantiate a Decision Tree classifier: tree\n",
    "tree = DecisionTreeClassifier()\n",
    "\n",
    "# Instantiate the RandomizedSearchCV object: tree_cv\n",
    "tree_cv = RandomizedSearchCV(tree, param_distributions=param_dist, cv=5, random_state=0)\n",
    "\n",
    "# Fit it to the data\n",
    "tree_cv.fit(X_train, y_train)\n",
    "\n",
    "# Print the tuned parameters and score\n",
    "print(\"Tuned Decision Tree Parameters: {}\".format(tree_cv.best_params_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Tree = DecisionTreeClassifier(criterion= 'gini', max_depth= 7, \n",
    "                                     max_features= 9, min_samples_leaf= 2, \n",
    "                                     random_state=0)\n",
    "Tree.fit(X_train, y_train)\n",
    "y_pred = Tree.predict(X_test)\n",
    "print('Accuracy:', metrics.accuracy_score(y_pred,y_test))\n",
    "\n",
    "## 5-fold cross-validation \n",
    "cv_scores =cross_val_score(Tree, X, y, cv=5)\n",
    "\n",
    "# Print the 5-fold cross-validation scores\n",
    "print()\n",
    "print(classification_report(y_test, y_pred))\n",
    "print()\n",
    "print(\"Average 5-Fold CV Score: {}\".format(round(np.mean(cv_scores),4)), \n",
    "      \", Standard deviation: {}\".format(round(np.std(cv_scores),4)))\n",
    "\n",
    "plt.figure(figsize=(4,3))\n",
    "ConfMatrix = confusion_matrix(y_test,Tree.predict(X_test))\n",
    "sns.heatmap(ConfMatrix,annot=True, cmap=\"Blues\", fmt=\"d\", \n",
    "            xticklabels = ['Non-default', 'Default'], \n",
    "            yticklabels = ['Non-default', 'Default'])\n",
    "plt.ylabel('True label')\n",
    "plt.xlabel('Predicted label')\n",
    "plt.title(\"Confusion Matrix - Decision Tree\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The performance of the decision tree model improved compared to the logistic regression model showed previously. However, the recall is still low (0.33)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='rf'></a>\n",
    "### Random Forest Classifier\n",
    "Random forest classifier is comprised of multiple decision trees. It creates different random subset of decision trees from the training set as its predictors and selects the best solution by means of voting. As a result, the Random Forest model avoids overfitting problems. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create the random grid\n",
    "param_dist = {'n_estimators': [50,100,150,200,250],\n",
    "               \"max_features\": [1,2,3,4,5,6,7,8,9],\n",
    "               'max_depth': [1,2,3,4,5,6,7,8,9],\n",
    "               \"criterion\": [\"gini\", \"entropy\"]}\n",
    "\n",
    "rf = RandomForestClassifier()\n",
    "\n",
    "rf_cv = RandomizedSearchCV(rf, param_distributions = param_dist, \n",
    "                           cv = 5, random_state=0, n_jobs = -1)\n",
    "\n",
    "rf_cv.fit(X, y)\n",
    "\n",
    "print(\"Tuned Random Forest Parameters: %s\" % (rf_cv.best_params_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "Ran = RandomForestClassifier(criterion= 'gini', max_depth= 6, \n",
    "                                     max_features= 5, n_estimators= 150, \n",
    "                                     random_state=0)\n",
    "Ran.fit(X_train, y_train)\n",
    "y_pred = Ran.predict(X_test)\n",
    "print('Accuracy:', metrics.accuracy_score(y_pred,y_test))\n",
    "\n",
    "## 5-fold cross-validation \n",
    "cv_scores =cross_val_score(Ran, X, y, cv=5)\n",
    "\n",
    "# Print the 5-fold cross-validation scores\n",
    "print()\n",
    "print(classification_report(y_test, y_pred))\n",
    "print()\n",
    "print(\"Average 5-Fold CV Score: {}\".format(round(np.mean(cv_scores),4)),\n",
    "      \", Standard deviation: {}\".format(round(np.std(cv_scores),4)))\n",
    "\n",
    "plt.figure(figsize=(4,3))\n",
    "ConfMatrix = confusion_matrix(y_test,Ran.predict(X_test))\n",
    "sns.heatmap(ConfMatrix,annot=True, cmap=\"Blues\", fmt=\"d\", \n",
    "            xticklabels = ['Non-default', 'Default'], \n",
    "            yticklabels = ['Non-default', 'Default'])\n",
    "plt.ylabel('True label')\n",
    "plt.xlabel('Predicted label')\n",
    "plt.title(\"Confusion Matrix - Random Forest\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='sum'></a>\n",
    "## Comparison of model performance\n",
    "The metrics used to evaluate performance of the different models: accuracy, precision, recall, f1-score, AUC (ROC), and confusion matrix were employed."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='roc'></a>\n",
    "### Receiver operating characteristic (ROC) Curve\n",
    "Receiver Operating Characteristic (ROC) curve is a plot of the true positive rate against the false positive rate. The AUC is the Area Under Curve. If the AUC is high, the model is better distinguishing between positive and negative class. The ROC curve is plotted with “True Positive Rate” or Recall (on the y-axis) against the “False Positive Rate” (on the x-axis). When the AUC is 0.5 means that the model has no discrimination capacity to distinguish between positive and negative class.\n",
    "\n",
    "The Receiver operating characteristic (ROC) Curve with the respective area under the curve (AUC) are shown below for each model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_proba_RF = Ran.predict_proba(X_test)[::,1]\n",
    "fpr1, tpr1, _ = metrics.roc_curve(y_test,  y_pred_proba_RF)\n",
    "auc1 = metrics.roc_auc_score(y_test, y_pred_proba_RF)\n",
    "\n",
    "y_pred_proba_DT = Tree.predict_proba(X_test)[::,1]\n",
    "fpr2, tpr2, _ = metrics.roc_curve(y_test,  y_pred_proba_DT)\n",
    "auc2 = metrics.roc_auc_score(y_test, y_pred_proba_DT)\n",
    "\n",
    "y_pred_proba_LR = LR.predict_proba(X_test)[::,1]\n",
    "fpr3, tpr3, _ = metrics.roc_curve(y_test,  y_pred_proba_LR)\n",
    "auc3 = metrics.roc_auc_score(y_test, y_pred_proba_LR)\n",
    "\n",
    "y_pred_proba_LRS = LRS.predict_proba(Xstd_test)[::,1]\n",
    "fpr4, tpr4, _ = metrics.roc_curve(ystd_test,  y_pred_proba_LRS)\n",
    "auc4 = metrics.roc_auc_score(ystd_test, y_pred_proba_LRS)\n",
    "\n",
    "y_pred_proba_LRimp = LR_imp.predict_proba(X_t)[::,1]\n",
    "fpr5, tpr5, _ = metrics.roc_curve(y_t,  y_pred_proba_LRimp)\n",
    "auc5 = metrics.roc_auc_score(y_t, y_pred_proba_LRimp)\n",
    "\n",
    "plt.figure(figsize=(10,7))\n",
    "plt.plot([0, 1], [0, 1], 'k--')\n",
    "plt.plot(fpr1,tpr1,label=\"Random Forest, auc=\"+str(round(auc1,2)))\n",
    "plt.plot(fpr2,tpr2,label=\"Decision Tree, auc=\"+str(round(auc2,2)))\n",
    "plt.plot(fpr3,tpr3,label=\"LogReg, auc=\"+str(round(auc3,2)))\n",
    "plt.plot(fpr4,tpr4,label=\"LogReg(std), auc=\"+str(round(auc4,2)))\n",
    "plt.plot(fpr5,tpr5,label=\"LogReg(Std&Imp), auc=\"+str(round(auc5,2)))\n",
    "plt.legend(loc=4, title='Models', facecolor='white')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC', size=15)\n",
    "plt.box(False)\n",
    "plt.savefig('ImageName', format='png', dpi=200, transparent=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The highest AUC is obtained for the Random Forest Classifier model, with a value of 0.77. This means there is 77% chance that the model will be able to distinguish between default class and non-default class."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='ac'></a>\n",
    "### Mean Accuracy (coss-validation)\n",
    "Accuracy is the ratio of correctly predicted observation to the overall observations and it is one of the most intuitive measurements of performance. However, a high accuracy rate does not always mean we have a perfect model. In fact, it only works well when the datasets are symmetric. It can be misleading when classes are imbalanced.\n",
    "\n",
    "Using K-fold cross-validation it is possible to obtain less biased models and avoid overfitting the data. In this case it was used a 5-fold cross-validation, as shown in the code below.\n",
    "\n",
    "cv_scores = cross_val_score(Model, X, y, cv=5)\n",
    "\n",
    "After cross-validation there are five values of accuracy, so it was calculated the mean and standard deviation of all results."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Append different models\n",
    "models = []\n",
    "\n",
    "# Logistic Regression\n",
    "models.append(('LogReg',\n",
    "               LogisticRegression(C=3.73, random_state=0),'none'))\n",
    "\n",
    "# Logistic Regression (with standardized data)\n",
    "models.append(('LogReg(Std)',\n",
    "               LogisticRegression(C=3.73, random_state=0),'Std'))\n",
    "\n",
    "# Logistic Regression with standardized and important features\n",
    "models.append(('LogReg(Std&Imp)',\n",
    "               LogisticRegression(C=3.73, random_state=0),'imp'))\n",
    "\n",
    "# Decision Tree\n",
    "models.append(('Decision Tree', \n",
    "              DecisionTreeClassifier(criterion= 'entropy', max_depth= 4, \n",
    "                                     max_features= 7, min_samples_leaf= 8, \n",
    "                                     random_state=0),'none'))\n",
    "\n",
    "# Random Forest Classifier\n",
    "models.append(('Random Forest', \n",
    "              RandomForestClassifier(criterion= 'gini', max_depth= 6, \n",
    "                                     max_features= 5, n_estimators= 150, \n",
    "                                     random_state=0), 'none'))\n",
    "\n",
    "# Evaluate each model\n",
    "results = []\n",
    "names = []\n",
    "scoring = 'accuracy'\n",
    "\n",
    "for name, model, Std in models:\n",
    "    if Std == 'Std':\n",
    "        cv_results = cross_val_score(model, stdX, y, cv=5, scoring=scoring)\n",
    "        results.append(cv_results)\n",
    "        names.append(name)\n",
    "        msg = \"%s: %f (%f)\" % (name, cv_results.mean(), cv_results.std())\n",
    "        print(msg)       \n",
    "    elif Std == 'none':\n",
    "        cv_results = cross_val_score(model, X, y, cv=5, scoring=scoring)\n",
    "        results.append(cv_results)\n",
    "        names.append(name)\n",
    "        msg = \"%s: %f (%f)\" % (name, cv_results.mean(), cv_results.std())\n",
    "        print(msg)\n",
    "    else:\n",
    "        cv_results = cross_val_score(model, Ximp, y, cv=5, scoring=scoring)\n",
    "        results.append(cv_results)\n",
    "        names.append(name)\n",
    "        msg = \"%s: %f (%f)\" % (name, cv_results.mean(), cv_results.std())\n",
    "        print(msg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib import pyplot\n",
    "# Plot all the accuracy results vs. each model \n",
    "#(model type on the x-axis and accuracy on the y-axis).\n",
    "fig = pyplot.figure(figsize=(10,5))\n",
    "sns.set_context('notebook', font_scale=1.1)\n",
    "fig.suptitle('Algorithm Comparison - Accuracy (cv=5)')\n",
    "ax = fig.add_subplot(111)\n",
    "pyplot.boxplot(results, showmeans=True)\n",
    "ax.set_xticklabels(names)\n",
    "ax.set_ylabel('Accuracy')\n",
    "ax.set_ylim([0.75,1])\n",
    "plt.box(False)\n",
    "plt.savefig('ImageName', format='png', dpi=200, transparent=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The best accuracy is obtained for the Random Forest Classifier with a mean accuracy of 0.82, yet it is the model with higher variation (0.0096). In general, all models have comparable mean accuracy. Nevertheless, because the classes are imbalanced (the proportion of non-default credit cards is higher than default) this metric is misleading. Furthermore, accuracy does not consider the rate of false positives (non-default credits cards that were predicted as default) and false negatives (default credit cards that were incorrectly predicted as non-default). Both cases have negative impact on the bank, since false positives leads to unsatisfied customers and false negatives leads to financial loss. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='m'></a>\n",
    "### Precision, Recall, F1-score\n",
    "The precision of a model is the ratio TP / (TP + FP). In this case, it is the ability of the classifier not to label as positive a sample that is negative. Precision is a good metric to use when the costs of false positive (FP) is high.\n",
    "\n",
    "The recall of a model is the ratio TP / (TP + FN). In this case, it is the ability of the classifier to find all the positive class. Recall is a good metric to use when the cost associated with false negative (FN) is high. In this classification problem there is a high cost for the bank when a default credit card is predicted as non-default, since no actions can be taken. Thus, recall is one important metric to pay attention to.\n",
    "\n",
    "F1-score is a weighted average of precision and recall. Thus, it considers FP and FN. This metric is very useful when we have uneven class distribution, as it seeks a balance between precision and recall."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       Model               Data        Precision Recall  F1 \n",
      "------------------- ------------------ --------- ------ ----\n",
      "Logistic Regression       Standardized      0.79   0.81 0.77\n",
      "Logistic Regression Important features      0.79   0.81 0.78\n",
      "      Decision Tree           original       0.8   0.82 0.79\n",
      "      Random Forest           original       0.8   0.82  0.8\n"
     ]
    }
   ],
   "source": [
    "from astropy.table import Table, Column\n",
    "data_rows = [('Logistic Regression', 'Standardized', 0.79, 0.81, 0.77),\n",
    "              ('Logistic Regression', 'Important features', 0.79, 0.81, 0.78),\n",
    "              ('Decision Tree', 'original', 0.80, 0.82, 0.79),\n",
    "             ('Random Forest', 'original', 0.80, 0.82, 0.80)\n",
    "            ]\n",
    "t = Table(rows=data_rows, names=('Model', 'Data', 'Precision', 'Recall', 'F1'))\n",
    "print(t)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}