Precision, Recall, F1score.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## sklearn.metrics.accuracy_score\n",
    "* sklearn.metrics.accuracy_score(y_true, y_pred, *, normalize=True, sample_weight=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## sklearn.metrics.precision_score\n",
    "* sklearn.metrics.precision_score(y_true, y_pred, *, labels=None, pos_label=1, average='binary', sample_weight=None, zero_division='warn')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## sklearn.metrics.recall_score\n",
    "* sklearn.metrics.recall_score(y_true, y_pred, *, labels=None, pos_label=1, average='binary', sample_weight=None, zero_division='warn')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## sklearn.metrics.confusion_matrix\n",
    "* sklearn.metrics.confusion_matrix(y_true, y_pred, *, labels=None, sample_weight=None, normalize=None)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## sklearn.metrics.f1_score\n",
    "* sklearn.metrics.f1_score(y_true, y_pred, *, labels=None, pos_label=1, average='binary', sample_weight=None, zero_division='warn')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, confusion_matrix, f1_score\n",
    "\n",
    "def get_clf_eval(y_test, pred):\n",
    "    confusion = confusion_matrix(y_test,pred)\n",
    "    accuracy = accuracy_score(y_test,pred)\n",
    "    precision = precision_score(y_test,pred)\n",
    "    recall = recall_score(y_test,pred)\n",
    "    f1 = f1_score(y_test, lr_pred)\n",
    "    print('오차 행렬')\n",
    "    print(confusion)\n",
    "    print(f'정확도: {accuracy:.4f}, 정밀도: {precision:.4f}, 재현율:{recall:.4f}, F1:{f1:.4f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>PassengerId</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PassengerId  Survived  Pclass  \\\n",
       "0            1         0       3   \n",
       "1            2         1       1   \n",
       "2            3         1       3   \n",
       "\n",
       "                                                Name     Sex   Age  SibSp  \\\n",
       "0                            Braund, Mr. Owen Harris    male  22.0      1   \n",
       "1  Cumings, Mrs. John Bradley (Florence Briggs Th...  female  38.0      1   \n",
       "2                             Heikkinen, Miss. Laina  female  26.0      0   \n",
       "\n",
       "   Parch            Ticket     Fare Cabin Embarked  \n",
       "0      0         A/5 21171   7.2500   NaN        S  \n",
       "1      0          PC 17599  71.2833   C85        C  \n",
       "2      0  STON/O2. 3101282   7.9250   NaN        S  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "titanic_df = pd.read_csv('./datasets/titanic_train.csv')\n",
    "titanic_df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fillna_features(df):\n",
    "    df['Age'].fillna(df['Age'].mean(),inplace=True)\n",
    "    return df\n",
    "\n",
    "def drop_features(df):\n",
    "    df.drop(['PassengerId', 'Name', 'SibSp',\n",
    "       'Parch', 'Ticket', 'Fare', 'Cabin', 'Embarked'],axis=1,inplace=True)\n",
    "    return df\n",
    "\n",
    "def format_features(df):\n",
    "    encoder = LabelEncoder()\n",
    "    encoder.fit(df['Sex'])\n",
    "    df['Sex'] = encoder.transform(df['Sex'])\n",
    "    return df\n",
    "\n",
    "def transform_features(df):\n",
    "    df = fillna_features(df)\n",
    "    df = drop_features(df)\n",
    "    df = format_features(df)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_titanic_df = titanic_df['Survived']\n",
    "X_titanic_df= titanic_df.drop('Survived',axis=1)\n",
    "X_titanic_df = transform_features(X_titanic_df)\n",
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test=train_test_split(X_titanic_df, y_titanic_df, \\\n",
    "                                                  test_size=0.2, random_state=11)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "오차 행렬\n",
      "[[100  18]\n",
      " [ 13  48]]\n",
      "정확도: 0.8268, 정밀도: 0.7273, 재현율:0.7869, F1:0.7559\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "lr_clf = LogisticRegression()\n",
    "lr_clf.fit(X_train , y_train)\n",
    "lr_pred = lr_clf.predict(X_test)\n",
    "get_clf_eval(y_test,lr_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.44781358, 0.55218642],\n",
       "       [0.89023075, 0.10976925],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.86466451, 0.13533549],\n",
       "       [0.75812794, 0.24187206],\n",
       "       [0.85434917, 0.14565083],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.33510303, 0.66489697],\n",
       "       [0.71971405, 0.28028595],\n",
       "       [0.32644462, 0.67355538],\n",
       "       [0.87119387, 0.12880613],\n",
       "       [0.86256142, 0.13743858],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.49595405, 0.50404595],\n",
       "       [0.87551989, 0.12448011],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.71971405, 0.28028595],\n",
       "       [0.63693828, 0.36306172],\n",
       "       [0.26366186, 0.73633814],\n",
       "       [0.73662511, 0.26337489],\n",
       "       [0.63032568, 0.36967432],\n",
       "       [0.87238203, 0.12761797],\n",
       "       [0.87551989, 0.12448011],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.7459071 , 0.2540929 ],\n",
       "       [0.8317686 , 0.1682314 ],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.70807897, 0.29192103],\n",
       "       [0.61020101, 0.38979899],\n",
       "       [0.08350277, 0.91649723],\n",
       "       [0.22744251, 0.77255749],\n",
       "       [0.89023075, 0.10976925],\n",
       "       [0.28058432, 0.71941568],\n",
       "       [0.63032568, 0.36967432],\n",
       "       [0.7459071 , 0.2540929 ],\n",
       "       [0.90585253, 0.09414747],\n",
       "       [0.38762445, 0.61237555],\n",
       "       [0.91513088, 0.08486912],\n",
       "       [0.90339492, 0.09660508],\n",
       "       [0.70807897, 0.29192103],\n",
       "       [0.88741603, 0.11258397],\n",
       "       [0.2104012 , 0.7895988 ],\n",
       "       [0.18771337, 0.81228663],\n",
       "       [0.44781358, 0.55218642],\n",
       "       [0.44781358, 0.55218642],\n",
       "       [0.13974673, 0.86025327],\n",
       "       [0.50673646, 0.49326354],\n",
       "       [0.07001181, 0.92998819],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.49595405, 0.50404595],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.88159724, 0.11840276],\n",
       "       [0.28637012, 0.71362988],\n",
       "       [0.6839814 , 0.3160186 ],\n",
       "       [0.78085036, 0.21914964],\n",
       "       [0.72896639, 0.27103361],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.86590363, 0.13409637],\n",
       "       [0.49595405, 0.50404595],\n",
       "       [0.70807897, 0.29192103],\n",
       "       [0.8845385 , 0.1154615 ],\n",
       "       [0.49595405, 0.50404595],\n",
       "       [0.49595405, 0.50404595],\n",
       "       [0.67546629, 0.32453371],\n",
       "       [0.9106002 , 0.0893998 ],\n",
       "       [0.39440739, 0.60559261],\n",
       "       [0.45699349, 0.54300651],\n",
       "       [0.05809047, 0.94190953],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.85914946, 0.14085054],\n",
       "       [0.85566697, 0.14433303],\n",
       "       [0.90339492, 0.09660508],\n",
       "       [0.06128733, 0.93871267],\n",
       "       [0.77840333, 0.22159667],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.56448579, 0.43551421],\n",
       "       [0.79279284, 0.20720716],\n",
       "       [0.17502936, 0.82497064],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.23503773, 0.76496227],\n",
       "       [0.44289439, 0.55710561],\n",
       "       [0.08350277, 0.91649723],\n",
       "       [0.80648026, 0.19351974],\n",
       "       [0.06639205, 0.93360795],\n",
       "       [0.05966885, 0.94033115],\n",
       "       [0.89433708, 0.10566292],\n",
       "       [0.89567546, 0.10432454],\n",
       "       [0.12464135, 0.87535865],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.7459071 , 0.2540929 ],\n",
       "       [0.73662511, 0.26337489],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.44781358, 0.55218642],\n",
       "       [0.9255586 , 0.0744414 ],\n",
       "       [0.06817967, 0.93182033],\n",
       "       [0.90585253, 0.09414747],\n",
       "       [0.56187069, 0.43812931],\n",
       "       [0.07188938, 0.92811062],\n",
       "       [0.38762445, 0.61237555],\n",
       "       [0.79453364, 0.20546636],\n",
       "       [0.08199128, 0.91800872],\n",
       "       [0.90825397, 0.09174603],\n",
       "       [0.43587775, 0.56412225],\n",
       "       [0.8845385 , 0.1154615 ],\n",
       "       [0.87551989, 0.12448011],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.32253081, 0.67746919],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.85434917, 0.14565083],\n",
       "       [0.89960109, 0.10039891],\n",
       "       [0.57147543, 0.42852457],\n",
       "       [0.46964837, 0.53035163],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.89298358, 0.10701642],\n",
       "       [0.84711578, 0.15288422],\n",
       "       [0.77840333, 0.22159667],\n",
       "       [0.12043586, 0.87956414],\n",
       "       [0.93123662, 0.06876338],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.88741603, 0.11258397],\n",
       "       [0.92153686, 0.07846314],\n",
       "       [0.57732555, 0.42267445],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.86256142, 0.13743858],\n",
       "       [0.67779219, 0.32220781],\n",
       "       [0.36754443, 0.63245557],\n",
       "       [0.761387  , 0.238613  ],\n",
       "       [0.07188938, 0.92811062],\n",
       "       [0.53365279, 0.46634721],\n",
       "       [0.34147941, 0.65852059],\n",
       "       [0.51385473, 0.48614527],\n",
       "       [0.42192392, 0.57807608],\n",
       "       [0.51385473, 0.48614527],\n",
       "       [0.23503773, 0.76496227],\n",
       "       [0.75617335, 0.24382665],\n",
       "       [0.90339492, 0.09660508],\n",
       "       [0.12348616, 0.87651384],\n",
       "       [0.13974673, 0.86025327],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.90088018, 0.09911982],\n",
       "       [0.8081339 , 0.1918661 ],\n",
       "       [0.91289218, 0.08710782],\n",
       "       [0.18771337, 0.81228663],\n",
       "       [0.9255586 , 0.0744414 ],\n",
       "       [0.67546629, 0.32453371],\n",
       "       [0.08350277, 0.91649723],\n",
       "       [0.76842062, 0.23157938],\n",
       "       [0.49595405, 0.50404595],\n",
       "       [0.40809364, 0.59190636],\n",
       "       [0.36754443, 0.63245557],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.18771337, 0.81228663],\n",
       "       [0.07188938, 0.92811062],\n",
       "       [0.49097623, 0.50902377],\n",
       "       [0.88159724, 0.11840276],\n",
       "       [0.31021021, 0.68978979],\n",
       "       [0.39440739, 0.60559261],\n",
       "       [0.86917689, 0.13082311],\n",
       "       [0.18771337, 0.81228663],\n",
       "       [0.91289218, 0.08710782],\n",
       "       [0.39440739, 0.60559261],\n",
       "       [0.44781358, 0.55218642],\n",
       "       [0.40809364, 0.59190636],\n",
       "       [0.1615851 , 0.8384149 ],\n",
       "       [0.60340466, 0.39659534],\n",
       "       [0.91289218, 0.08710782],\n",
       "       [0.09098629, 0.90901371],\n",
       "       [0.89752171, 0.10247829],\n",
       "       [0.1156126 , 0.8843874 ],\n",
       "       [0.7421142 , 0.2578858 ],\n",
       "       [0.69616481, 0.30383519],\n",
       "       [0.52807156, 0.47192844],\n",
       "       [0.93651127, 0.06348873],\n",
       "       [0.87551989, 0.12448011],\n",
       "       [0.49595405, 0.50404595],\n",
       "       [0.4625598 , 0.5374402 ]])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_clf.predict_proba(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.55218642, 0.10976925, 0.10247829, 0.13533549, 0.24187206,\n",
       "       0.14565083, 0.10247829, 0.66489697, 0.28028595, 0.67355538,\n",
       "       0.12880613, 0.13743858, 0.10247829, 0.10247829, 0.50404595,\n",
       "       0.12448011, 0.10247829, 0.28028595, 0.36306172, 0.73633814,\n",
       "       0.26337489, 0.36967432, 0.12761797, 0.12448011, 0.10247829,\n",
       "       0.2540929 , 0.1682314 , 0.10247829, 0.29192103, 0.38979899,\n",
       "       0.91649723, 0.77255749, 0.10976925, 0.71941568, 0.36967432,\n",
       "       0.2540929 , 0.09414747, 0.61237555, 0.08486912, 0.09660508,\n",
       "       0.29192103, 0.11258397, 0.7895988 , 0.81228663, 0.55218642,\n",
       "       0.55218642, 0.86025327, 0.49326354, 0.92998819, 0.10247829,\n",
       "       0.50404595, 0.10247829, 0.11840276, 0.71362988, 0.3160186 ,\n",
       "       0.21914964, 0.27103361, 0.10247829, 0.13409637, 0.50404595,\n",
       "       0.29192103, 0.1154615 , 0.50404595, 0.50404595, 0.32453371,\n",
       "       0.0893998 , 0.60559261, 0.54300651, 0.94190953, 0.10247829,\n",
       "       0.14085054, 0.14433303, 0.09660508, 0.93871267, 0.22159667,\n",
       "       0.10247829, 0.43551421, 0.20720716, 0.82497064, 0.10247829,\n",
       "       0.76496227, 0.55710561, 0.91649723, 0.19351974, 0.93360795,\n",
       "       0.94033115, 0.10566292, 0.10432454, 0.87535865, 0.10247829,\n",
       "       0.10247829, 0.2540929 , 0.26337489, 0.10247829, 0.55218642,\n",
       "       0.0744414 , 0.93182033, 0.09414747, 0.43812931, 0.92811062,\n",
       "       0.61237555, 0.20546636, 0.91800872, 0.09174603, 0.56412225,\n",
       "       0.1154615 , 0.12448011, 0.10247829, 0.67746919, 0.10247829,\n",
       "       0.14565083, 0.10039891, 0.42852457, 0.53035163, 0.10247829,\n",
       "       0.10701642, 0.15288422, 0.22159667, 0.87956414, 0.06876338,\n",
       "       0.10247829, 0.11258397, 0.07846314, 0.42267445, 0.10247829,\n",
       "       0.10247829, 0.13743858, 0.32220781, 0.63245557, 0.238613  ,\n",
       "       0.92811062, 0.46634721, 0.65852059, 0.48614527, 0.57807608,\n",
       "       0.48614527, 0.76496227, 0.24382665, 0.09660508, 0.87651384,\n",
       "       0.86025327, 0.10247829, 0.09911982, 0.1918661 , 0.08710782,\n",
       "       0.81228663, 0.0744414 , 0.32453371, 0.91649723, 0.23157938,\n",
       "       0.50404595, 0.59190636, 0.63245557, 0.10247829, 0.81228663,\n",
       "       0.92811062, 0.50902377, 0.11840276, 0.68978979, 0.60559261,\n",
       "       0.13082311, 0.81228663, 0.08710782, 0.60559261, 0.55218642,\n",
       "       0.59190636, 0.8384149 , 0.39659534, 0.08710782, 0.90901371,\n",
       "       0.10247829, 0.8843874 , 0.2578858 , 0.30383519, 0.47192844,\n",
       "       0.06348873, 0.12448011, 0.50404595, 0.5374402 ])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lr_clf.predict_proba(X_test)[:,1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## sklearn.metrics.precision_recall_curve\n",
    "* sklearn.metrics.precision_recall_curve(y_true, probas_pred, *, pos_label=None, sample_weight=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "precisions, recalls, thresholds = precision_recall_curve(y_test, lr_clf.predict_proba(X_test)[:,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(98,)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precisions.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(98,)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recalls.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(97,)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "thresholds.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def precision_recall_curve_plot(y_test=None, pred_proba_c1=None):\n",
    "    # threshold ndarray와 이 threshold에 따른 정밀도, 재현율 ndarray 추출. \n",
    "    precisions, recalls, thresholds = precision_recall_curve( y_test, pred_proba_c1)\n",
    "    \n",
    "    # X축을 threshold값으로, Y축은 정밀도, 재현율 값으로 각각 Plot 수행. 정밀도는 점선으로 표시\n",
    "    plt.figure(figsize=(8,6))\n",
    "    threshold_boundary = thresholds.shape[0]\n",
    "    plt.plot(thresholds, precisions[0:threshold_boundary], linestyle='--', label='precision')\n",
    "    plt.plot(thresholds, recalls[0:threshold_boundary],label='recall')\n",
    "    \n",
    "    # threshold 값 X 축의 Scale을 0.1 단위로 변경\n",
    "    start, end = plt.xlim()\n",
    "    plt.xticks(np.round(np.arange(start, end, 0.1),2))\n",
    "    \n",
    "    # x축, y축 label과 legend, 그리고 grid 설정\n",
    "    plt.xlabel('Threshold value'); plt.ylabel('Precision and Recall value')\n",
    "    plt.legend(); plt.grid()\n",
    "    plt.show()\n",
    "\n",
    "precision_recall_curve_plot(y_test, lr_clf.predict_proba(X_test)[:,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.1 64-bit",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "31f8360559af503f7d1219287c80e0cc362e8a430fe9b8335f384415946613ec"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}