SVM-2.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\sklearn\\externals\\six.py:31: DeprecationWarning: The module is deprecated in version 0.21 and will be removed in version 0.23 since we've dropped support for Python 2.7. Please rely on the official version of six (https://pypi.org/project/six/).\n",
      "  \"(https://pypi.org/project/six/).\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np #linear algerbra\n",
    "import pandas as pd #data processing, csv file r/o\n",
    "\n",
    "from sklearn.utils import shuffle\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import classification_report,confusion_matrix, roc_curve, auc, precision_recall_curve, roc_curve\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.model_selection import StratifiedKFold\n",
    "from sklearn.feature_selection import RFECV\n",
    "from sklearn.datasets import make_classification\n",
    "\n",
    "from sklearn import discriminant_analysis\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "from imblearn.combine import SMOTEENN\n",
    "\n",
    "from sklearn.metrics import confusion_matrix\n",
    "import itertools\n",
    "from itertools import cycle\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Time</th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Amount</th>\n",
       "      <th>Class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>149.62</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>2.69</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>...</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>378.66</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>123.50</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>69.99</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Time        V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0   0.0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1   0.0  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2   1.0 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3   1.0 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4   2.0 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9  ...       V21       V22       V23       V24       V25  \\\n",
       "0  0.098698  0.363787  ... -0.018307  0.277838 -0.110474  0.066928  0.128539   \n",
       "1  0.085102 -0.255425  ... -0.225775 -0.638672  0.101288 -0.339846  0.167170   \n",
       "2  0.247676 -1.514654  ...  0.247998  0.771679  0.909412 -0.689281 -0.327642   \n",
       "3  0.377436 -1.387024  ... -0.108300  0.005274 -0.190321 -1.175575  0.647376   \n",
       "4 -0.270533  0.817739  ... -0.009431  0.798278 -0.137458  0.141267 -0.206010   \n",
       "\n",
       "        V26       V27       V28  Amount  Class  \n",
       "0 -0.189115  0.133558 -0.021053  149.62      0  \n",
       "1  0.125895 -0.008983  0.014724    2.69      0  \n",
       "2 -0.139097 -0.055353 -0.059752  378.66      0  \n",
       "3 -0.221929  0.062723  0.061458  123.50      0  \n",
       "4  0.502292  0.219422  0.215153   69.99      0  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# file located at '/data' in parent directory\n",
    "path = os.path.abspath(os.path.join(os.getcwd(), '..', 'data', 'creditcard.csv')) \n",
    "df = pd.read_csv(path)\n",
    "np.array(df.isnull().sum())\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['new_amout'] = StandardScaler().fit_transform(df['Amount'].values.reshape(-1, 1))\n",
    "\n",
    "df.drop(['Time', 'Amount'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>V10</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Class</th>\n",
       "      <th>new_amout</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.359807</td>\n",
       "      <td>-0.072781</td>\n",
       "      <td>2.536347</td>\n",
       "      <td>1.378155</td>\n",
       "      <td>-0.338321</td>\n",
       "      <td>0.462388</td>\n",
       "      <td>0.239599</td>\n",
       "      <td>0.098698</td>\n",
       "      <td>0.363787</td>\n",
       "      <td>0.090794</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.018307</td>\n",
       "      <td>0.277838</td>\n",
       "      <td>-0.110474</td>\n",
       "      <td>0.066928</td>\n",
       "      <td>0.128539</td>\n",
       "      <td>-0.189115</td>\n",
       "      <td>0.133558</td>\n",
       "      <td>-0.021053</td>\n",
       "      <td>0</td>\n",
       "      <td>0.244964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.191857</td>\n",
       "      <td>0.266151</td>\n",
       "      <td>0.166480</td>\n",
       "      <td>0.448154</td>\n",
       "      <td>0.060018</td>\n",
       "      <td>-0.082361</td>\n",
       "      <td>-0.078803</td>\n",
       "      <td>0.085102</td>\n",
       "      <td>-0.255425</td>\n",
       "      <td>-0.166974</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.225775</td>\n",
       "      <td>-0.638672</td>\n",
       "      <td>0.101288</td>\n",
       "      <td>-0.339846</td>\n",
       "      <td>0.167170</td>\n",
       "      <td>0.125895</td>\n",
       "      <td>-0.008983</td>\n",
       "      <td>0.014724</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.342475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.358354</td>\n",
       "      <td>-1.340163</td>\n",
       "      <td>1.773209</td>\n",
       "      <td>0.379780</td>\n",
       "      <td>-0.503198</td>\n",
       "      <td>1.800499</td>\n",
       "      <td>0.791461</td>\n",
       "      <td>0.247676</td>\n",
       "      <td>-1.514654</td>\n",
       "      <td>0.207643</td>\n",
       "      <td>...</td>\n",
       "      <td>0.247998</td>\n",
       "      <td>0.771679</td>\n",
       "      <td>0.909412</td>\n",
       "      <td>-0.689281</td>\n",
       "      <td>-0.327642</td>\n",
       "      <td>-0.139097</td>\n",
       "      <td>-0.055353</td>\n",
       "      <td>-0.059752</td>\n",
       "      <td>0</td>\n",
       "      <td>1.160686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.966272</td>\n",
       "      <td>-0.185226</td>\n",
       "      <td>1.792993</td>\n",
       "      <td>-0.863291</td>\n",
       "      <td>-0.010309</td>\n",
       "      <td>1.247203</td>\n",
       "      <td>0.237609</td>\n",
       "      <td>0.377436</td>\n",
       "      <td>-1.387024</td>\n",
       "      <td>-0.054952</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.108300</td>\n",
       "      <td>0.005274</td>\n",
       "      <td>-0.190321</td>\n",
       "      <td>-1.175575</td>\n",
       "      <td>0.647376</td>\n",
       "      <td>-0.221929</td>\n",
       "      <td>0.062723</td>\n",
       "      <td>0.061458</td>\n",
       "      <td>0</td>\n",
       "      <td>0.140534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.158233</td>\n",
       "      <td>0.877737</td>\n",
       "      <td>1.548718</td>\n",
       "      <td>0.403034</td>\n",
       "      <td>-0.407193</td>\n",
       "      <td>0.095921</td>\n",
       "      <td>0.592941</td>\n",
       "      <td>-0.270533</td>\n",
       "      <td>0.817739</td>\n",
       "      <td>0.753074</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.009431</td>\n",
       "      <td>0.798278</td>\n",
       "      <td>-0.137458</td>\n",
       "      <td>0.141267</td>\n",
       "      <td>-0.206010</td>\n",
       "      <td>0.502292</td>\n",
       "      <td>0.219422</td>\n",
       "      <td>0.215153</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.073403</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         V1        V2        V3        V4        V5        V6        V7  \\\n",
       "0 -1.359807 -0.072781  2.536347  1.378155 -0.338321  0.462388  0.239599   \n",
       "1  1.191857  0.266151  0.166480  0.448154  0.060018 -0.082361 -0.078803   \n",
       "2 -1.358354 -1.340163  1.773209  0.379780 -0.503198  1.800499  0.791461   \n",
       "3 -0.966272 -0.185226  1.792993 -0.863291 -0.010309  1.247203  0.237609   \n",
       "4 -1.158233  0.877737  1.548718  0.403034 -0.407193  0.095921  0.592941   \n",
       "\n",
       "         V8        V9       V10  ...       V21       V22       V23       V24  \\\n",
       "0  0.098698  0.363787  0.090794  ... -0.018307  0.277838 -0.110474  0.066928   \n",
       "1  0.085102 -0.255425 -0.166974  ... -0.225775 -0.638672  0.101288 -0.339846   \n",
       "2  0.247676 -1.514654  0.207643  ...  0.247998  0.771679  0.909412 -0.689281   \n",
       "3  0.377436 -1.387024 -0.054952  ... -0.108300  0.005274 -0.190321 -1.175575   \n",
       "4 -0.270533  0.817739  0.753074  ... -0.009431  0.798278 -0.137458  0.141267   \n",
       "\n",
       "        V25       V26       V27       V28  Class  new_amout  \n",
       "0  0.128539 -0.189115  0.133558 -0.021053      0   0.244964  \n",
       "1  0.167170  0.125895 -0.008983  0.014724      0  -0.342475  \n",
       "2 -0.327642 -0.139097 -0.055353 -0.059752      0   1.160686  \n",
       "3  0.647376 -0.221929  0.062723  0.061458      0   0.140534  \n",
       "4 -0.206010  0.502292  0.219422  0.215153      0  -0.073403  \n",
       "\n",
       "[5 rows x 30 columns]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. Sampling method 1 : Fraud transaction 데이터 개수에 맞게 Normal transaction을 sampling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\pandas\\core\\frame.py:3940: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  errors=errors)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "443\n",
      "(443, 30)\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "train_set_p = 0.1\n",
    "\n",
    "fraud_series = df[df['Class'] == 1]\n",
    "\n",
    "idx = fraud_series.index.values\n",
    "\n",
    "np.random.shuffle(idx)\n",
    "fraud_series.drop(idx[:int(idx.shape[0]*train_set_p)], inplace=True)\n",
    "df.drop(fraud_series.index.values, inplace=True)\n",
    "\n",
    "print(len(fraud_series))\n",
    "print(fraud_series.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>V1</th>\n",
       "      <th>V2</th>\n",
       "      <th>V3</th>\n",
       "      <th>V4</th>\n",
       "      <th>V5</th>\n",
       "      <th>V6</th>\n",
       "      <th>V7</th>\n",
       "      <th>V8</th>\n",
       "      <th>V9</th>\n",
       "      <th>V10</th>\n",
       "      <th>...</th>\n",
       "      <th>V21</th>\n",
       "      <th>V22</th>\n",
       "      <th>V23</th>\n",
       "      <th>V24</th>\n",
       "      <th>V25</th>\n",
       "      <th>V26</th>\n",
       "      <th>V27</th>\n",
       "      <th>V28</th>\n",
       "      <th>Class</th>\n",
       "      <th>new_amout</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>154684</th>\n",
       "      <td>-14.724627</td>\n",
       "      <td>7.875157</td>\n",
       "      <td>-21.872317</td>\n",
       "      <td>11.906170</td>\n",
       "      <td>-8.348734</td>\n",
       "      <td>-2.262846</td>\n",
       "      <td>-15.833443</td>\n",
       "      <td>0.077874</td>\n",
       "      <td>-6.356833</td>\n",
       "      <td>-13.261652</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.362345</td>\n",
       "      <td>1.099557</td>\n",
       "      <td>1.037199</td>\n",
       "      <td>-1.036359</td>\n",
       "      <td>-0.254777</td>\n",
       "      <td>0.642343</td>\n",
       "      <td>2.161129</td>\n",
       "      <td>-1.401282</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.349231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44556</th>\n",
       "      <td>-15.271362</td>\n",
       "      <td>8.326581</td>\n",
       "      <td>-22.338591</td>\n",
       "      <td>11.885313</td>\n",
       "      <td>-8.721334</td>\n",
       "      <td>-2.324307</td>\n",
       "      <td>-16.196419</td>\n",
       "      <td>0.512882</td>\n",
       "      <td>-6.333685</td>\n",
       "      <td>-13.175198</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.356896</td>\n",
       "      <td>1.068019</td>\n",
       "      <td>1.085617</td>\n",
       "      <td>-1.039797</td>\n",
       "      <td>-0.182006</td>\n",
       "      <td>0.649921</td>\n",
       "      <td>2.149247</td>\n",
       "      <td>-1.406811</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.349231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>218442</th>\n",
       "      <td>-15.819179</td>\n",
       "      <td>8.775997</td>\n",
       "      <td>-22.804686</td>\n",
       "      <td>11.864868</td>\n",
       "      <td>-9.092361</td>\n",
       "      <td>-2.386893</td>\n",
       "      <td>-16.560368</td>\n",
       "      <td>0.948349</td>\n",
       "      <td>-6.310658</td>\n",
       "      <td>-13.088891</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.350634</td>\n",
       "      <td>1.036362</td>\n",
       "      <td>1.136051</td>\n",
       "      <td>-1.043414</td>\n",
       "      <td>-0.108923</td>\n",
       "      <td>0.657437</td>\n",
       "      <td>2.136424</td>\n",
       "      <td>-1.411945</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.349231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135718</th>\n",
       "      <td>-16.367923</td>\n",
       "      <td>9.223692</td>\n",
       "      <td>-23.270631</td>\n",
       "      <td>11.844777</td>\n",
       "      <td>-9.462037</td>\n",
       "      <td>-2.450444</td>\n",
       "      <td>-16.925152</td>\n",
       "      <td>1.384208</td>\n",
       "      <td>-6.287736</td>\n",
       "      <td>-13.002709</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.343674</td>\n",
       "      <td>1.004602</td>\n",
       "      <td>1.188212</td>\n",
       "      <td>-1.047184</td>\n",
       "      <td>-0.035573</td>\n",
       "      <td>0.664900</td>\n",
       "      <td>2.122796</td>\n",
       "      <td>-1.416741</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.349231</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77348</th>\n",
       "      <td>-16.917468</td>\n",
       "      <td>9.669900</td>\n",
       "      <td>-23.736443</td>\n",
       "      <td>11.824990</td>\n",
       "      <td>-9.830548</td>\n",
       "      <td>-2.514829</td>\n",
       "      <td>-17.290657</td>\n",
       "      <td>1.820408</td>\n",
       "      <td>-6.264903</td>\n",
       "      <td>-12.916636</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.336111</td>\n",
       "      <td>0.972755</td>\n",
       "      <td>1.241866</td>\n",
       "      <td>-1.051086</td>\n",
       "      <td>0.038009</td>\n",
       "      <td>0.672317</td>\n",
       "      <td>2.108471</td>\n",
       "      <td>-1.421243</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.349231</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 30 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               V1        V2         V3         V4        V5        V6  \\\n",
       "154684 -14.724627  7.875157 -21.872317  11.906170 -8.348734 -2.262846   \n",
       "44556  -15.271362  8.326581 -22.338591  11.885313 -8.721334 -2.324307   \n",
       "218442 -15.819179  8.775997 -22.804686  11.864868 -9.092361 -2.386893   \n",
       "135718 -16.367923  9.223692 -23.270631  11.844777 -9.462037 -2.450444   \n",
       "77348  -16.917468  9.669900 -23.736443  11.824990 -9.830548 -2.514829   \n",
       "\n",
       "               V7        V8        V9        V10  ...       V21       V22  \\\n",
       "154684 -15.833443  0.077874 -6.356833 -13.261652  ... -2.362345  1.099557   \n",
       "44556  -16.196419  0.512882 -6.333685 -13.175198  ... -2.356896  1.068019   \n",
       "218442 -16.560368  0.948349 -6.310658 -13.088891  ... -2.350634  1.036362   \n",
       "135718 -16.925152  1.384208 -6.287736 -13.002709  ... -2.343674  1.004602   \n",
       "77348  -17.290657  1.820408 -6.264903 -12.916636  ... -2.336111  0.972755   \n",
       "\n",
       "             V23       V24       V25       V26       V27       V28  Class  \\\n",
       "154684  1.037199 -1.036359 -0.254777  0.642343  2.161129 -1.401282      1   \n",
       "44556   1.085617 -1.039797 -0.182006  0.649921  2.149247 -1.406811      1   \n",
       "218442  1.136051 -1.043414 -0.108923  0.657437  2.136424 -1.411945      1   \n",
       "135718  1.188212 -1.047184 -0.035573  0.664900  2.122796 -1.416741      1   \n",
       "77348   1.241866 -1.051086  0.038009  0.672317  2.108471 -1.421243      1   \n",
       "\n",
       "        new_amout  \n",
       "154684  -0.349231  \n",
       "44556   -0.349231  \n",
       "218442  -0.349231  \n",
       "135718  -0.349231  \n",
       "77348   -0.349231  \n",
       "\n",
       "[5 rows x 30 columns]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fraud_series.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "#normal dataset with the same size of the fraud_series sets\n",
    "\n",
    "normal_series = df[df['Class'] == 0]\n",
    "idx = normal_series.index.values\n",
    "np.random.shuffle(idx)\n",
    "normal_series.drop(idx[fraud_series.shape[0]: ], inplace=True)\n",
    "df.drop(normal_series.index.values, inplace=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "443\n"
     ]
    }
   ],
   "source": [
    "print(len(normal_series))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "#build trainig dataset\n",
    "\n",
    "new_df = pd.concat([normal_series, fraud_series])\n",
    "new_df.reset_index(inplace=True, drop=True)\n",
    "y = new_df['Class'].values.reshape(-1, 1)\n",
    "new_df.drop(['Class'], axis = 1, inplace=True)\n",
    "X = new_df\n",
    "\n",
    "#len(df[df['Class']==0])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2. 전체 데이터셋을 활용한다. (시간이 너무 오래걸린다.) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "#build new trainig dataset\n",
    "\n",
    "#y = df['Class'].values.reshape(-1, 1)\n",
    "#df.drop(['Class'], axis = 1, inplace=True)\n",
    "#X = df\n",
    "\n",
    "#len(df[df['Class']==0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### - Fraud transaction과 normal transaction의 수가 같도록 sampling 한 뒤, 그 set을 train/test로 split해서 SVM model을 적합시켰다. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train.shape:  (620, 29)\n",
      "X_test.shape:  (266, 29)\n",
      "y_train.shape:  (620,)\n",
      "y_test.shape:  (266,)\n"
     ]
    }
   ],
   "source": [
    "#attributes that will be used by the gridsearchCV algorithm\n",
    "attr = {'C' : [0.1, 1, 2, 5, 10, 25, 100],\n",
    "       'gamma' : [1e-1, 1e-2, 1e-3]}\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y.ravel(), test_size=0.3, random_state = 10)\n",
    "\n",
    "print(\"X_train.shape: \", X_train.shape)\n",
    "print(\"X_test.shape: \", X_test.shape)\n",
    "print(\"y_train.shape: \", y_train.shape)\n",
    "print(\"y_test.shape: \", y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy:  0.9548872180451128\n"
     ]
    }
   ],
   "source": [
    "model = SVC()\n",
    "classif = GridSearchCV(model, attr, cv=5)\n",
    "classif.fit(X_train, y_train)\n",
    "y_pred = classif.predict(X_test)\n",
    "print('Accuracy: ', accuracy_score(y_pred, y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 아래는 fraud/normal 데이터 수가 같도록 조정한 뒤의 new_df set을 기준으로 SVM model을 학습시킨 뒤, 그 모델을 전체 데이터셋에 적용해 본 것이다. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[236660  47212]\n",
      " [     3     46]]\n"
     ]
    }
   ],
   "source": [
    "y_all = df['Class'].values.reshape(-1, 1)\n",
    "df.drop(['Class'], axis=1, inplace=True)\n",
    "X_all = df\n",
    "y_pred_all = classif.predict(X_all)\n",
    "print(confusion_matrix(y_all, y_pred_all))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9387755102040817\n"
     ]
    }
   ],
   "source": [
    "print(recall_score(y_all, y_pred_all))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "####  * Fraud와 normal의 비율이 같은 set의 test set에 SVM을 적용하여 confusion matrix, average precision score, auc, ROC 등을 구해보았다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import roc_auc_score, roc_curve, auc, average_precision_score\n",
    "from mlxtend.plotting import plot_confusion_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "con_mat = confusion_matrix(y_test, y_pred)\n",
    "average_precision = average_precision_score(y_test, y_pred)\n",
    "cls_report = classification_report(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*****************************************************************\n",
      "Area under the curve : 0.953691\n",
      "Average precision-recall score RF: 0.9441044016290727\n",
      "[[136   2]\n",
      " [ 10 118]]\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.93      0.99      0.96       138\n",
      "           1       0.98      0.92      0.95       128\n",
      "\n",
      "    accuracy                           0.95       266\n",
      "   macro avg       0.96      0.95      0.95       266\n",
      "weighted avg       0.96      0.95      0.95       266\n",
      "\n",
      "*****************************************************************\n"
     ]
    }
   ],
   "source": [
    "print(\"*****************************************************************\")\n",
    "print(\"Area under the curve : %f\" % (roc_auc_score(y_test, y_pred)))\n",
    "print(\"Average precision-recall score RF: {}\".format(average_precision))\n",
    "print(con_mat)\n",
    "print(cls_report)\n",
    "print(\"*****************************************************************\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The accuracy is 95.48872180451127 %\n",
      "The recall from the confusion matrix is 92.1875 %\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\matplotlib\\figure.py:98: MatplotlibDeprecationWarning: \n",
      "Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  \"Adding an axes using the same arguments as a previous axes \"\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def confus_matrix(CM):\n",
    "    fig, ax = plot_confusion_matrix(conf_mat= CM)\n",
    "    plt.title(\"The Confusion Matrix of full dataset using best_parameters\")\n",
    "    plt.ylabel(\"Actual\")\n",
    "    plt.xlabel(\"Predicted\")\n",
    "    plt.show()\n",
    "    print(\"The accuracy is \"+str((CM[1,1]+CM[0,0])/(CM[0,0] + CM[0,1]+CM[1,0] + CM[1,1])*100) + \" %\")\n",
    "    print(\"The recall from the confusion matrix is \"+ str(CM[1,1]/(CM[1,0] + CM[1,1])*100) +\" %\")\n",
    "confus_matrix(con_mat)\n",
    "\n",
    "precision, recall, _ = precision_recall_curve(y_test, y_pred)\n",
    "plt.step(recall, precision, color='b', alpha=0.2, where='post')\n",
    "plt.fill_between(recall, precision, step='post', alpha=0.2, color='b')\n",
    "\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.title('2-class Precision-Recall curve: AP={0:0.2f}'.format(average_precision))\n",
    "\n",
    "fpr_rf, tpr_rf, _ = roc_curve(y_test, y_pred)\n",
    "roc_auc_rf = auc(fpr_rf, tpr_rf)\n",
    "plt.figure(figsize=(8,8))\n",
    "plt.xlim([-0.01, 1.00])\n",
    "plt.ylim([-0.01, 1.01])\n",
    "plt.plot(fpr_rf, tpr_rf, lw=1, label='{} curve (AUC = {:0.2f})'.format('RF',roc_auc_rf))\n",
    "\n",
    "plt.xlabel('False Positive Rate', fontsize=16)\n",
    "plt.ylabel('True Positive Rate', fontsize=16)\n",
    "plt.title('ROC curve', fontsize=16)\n",
    "plt.legend(loc='lower right', fontsize=13)\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=1, linestyle='--')\n",
    "plt.axes().set_aspect('equal')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Fraud 개수에 맞춰 적합시킨 SVM model을 whole dataset에 대해 적용하여 auc, confusion matrix, ROC curve를 그려보았다. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*****************************************************************\n",
      "Area under the curve : 0.886231\n",
      "Average precision-recall score RF: 0.0009243517837514333\n",
      "[[236660  47212]\n",
      " [     3     46]]\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       1.00      0.83      0.91    283872\n",
      "           1       0.00      0.94      0.00        49\n",
      "\n",
      "    accuracy                           0.83    283921\n",
      "   macro avg       0.50      0.89      0.46    283921\n",
      "weighted avg       1.00      0.83      0.91    283921\n",
      "\n",
      "*****************************************************************\n"
     ]
    }
   ],
   "source": [
    "con_mat = confusion_matrix(y_all, y_pred_all)\n",
    "average_precision = average_precision_score(y_all, y_pred_all)\n",
    "cls_report = classification_report(y_all, y_pred_all)\n",
    "\n",
    "print(\"*****************************************************************\")\n",
    "print(\"Area under the curve : %f\" % (roc_auc_score(y_all, y_pred_all)))\n",
    "print(\"Average precision-recall score RF: {}\".format(average_precision))\n",
    "print(con_mat)\n",
    "print(cls_report)\n",
    "print(\"*****************************************************************\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The accuracy is 83.37037415337365 %\n",
      "The recall from the confusion matrix is 93.87755102040816 %\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Anaconda3\\lib\\site-packages\\matplotlib\\figure.py:98: MatplotlibDeprecationWarning: \n",
      "Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  \"Adding an axes using the same arguments as a previous axes \"\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def confus_matrix(CM):\n",
    "    fig, ax = plot_confusion_matrix(conf_mat= CM)\n",
    "    plt.title(\"The Confusion Matrix of full dataset using best_parameters\")\n",
    "    plt.ylabel(\"Actual\")\n",
    "    plt.xlabel(\"Predicted\")\n",
    "    plt.show()\n",
    "    print(\"The accuracy is \"+str((CM[1,1]+CM[0,0])/(CM[0,0] + CM[0,1]+CM[1,0] + CM[1,1])*100) + \" %\")\n",
    "    print(\"The recall from the confusion matrix is \"+ str(CM[1,1]/(CM[1,0] + CM[1,1])*100) +\" %\")\n",
    "confus_matrix(con_mat)\n",
    "\n",
    "fpr_rf, tpr_rf, _ = roc_curve(y_all, y_pred_all)\n",
    "roc_auc_rf = auc(fpr_rf, tpr_rf)\n",
    "plt.figure(figsize=(8,8))\n",
    "plt.xlim([-0.01, 1.00])\n",
    "plt.ylim([-0.01, 1.01])\n",
    "plt.plot(fpr_rf, tpr_rf, lw=1, label='{} curve (AUC = {:0.2f})'.format('RF',roc_auc_rf))\n",
    "\n",
    "plt.xlabel('False Positive Rate', fontsize=16)\n",
    "plt.ylabel('True Positive Rate', fontsize=16)\n",
    "plt.title('ROC curve', fontsize=16)\n",
    "plt.legend(loc='lower right', fontsize=13)\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=1, linestyle='--')\n",
    "plt.axes().set_aspect('equal')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}