lab_1_1_getting started for scikit-learn machine learning_jaes.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Getting Started\n",
    "https://scikit-learn.org/stable/getting_started.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>.container { width:100% !important; }</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from IPython.core.display import display, HTML\n",
    "display(HTML(\"<style>.container { width:100% !important; }</style>\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.24.2\n"
     ]
    }
   ],
   "source": [
    "import sklearn\n",
    "print(sklearn.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 필요 패키지 import\n",
    "import numpy as np \n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['font.family'] = 'Malgun Gothic'\n",
    "plt.rcParams['axes.unicode_minus'] = False "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 실습용 데이터 - size_korea\n",
    "https://sizekorea.kr/page/report/1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "size_korea = pd.read_csv('../data/size_korea_small.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>키</th>\n",
       "      <th>몸무게</th>\n",
       "      <th>체지방율</th>\n",
       "      <th>성별</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1736.0</td>\n",
       "      <td>72.1</td>\n",
       "      <td>18.9</td>\n",
       "      <td>남</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1833.0</td>\n",
       "      <td>106.2</td>\n",
       "      <td>27.0</td>\n",
       "      <td>남</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1744.0</td>\n",
       "      <td>69.1</td>\n",
       "      <td>8.1</td>\n",
       "      <td>남</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1751.0</td>\n",
       "      <td>68.4</td>\n",
       "      <td>10.3</td>\n",
       "      <td>남</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1851.0</td>\n",
       "      <td>81.9</td>\n",
       "      <td>14.1</td>\n",
       "      <td>남</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        키    몸무게  체지방율 성별\n",
       "0  1736.0   72.1  18.9  남\n",
       "1  1833.0  106.2  27.0  남\n",
       "2  1744.0   69.1   8.1  남\n",
       "3  1751.0   68.4  10.3  남\n",
       "4  1851.0   81.9  14.1  남"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "size_korea.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='몸무게', ylabel='키'>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "size_korea.replace({'남':'blue','여':'red'}).plot.scatter('몸무게','키',c='성별', alpha=0.1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 성별: 키 몸무게 평균값"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>키</th>\n",
       "      <th>몸무게</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>성별</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>남</th>\n",
       "      <td>1722.415307</td>\n",
       "      <td>70.952886</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>여</th>\n",
       "      <td>1587.070650</td>\n",
       "      <td>56.806660</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              키        몸무게\n",
       "성별                        \n",
       "남   1722.415307  70.952886\n",
       "여   1587.070650  56.806660"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "size_korea.groupby('성별')[['키','몸무게']].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 평균값만을 가지고  예측해 보기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = size_korea.groupby('성별')[['키','몸무게']].mean().values\n",
    "y = [0, 1]  # classes of each sample"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1722.4153074 ,   70.95288582],\n",
       "       [1587.07065048,   56.80666044]])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "clf = RandomForestClassifier(random_state=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 학습"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(random_state=0)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X,y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 예측 \n",
    "- 내 키와 몸무게를 넣어 봅시다"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.predict([[170, 75]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 원래 데이터에서는 얼마나 맞췄을까?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 1, 1, 1])"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.predict(size_korea[['키','몸무게']].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8772066864552414"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8772066864552414"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.score(size_korea[['키','몸무게']].values, size_korea.성별.replace({'남':0,'여':1}))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8772066864552414"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 정확도 계산\n",
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(size_korea.성별.replace({'남':0,'여':1}), clf.predict(size_korea[['키','몸무게']].values))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 다른 모델들에도 적용해 보면 ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "clf = LogisticRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression()"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6917669114200906"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GaussianNB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "clf = GaussianNB()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GaussianNB()"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8795500703015154"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC\n",
    "clf = SVC()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC()"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8795500703015154"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "clf = KNeighborsClassifier(n_neighbors=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(n_neighbors=1)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8795500703015154"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 전체 데이터를 이용하면 ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = size_korea[['키','몸무게']].values\n",
    "y = size_korea.성별.replace({'남':0,'여':1})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RandomForestClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9923449461021715"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "clf = RandomForestClassifier(random_state=0)\n",
    "clf.fit(X,y)\n",
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 다른 모델들에도 적용해 보면 ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8864239962505859"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "clf = LogisticRegression()\n",
    "clf.fit(X,y)\n",
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### GaussianNB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.884393063583815"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "clf = GaussianNB()\n",
    "clf.fit(X,y)\n",
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8826745820965474"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "clf = SVC()\n",
    "clf.fit(X,y)\n",
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### KNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9181377909701609"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "clf = KNeighborsClassifier(n_neighbors=3)\n",
    "clf.fit(X,y)\n",
    "(clf.predict(size_korea[['키','몸무게']].values) == size_korea.성별.replace({'남':0,'여':1})).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 같이 비교해 보기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "clfs = []\n",
    "clfs.append(RandomForestClassifier(random_state=0))\n",
    "clfs.append(LogisticRegression())\n",
    "clfs.append(GaussianNB())\n",
    "clfs.append(SVC())\n",
    "clfs.append(KNeighborsClassifier(n_neighbors=3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "sk_data = {}\n",
    "sk_data['mean'] = {'X': size_korea.groupby('성별')[['키','몸무게']].mean().values, 'y': [0,1]}\n",
    "sk_data['all'] = {'X': size_korea[['키','몸무게']].values, 'y': size_korea.성별.replace({'남':0,'여':1})}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Expected n_neighbors <= n_samples,  but n_samples = 2, n_neighbors = 3\n"
     ]
    }
   ],
   "source": [
    "result = {}\n",
    "for data_type in ['mean', 'all']:\n",
    "    X = sk_data[data_type]['X']\n",
    "    y = sk_data[data_type]['y']\n",
    "    for clf in clfs:\n",
    "        try:\n",
    "            clf.fit(X,y)\n",
    "            result[data_type, clf.__class__.__name__] = (clf.predict(size_korea[['키','몸무게']].values) \n",
    "                                                         == size_korea.성별.replace({'남':0,'여':1})).mean()\n",
    "        except Exception as e:\n",
    "            print(e)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "scrolled": true
   },
   "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>all</th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>GaussianNB</th>\n",
       "      <td>0.884393</td>\n",
       "      <td>0.879550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KNeighborsClassifier</th>\n",
       "      <td>0.918138</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LogisticRegression</th>\n",
       "      <td>0.886424</td>\n",
       "      <td>0.691767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RandomForestClassifier</th>\n",
       "      <td>0.992345</td>\n",
       "      <td>0.877207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SVC</th>\n",
       "      <td>0.882675</td>\n",
       "      <td>0.879550</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             all      mean\n",
       "GaussianNB              0.884393  0.879550\n",
       "KNeighborsClassifier    0.918138       NaN\n",
       "LogisticRegression      0.886424  0.691767\n",
       "RandomForestClassifier  0.992345  0.877207\n",
       "SVC                     0.882675  0.879550"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Series(result).unstack().T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 메모리에 저장한 후 그 결과를 보여주는 것은 아닌가?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = size_korea[['키','몸무게','체지방율']].values\n",
    "y = size_korea.성별.replace({'남':0,'여':1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([[1736. ,   72.1,   18.9],\n",
       "        [1833. ,  106.2,   27. ],\n",
       "        [1744. ,   69.1,    8.1],\n",
       "        [1751. ,   68.4,   10.3],\n",
       "        [1851. ,   81.9,   14.1]]),\n",
       " 0    0\n",
       " 1    0\n",
       " 2    0\n",
       " 3    0\n",
       " 4    0\n",
       " Name: 성별, dtype: int64)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X[:5], y[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 짝수번째 데이터만 추출"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = X[::2]\n",
    "y_train = y[::2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((3201, 3), (3201,))"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.shape, y_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    1604\n",
       "0    1597\n",
       "Name: 성별, dtype: int64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(n_neighbors=3)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 홀수번째 데이터만 추출"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test = X[1::2]\n",
    "y_test = y[1::2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((3200, 3), (3200,))"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 1, 1, 1], dtype=int64)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.predict(X_test) \n",
    "y_test.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3200"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(X_test) == y_test).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.96125"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(X_test) == y_test).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 어떤 항목들이 맞지 않나?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "61      False\n",
       "67      False\n",
       "73      False\n",
       "211     False\n",
       "227     False\n",
       "        ...  \n",
       "5981    False\n",
       "6019    False\n",
       "6319    False\n",
       "6361    False\n",
       "6379    False\n",
       "Name: 성별, Length: 124, dtype: bool"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(X_test) == y_test).loc[lambda x: x==False]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## train_test_split 활용"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = size_korea[['키','몸무게','체지방율']].values\n",
    "y = size_korea.성별.replace({'남':0,'여':1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9633099141295862"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X_train, y_train)\n",
    "clf.predict(X_test) == y_test.values\n",
    "(clf.predict(X_test) == y_test.values).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 데이터 변환\n",
    "### Transformers and pre-processors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 표준화 n(0,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1722.4153074    70.95288582   19.11718946]\n",
      " [1587.07065048   56.80666044   30.4881108 ]]\n"
     ]
    }
   ],
   "source": [
    "X = size_korea.groupby('성별')[['키','몸무게','체지방율']].mean().values\n",
    "print(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 1.,  1., -1.],\n",
       "       [-1., -1.,  1.]])"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "# scale data according to computed scaling values\n",
    "StandardScaler().fit(X).transform(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "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>키</th>\n",
       "      <th>몸무게</th>\n",
       "      <th>체지방율</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1736.0</td>\n",
       "      <td>72.1</td>\n",
       "      <td>18.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1833.0</td>\n",
       "      <td>106.2</td>\n",
       "      <td>27.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1744.0</td>\n",
       "      <td>69.1</td>\n",
       "      <td>8.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1751.0</td>\n",
       "      <td>68.4</td>\n",
       "      <td>10.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1851.0</td>\n",
       "      <td>81.9</td>\n",
       "      <td>14.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        키    몸무게  체지방율\n",
       "0  1736.0   72.1  18.9\n",
       "1  1833.0  106.2  27.0\n",
       "2  1744.0   69.1   8.1\n",
       "3  1751.0   68.4  10.3\n",
       "4  1851.0   81.9  14.1"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "size_korea[['키','몸무게','체지방율']].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 왜 데이터 변환이나 표준화를 하는가?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### data transform 없이"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = size_korea[['키','몸무게','체지방율']].values\n",
    "y = size_korea.성별.replace({'남':0,'여':1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(n_neighbors=3)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9633099141295862"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.predict(X_test)\n",
    "(clf.predict(X_test) == y_test).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### data transform 이후 fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "X_train_std = StandardScaler().fit(X_train).transform(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:ylabel='Frequency'>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.Series(X_train_std[:,0]).plot.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:ylabel='Frequency'>"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.Series(X_train[:,0]).plot.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KNeighborsClassifier(n_neighbors=3)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X_train_std, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test_std = StandardScaler().fit(X_train).transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9648711943793911"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.predict(X_test_std)\n",
    "(clf.predict(X_test_std) == y_test).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 데이터 변환과 fitting을 한번에 하는 방법은 없나?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Pipelines: chaining pre-processors and estimators\n",
    "<font color='blue'> Transformers and estimators (predictors) can be combined together into a single unifying object: a Pipeline."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "# X, y 정의\n",
    "X = size_korea[['키','몸무게','체지방율']].values\n",
    "y = size_korea.성별.replace({'남':0,'여':1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1    0.501953\n",
      "0    0.498047\n",
      "Name: 성별, dtype: float64\n",
      "1    0.503958\n",
      "0    0.496042\n",
      "Name: 성별, dtype: float64\n",
      "0    0.50406\n",
      "1    0.49594\n",
      "Name: 성별, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape\n",
    "\n",
    "print(pd.Series(y).value_counts(normalize=True))\n",
    "print(pd.Series(y_train).value_counts(normalize=True))\n",
    "print(pd.Series(y_test).value_counts(normalize=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 층화추출: stratify option"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1    0.501953\n",
      "0    0.498047\n",
      "Name: 성별, dtype: float64\n",
      "1    0.501875\n",
      "0    0.498125\n",
      "Name: 성별, dtype: float64\n",
      "1    0.502186\n",
      "0    0.497814\n",
      "Name: 성별, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, random_state=0)\n",
    "\n",
    "print(pd.Series(y).value_counts(normalize=True))\n",
    "print(pd.Series(y_train).value_counts(normalize=True))\n",
    "print(pd.Series(y_test).value_counts(normalize=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### make_pipeline 활용"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('standardscaler', StandardScaler()),\n",
       "                ('logisticregression', LogisticRegression())])"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "# create a pipeline object\n",
    "pipe = make_pipeline(\n",
    "    StandardScaler(),\n",
    "    LogisticRegression()\n",
    ")\n",
    "\n",
    "# fit the whole pipeline\n",
    "pipe.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3406    0\n",
       "6222    1\n",
       "194     1\n",
       "756     0\n",
       "5531    1\n",
       "       ..\n",
       "3504    0\n",
       "4449    0\n",
       "4528    0\n",
       "2442    0\n",
       "4049    0\n",
       "Name: 성별, Length: 1601, dtype: int64"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.predict(X_test)\n",
    "y_test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 얼마나 맞았나?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9537788881948782"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(pipe.predict(X_test) == y_test).mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### accuracy_score 활용"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9537788881948782"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(pipe.predict(X_test), y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### model.score 활용"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9537788881948782"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 운 좋게 맞은 것은 아닌가?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cross validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "clf = RandomForestClassifier(random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = size_korea[['키','몸무게','체지방율']].values\n",
    "y = size_korea.성별.replace({'남':0,'여':1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "kf = KFold(n_splits=5, shuffle=True, random_state=11)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(array([   0,    2,    3, ..., 6396, 6399, 6400]),\n",
       "  array([   1,    5,   11, ..., 6387, 6397, 6398])),\n",
       " (array([   0,    1,    2, ..., 6397, 6398, 6400]),\n",
       "  array([  12,   18,   21, ..., 6381, 6389, 6399])),\n",
       " (array([   0,    1,    2, ..., 6398, 6399, 6400]),\n",
       "  array([   3,    6,    8, ..., 6393, 6395, 6396])),\n",
       " (array([   0,    1,    3, ..., 6398, 6399, 6400]),\n",
       "  array([   2,   13,   20, ..., 6390, 6391, 6392])),\n",
       " (array([   1,    2,    3, ..., 6397, 6398, 6399]),\n",
       "  array([   0,    4,    7, ..., 6386, 6394, 6400]))]"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(kf.split(X))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((5120,), (1281,))"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "idx = list(kf.split(X))\n",
    "idx[0][0].shape, idx[0][1].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6401"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(idx[0][0]) | set(idx[0][1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### MECE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6401"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(set(idx[0][1]) | set(idx[1][1]) | set(idx[2][1]) | set(idx[3][1]) | set(idx[4][1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### kfold index들을 이용해서 결과 --> 평균"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.9633099141295862, 0.9671875, 0.95859375, 0.95859375, 0.965625]\n",
      "0.9626619828259171\n"
     ]
    }
   ],
   "source": [
    "accuracy_list = []\n",
    "for train_index, test_index in kf.split(X):\n",
    "    X_train, y_train = X[train_index], y[train_index]\n",
    "    X_test,  y_test  = X[test_index],  y[test_index]    \n",
    "    y_predict = clf.fit(X_train, y_train).predict(X_test)\n",
    "    accuracy_list.append(accuracy_score(y_test, y_predict))\n",
    "    \n",
    "print(accuracy_list)\n",
    "print(np.array(accuracy_list).mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### cross_validate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'fit_time': array([1.04322648, 1.04491305, 1.03525901, 0.99767184, 0.98934841]),\n",
       " 'score_time': array([0.05774903, 0.06580758, 0.04149842, 0.06094098, 0.06660724]),\n",
       " 'test_score': array([0.96330991, 0.9671875 , 0.95859375, 0.95859375, 0.965625  ])}"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_validate\n",
    "result = cross_validate(clf, X, y, cv=kf)  # defaults to 5-fold CV\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9626619828259171"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result['test_score'].mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 좋은 결과를 내는 모델 튜닝\n",
    "### Automatic parameter searches"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### RandomizedSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import RandomizedSearchCV, train_test_split\n",
    "from scipy.stats import randint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = size_korea[['키','몸무게','체지방율']].values\n",
    "y = size_korea.성별.replace({'남':0,'여':1})\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bootstrap': True,\n",
       " 'ccp_alpha': 0.0,\n",
       " 'class_weight': None,\n",
       " 'criterion': 'gini',\n",
       " 'max_depth': None,\n",
       " 'max_features': 'auto',\n",
       " 'max_leaf_nodes': None,\n",
       " 'max_samples': None,\n",
       " 'min_impurity_decrease': 0.0,\n",
       " 'min_impurity_split': None,\n",
       " 'min_samples_leaf': 1,\n",
       " 'min_samples_split': 2,\n",
       " 'min_weight_fraction_leaf': 0.0,\n",
       " 'n_estimators': 100,\n",
       " 'n_jobs': None,\n",
       " 'oob_score': False,\n",
       " 'random_state': 0,\n",
       " 'verbose': 0,\n",
       " 'warm_start': False}"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "RandomForestClassifier(random_state=0).get_params()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 5 candidates, totalling 25 fits\n",
      "[CV] END ........................................max_depth=9; total time=   0.8s\n",
      "[CV] END ........................................max_depth=9; total time=   0.8s\n",
      "[CV] END ........................................max_depth=9; total time=   0.8s\n",
      "[CV] END ........................................max_depth=9; total time=   0.8s\n",
      "[CV] END ........................................max_depth=9; total time=   0.9s\n",
      "[CV] END ........................................max_depth=5; total time=   0.6s\n",
      "[CV] END ........................................max_depth=5; total time=   0.6s\n",
      "[CV] END ........................................max_depth=5; total time=   0.6s\n",
      "[CV] END ........................................max_depth=5; total time=   0.6s\n",
      "[CV] END ........................................max_depth=5; total time=   0.6s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.7s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.7s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.7s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n",
      "[CV] END ........................................max_depth=8; total time=   0.8s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.9643972517176764"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# define the parameter space that will be randomizeded over\n",
    "param_distributions = {'max_depth': randint(5, 10)}\n",
    "\n",
    "# now create a randomizedCV object and fit it to the data\n",
    "randomized = RandomizedSearchCV(estimator=RandomForestClassifier(random_state=0),\n",
    "                            n_iter=5,\n",
    "                            param_distributions=param_distributions,\n",
    "                            random_state=0, verbose=2)\n",
    "randomized.fit(X_train, y_train)\n",
    "\n",
    "randomized.best_params_\n",
    "\n",
    "# the randomized object now acts like a normal random forest estimator\n",
    "# with max_depth=9 and n_estimators=4\n",
    "randomized.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0, 1, ..., 0, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "randomized.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### GridSearchCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "knn = KNeighborsClassifier(n_neighbors=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "hyperparameters = {'n_neighbors': [1, 2, 3, 11, 17],\n",
    "                   'weights': ['uniform', 'distance'],\n",
    "                   'leaf_size': [4, 6, 8, 10]}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, estimator=KNeighborsClassifier(n_neighbors=3),\n",
       "             param_grid={'leaf_size': [4, 6, 8, 10],\n",
       "                         'n_neighbors': [1, 2, 3, 11, 17],\n",
       "                         'weights': ['uniform', 'distance']},\n",
       "             scoring='accuracy')"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "grid = GridSearchCV(knn, param_grid = hyperparameters,\n",
    "                           cv=5, scoring='accuracy')\n",
    "grid.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'algorithm': 'auto', 'leaf_size': 4, 'metric': 'minkowski', 'metric_params': None, 'n_jobs': None, 'n_neighbors': 3, 'p': 2, 'weights': 'uniform'}\n",
      "optimal train score: 0.960\n",
      "test score: 0.964\n",
      "optimal parameter: {'leaf_size': 4, 'n_neighbors': 3, 'weights': 'uniform'}\n",
      "{'mean_fit_time': array([0.01348696, 0.01409655, 0.01374249, 0.01293592, 0.01561203,\n",
      "       0.01280661, 0.01310496, 0.01444745, 0.0131619 , 0.01137433,\n",
      "       0.01338525, 0.01414475, 0.01504083, 0.01219964, 0.01407762,\n",
      "       0.01191216, 0.01133919, 0.0124042 , 0.01280227, 0.01444235,\n",
      "       0.01195507, 0.01079011, 0.01278644, 0.01193871, 0.01083179,\n",
      "       0.01083784, 0.01120176, 0.01242375, 0.01224656, 0.01212802,\n",
      "       0.01188974, 0.01233621, 0.01216798, 0.01189861, 0.01246867,\n",
      "       0.01204872, 0.01196327, 0.0121253 , 0.01221166, 0.01229758]), 'std_fit_time': array([0.00267374, 0.00207761, 0.001377  , 0.00141543, 0.00309978,\n",
      "       0.00218967, 0.00178592, 0.00021973, 0.00271612, 0.00149698,\n",
      "       0.00129275, 0.00079712, 0.00103638, 0.00280274, 0.00050456,\n",
      "       0.00177571, 0.00181879, 0.00191864, 0.00239393, 0.00050359,\n",
      "       0.00074711, 0.00309844, 0.0004447 , 0.00032719, 0.00215367,\n",
      "       0.00247523, 0.00227186, 0.00071311, 0.0005601 , 0.00022281,\n",
      "       0.00038462, 0.00017093, 0.00051049, 0.00050767, 0.00021098,\n",
      "       0.00065609, 0.00070659, 0.00017389, 0.00066486, 0.00038854]), 'mean_score_time': array([0.06480379, 0.01120329, 0.07686639, 0.01198812, 0.07248249,\n",
      "       0.01414552, 0.07668233, 0.02180004, 0.08237176, 0.02359982,\n",
      "       0.0729207 , 0.00951233, 0.06817517, 0.01143169, 0.06985011,\n",
      "       0.01336961, 0.07099895, 0.02098265, 0.08727183, 0.02341866,\n",
      "       0.06400967, 0.02464018, 0.06482797, 0.01066246, 0.0694037 ,\n",
      "       0.01125264, 0.07666111, 0.01765246, 0.07420092, 0.02129207,\n",
      "       0.0586082 , 0.00879612, 0.06136031, 0.01015706, 0.06221032,\n",
      "       0.01103244, 0.06733651, 0.01715178, 0.07166362, 0.02151904]), 'std_score_time': array([0.00376483, 0.00032351, 0.00280572, 0.00059523, 0.00283079,\n",
      "       0.00087031, 0.00497307, 0.00081589, 0.00626143, 0.00302833,\n",
      "       0.00555427, 0.00227885, 0.00285529, 0.00197401, 0.00627356,\n",
      "       0.00076697, 0.01117041, 0.00104513, 0.00264381, 0.00351518,\n",
      "       0.00522014, 0.03142773, 0.00680914, 0.00185709, 0.00338267,\n",
      "       0.00196014, 0.00292505, 0.00037046, 0.00391924, 0.00089665,\n",
      "       0.00068076, 0.00041984, 0.00147289, 0.00016128, 0.00188673,\n",
      "       0.00074366, 0.0007941 , 0.00040305, 0.00104062, 0.00063826]), 'param_leaf_size': masked_array(data=[4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 6, 6, 6, 6, 6, 6, 6, 6,\n",
      "                   6, 6, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 10, 10, 10, 10, 10,\n",
      "                   10, 10, 10, 10, 10],\n",
      "             mask=[False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False],\n",
      "       fill_value='?',\n",
      "            dtype=object), 'param_n_neighbors': masked_array(data=[1, 1, 2, 2, 3, 3, 11, 11, 17, 17, 1, 1, 2, 2, 3, 3, 11,\n",
      "                   11, 17, 17, 1, 1, 2, 2, 3, 3, 11, 11, 17, 17, 1, 1, 2,\n",
      "                   2, 3, 3, 11, 11, 17, 17],\n",
      "             mask=[False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False],\n",
      "       fill_value='?',\n",
      "            dtype=object), 'param_weights': masked_array(data=['uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance',\n",
      "                   'uniform', 'distance', 'uniform', 'distance'],\n",
      "             mask=[False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False,\n",
      "                   False, False, False, False, False, False, False, False],\n",
      "       fill_value='?',\n",
      "            dtype=object), 'params': [{'leaf_size': 4, 'n_neighbors': 1, 'weights': 'uniform'}, {'leaf_size': 4, 'n_neighbors': 1, 'weights': 'distance'}, {'leaf_size': 4, 'n_neighbors': 2, 'weights': 'uniform'}, {'leaf_size': 4, 'n_neighbors': 2, 'weights': 'distance'}, {'leaf_size': 4, 'n_neighbors': 3, 'weights': 'uniform'}, {'leaf_size': 4, 'n_neighbors': 3, 'weights': 'distance'}, {'leaf_size': 4, 'n_neighbors': 11, 'weights': 'uniform'}, {'leaf_size': 4, 'n_neighbors': 11, 'weights': 'distance'}, {'leaf_size': 4, 'n_neighbors': 17, 'weights': 'uniform'}, {'leaf_size': 4, 'n_neighbors': 17, 'weights': 'distance'}, {'leaf_size': 6, 'n_neighbors': 1, 'weights': 'uniform'}, {'leaf_size': 6, 'n_neighbors': 1, 'weights': 'distance'}, {'leaf_size': 6, 'n_neighbors': 2, 'weights': 'uniform'}, {'leaf_size': 6, 'n_neighbors': 2, 'weights': 'distance'}, {'leaf_size': 6, 'n_neighbors': 3, 'weights': 'uniform'}, {'leaf_size': 6, 'n_neighbors': 3, 'weights': 'distance'}, {'leaf_size': 6, 'n_neighbors': 11, 'weights': 'uniform'}, {'leaf_size': 6, 'n_neighbors': 11, 'weights': 'distance'}, {'leaf_size': 6, 'n_neighbors': 17, 'weights': 'uniform'}, {'leaf_size': 6, 'n_neighbors': 17, 'weights': 'distance'}, {'leaf_size': 8, 'n_neighbors': 1, 'weights': 'uniform'}, {'leaf_size': 8, 'n_neighbors': 1, 'weights': 'distance'}, {'leaf_size': 8, 'n_neighbors': 2, 'weights': 'uniform'}, {'leaf_size': 8, 'n_neighbors': 2, 'weights': 'distance'}, {'leaf_size': 8, 'n_neighbors': 3, 'weights': 'uniform'}, {'leaf_size': 8, 'n_neighbors': 3, 'weights': 'distance'}, {'leaf_size': 8, 'n_neighbors': 11, 'weights': 'uniform'}, {'leaf_size': 8, 'n_neighbors': 11, 'weights': 'distance'}, {'leaf_size': 8, 'n_neighbors': 17, 'weights': 'uniform'}, {'leaf_size': 8, 'n_neighbors': 17, 'weights': 'distance'}, {'leaf_size': 10, 'n_neighbors': 1, 'weights': 'uniform'}, {'leaf_size': 10, 'n_neighbors': 1, 'weights': 'distance'}, {'leaf_size': 10, 'n_neighbors': 2, 'weights': 'uniform'}, {'leaf_size': 10, 'n_neighbors': 2, 'weights': 'distance'}, {'leaf_size': 10, 'n_neighbors': 3, 'weights': 'uniform'}, {'leaf_size': 10, 'n_neighbors': 3, 'weights': 'distance'}, {'leaf_size': 10, 'n_neighbors': 11, 'weights': 'uniform'}, {'leaf_size': 10, 'n_neighbors': 11, 'weights': 'distance'}, {'leaf_size': 10, 'n_neighbors': 17, 'weights': 'uniform'}, {'leaf_size': 10, 'n_neighbors': 17, 'weights': 'distance'}], 'split0_test_score': array([0.94166667, 0.94166667, 0.93541667, 0.94166667, 0.94895833,\n",
      "       0.946875  , 0.94166667, 0.94479167, 0.93958333, 0.94375   ,\n",
      "       0.94166667, 0.94166667, 0.93541667, 0.94166667, 0.94895833,\n",
      "       0.946875  , 0.94166667, 0.94479167, 0.93958333, 0.94375   ,\n",
      "       0.94166667, 0.94166667, 0.93541667, 0.94166667, 0.94895833,\n",
      "       0.946875  , 0.94166667, 0.94479167, 0.93958333, 0.94375   ,\n",
      "       0.94166667, 0.94166667, 0.93541667, 0.94166667, 0.94895833,\n",
      "       0.946875  , 0.94166667, 0.94479167, 0.93958333, 0.94375   ]), 'split1_test_score': array([0.94791667, 0.94791667, 0.94270833, 0.94791667, 0.95625   ,\n",
      "       0.953125  , 0.95833333, 0.96041667, 0.95520833, 0.95729167,\n",
      "       0.94791667, 0.94791667, 0.94270833, 0.94791667, 0.95625   ,\n",
      "       0.953125  , 0.95833333, 0.96041667, 0.95520833, 0.95729167,\n",
      "       0.94791667, 0.94791667, 0.94270833, 0.94791667, 0.95625   ,\n",
      "       0.953125  , 0.95833333, 0.96041667, 0.95520833, 0.95729167,\n",
      "       0.94791667, 0.94791667, 0.94270833, 0.94791667, 0.95625   ,\n",
      "       0.953125  , 0.95833333, 0.96041667, 0.95520833, 0.95729167]), 'split2_test_score': array([0.9625    , 0.9625    , 0.94583333, 0.9625    , 0.959375  ,\n",
      "       0.96041667, 0.9625    , 0.96666667, 0.953125  , 0.96041667,\n",
      "       0.9625    , 0.9625    , 0.94583333, 0.9625    , 0.959375  ,\n",
      "       0.96041667, 0.9625    , 0.96666667, 0.953125  , 0.96041667,\n",
      "       0.9625    , 0.9625    , 0.94583333, 0.9625    , 0.959375  ,\n",
      "       0.96041667, 0.9625    , 0.96666667, 0.953125  , 0.96041667,\n",
      "       0.9625    , 0.9625    , 0.94583333, 0.9625    , 0.959375  ,\n",
      "       0.96041667, 0.9625    , 0.96666667, 0.953125  , 0.96041667]), 'split3_test_score': array([0.95208333, 0.95208333, 0.94375   , 0.95208333, 0.9625    ,\n",
      "       0.959375  , 0.95729167, 0.95416667, 0.95520833, 0.95729167,\n",
      "       0.95208333, 0.95208333, 0.94375   , 0.95208333, 0.9625    ,\n",
      "       0.959375  , 0.95729167, 0.95416667, 0.95520833, 0.95729167,\n",
      "       0.95208333, 0.95208333, 0.94375   , 0.95208333, 0.9625    ,\n",
      "       0.959375  , 0.95729167, 0.95416667, 0.95520833, 0.95729167,\n",
      "       0.95208333, 0.95208333, 0.94375   , 0.95208333, 0.9625    ,\n",
      "       0.959375  , 0.95729167, 0.95416667, 0.95520833, 0.95729167]), 'split4_test_score': array([0.946875  , 0.946875  , 0.95520833, 0.946875  , 0.97083333,\n",
      "       0.96979167, 0.96041667, 0.96458333, 0.95729167, 0.959375  ,\n",
      "       0.946875  , 0.946875  , 0.95520833, 0.946875  , 0.97083333,\n",
      "       0.96979167, 0.96041667, 0.96458333, 0.95729167, 0.959375  ,\n",
      "       0.946875  , 0.946875  , 0.95520833, 0.946875  , 0.97083333,\n",
      "       0.96979167, 0.96041667, 0.96458333, 0.95729167, 0.959375  ,\n",
      "       0.946875  , 0.946875  , 0.95520833, 0.946875  , 0.97083333,\n",
      "       0.96979167, 0.96041667, 0.96458333, 0.95729167, 0.959375  ]), 'mean_test_score': array([0.95020833, 0.95020833, 0.94458333, 0.95020833, 0.95958333,\n",
      "       0.95791667, 0.95604167, 0.958125  , 0.95208333, 0.955625  ,\n",
      "       0.95020833, 0.95020833, 0.94458333, 0.95020833, 0.95958333,\n",
      "       0.95791667, 0.95604167, 0.958125  , 0.95208333, 0.955625  ,\n",
      "       0.95020833, 0.95020833, 0.94458333, 0.95020833, 0.95958333,\n",
      "       0.95791667, 0.95604167, 0.958125  , 0.95208333, 0.955625  ,\n",
      "       0.95020833, 0.95020833, 0.94458333, 0.95020833, 0.95958333,\n",
      "       0.95791667, 0.95604167, 0.958125  , 0.95208333, 0.955625  ]), 'std_test_score': array([0.00698461, 0.00698461, 0.00636696, 0.00698461, 0.00719881,\n",
      "       0.00767165, 0.00740683, 0.00791667, 0.00638738, 0.0060596 ,\n",
      "       0.00698461, 0.00698461, 0.00636696, 0.00698461, 0.00719881,\n",
      "       0.00767165, 0.00740683, 0.00791667, 0.00638738, 0.0060596 ,\n",
      "       0.00698461, 0.00698461, 0.00636696, 0.00698461, 0.00719881,\n",
      "       0.00767165, 0.00740683, 0.00791667, 0.00638738, 0.0060596 ,\n",
      "       0.00698461, 0.00698461, 0.00636696, 0.00698461, 0.00719881,\n",
      "       0.00767165, 0.00740683, 0.00791667, 0.00638738, 0.0060596 ]), 'rank_test_score': array([25, 25, 37, 25,  1,  9, 13,  5, 21, 17, 25, 25, 37, 25,  1,  9, 13,\n",
      "        5, 21, 17, 25, 25, 37, 25,  1,  9, 13,  5, 21, 17, 25, 25, 37, 25,\n",
      "        1,  9, 13,  5, 21, 17])}\n"
     ]
    }
   ],
   "source": [
    "print(grid.best_estimator_.get_params())\n",
    "print(f'optimal train score: {grid.best_score_:.3f}')\n",
    "print(f'test score: {grid.score(X_test, y_test):.3f}')\n",
    "print(f'optimal parameter: {grid.best_params_}')\n",
    "print(grid.cv_results_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 모델로 예측하기 위해 매번 학습시켜야 하나?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['../working/randomized.joblib']"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from joblib import dump, load\n",
    "dump(randomized, '../working/randomized.joblib')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " D 드라이브의 볼륨에는 이름이 없습니다.\n",
      " 볼륨 일련 번호: 72F1-9AEA\n",
      "\n",
      " d:\\2021_데이터솔루션\\working 디렉터리\n",
      "\n",
      "2021-03-30 화  오후 08:06    <DIR>          .\n",
      "2021-03-30 화  오후 08:06    <DIR>          ..\n",
      "2021-03-30 화  오후 08:11             9,743 인체.xlsx\n",
      "2021-03-30 화  오후 08:11            14,894 인체_choice.xlsx\n",
      "2021-08-10 화  오후 10:29             9,468 Feature_importance.png\n",
      "2021-08-20 금  오후 07:12           300,690 submission.csv\n",
      "2021-08-22 일  오전 11:05         1,818,124 randomized.joblib\n",
      "               5개 파일           2,152,919 바이트\n",
      "               2개 디렉터리  93,063,217,152 바이트 남음\n"
     ]
    }
   ],
   "source": [
    "!dir ..\\working"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'cv': None,\n",
       " 'error_score': nan,\n",
       " 'estimator__bootstrap': True,\n",
       " 'estimator__ccp_alpha': 0.0,\n",
       " 'estimator__class_weight': None,\n",
       " 'estimator__criterion': 'gini',\n",
       " 'estimator__max_depth': None,\n",
       " 'estimator__max_features': 'auto',\n",
       " 'estimator__max_leaf_nodes': None,\n",
       " 'estimator__max_samples': None,\n",
       " 'estimator__min_impurity_decrease': 0.0,\n",
       " 'estimator__min_impurity_split': None,\n",
       " 'estimator__min_samples_leaf': 1,\n",
       " 'estimator__min_samples_split': 2,\n",
       " 'estimator__min_weight_fraction_leaf': 0.0,\n",
       " 'estimator__n_estimators': 100,\n",
       " 'estimator__n_jobs': None,\n",
       " 'estimator__oob_score': False,\n",
       " 'estimator__random_state': 0,\n",
       " 'estimator__verbose': 0,\n",
       " 'estimator__warm_start': False,\n",
       " 'estimator': RandomForestClassifier(random_state=0),\n",
       " 'n_iter': 5,\n",
       " 'n_jobs': None,\n",
       " 'param_distributions': {'max_depth': <scipy.stats._distn_infrastructure.rv_frozen at 0x22a57b2d610>},\n",
       " 'pre_dispatch': '2*n_jobs',\n",
       " 'random_state': 0,\n",
       " 'refit': True,\n",
       " 'return_train_score': False,\n",
       " 'scoring': None,\n",
       " 'verbose': 2}"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = load('../working/randomized.joblib')\n",
    "model.get_params()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1591. ,   62.2,   38.6],\n",
       "       [1667. ,   67.5,   21.8],\n",
       "       [1495. ,   43. ,   25.9],\n",
       "       ...,\n",
       "       [1737. ,   71.8,   23.4],\n",
       "       [1699. ,   62.1,   13.4],\n",
       "       [1801. ,   91.6,   26.8]])"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 0, 1, ..., 0, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(max_depth=8, random_state=0)"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.best_estimator_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# End"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 실습: pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv('../data/size_korea_w_null.csv')\n",
    "# 데이터 정제 및 변환\n",
    "X = data.drop('성별', axis=1).replace({'표준': -1, '표준이하': 0, '표준이상': 1})\n",
    "X = X.fillna(X.mean())\n",
    "y = data['성별'].replace({'남':0,'여':1})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{-1.0: -1, 0.0: 0, 1.0: 1}"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict(zip(X.기초대사량평가.value_counts().index,[-1,0,1]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_selection import RFE\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "lr = LogisticRegression()\n",
    "rfe_selector = RFE(estimator=lr, n_features_to_select=5, step=20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "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>키</th>\n",
       "      <th>몸무게</th>\n",
       "      <th>목둘레</th>\n",
       "      <th>허리둘레</th>\n",
       "      <th>BMI</th>\n",
       "      <th>체지방율</th>\n",
       "      <th>기초대사량</th>\n",
       "      <th>기초대사량평가</th>\n",
       "      <th>머리수직길이</th>\n",
       "      <th>머리두께</th>\n",
       "      <th>머리너비</th>\n",
       "      <th>발너비</th>\n",
       "      <th>발직선길이</th>\n",
       "      <th>나이</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1736.0</td>\n",
       "      <td>72.1</td>\n",
       "      <td>385.0</td>\n",
       "      <td>819.0</td>\n",
       "      <td>23.924112</td>\n",
       "      <td>18.9</td>\n",
       "      <td>1633.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>224.0</td>\n",
       "      <td>188.0</td>\n",
       "      <td>159.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>228.0</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1833.0</td>\n",
       "      <td>106.2</td>\n",
       "      <td>418.0</td>\n",
       "      <td>1036.0</td>\n",
       "      <td>31.608187</td>\n",
       "      <td>27.0</td>\n",
       "      <td>2044.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>247.0</td>\n",
       "      <td>201.0</td>\n",
       "      <td>161.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>269.0</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1744.0</td>\n",
       "      <td>69.1</td>\n",
       "      <td>400.0</td>\n",
       "      <td>754.0</td>\n",
       "      <td>22.718784</td>\n",
       "      <td>8.1</td>\n",
       "      <td>1742.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>251.0</td>\n",
       "      <td>201.0</td>\n",
       "      <td>162.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>244.0</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1751.0</td>\n",
       "      <td>68.4</td>\n",
       "      <td>355.0</td>\n",
       "      <td>720.0</td>\n",
       "      <td>22.309190</td>\n",
       "      <td>10.3</td>\n",
       "      <td>1695.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>234.0</td>\n",
       "      <td>199.0</td>\n",
       "      <td>154.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>231.0</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1851.0</td>\n",
       "      <td>81.9</td>\n",
       "      <td>394.0</td>\n",
       "      <td>846.0</td>\n",
       "      <td>23.904027</td>\n",
       "      <td>14.1</td>\n",
       "      <td>1890.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>223.0</td>\n",
       "      <td>195.0</td>\n",
       "      <td>158.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>265.0</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6415</th>\n",
       "      <td>1706.0</td>\n",
       "      <td>59.6</td>\n",
       "      <td>340.0</td>\n",
       "      <td>699.0</td>\n",
       "      <td>20.478031</td>\n",
       "      <td>30.9</td>\n",
       "      <td>1259.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>222.0</td>\n",
       "      <td>175.0</td>\n",
       "      <td>161.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>242.0</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6416</th>\n",
       "      <td>1624.0</td>\n",
       "      <td>54.9</td>\n",
       "      <td>321.0</td>\n",
       "      <td>765.0</td>\n",
       "      <td>20.816145</td>\n",
       "      <td>27.7</td>\n",
       "      <td>1228.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>243.0</td>\n",
       "      <td>179.0</td>\n",
       "      <td>151.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>241.0</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6417</th>\n",
       "      <td>1508.0</td>\n",
       "      <td>50.1</td>\n",
       "      <td>325.0</td>\n",
       "      <td>782.0</td>\n",
       "      <td>22.031042</td>\n",
       "      <td>30.7</td>\n",
       "      <td>1120.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>214.0</td>\n",
       "      <td>170.0</td>\n",
       "      <td>142.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>228.0</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6418</th>\n",
       "      <td>1574.0</td>\n",
       "      <td>44.6</td>\n",
       "      <td>305.0</td>\n",
       "      <td>630.0</td>\n",
       "      <td>18.002193</td>\n",
       "      <td>25.4</td>\n",
       "      <td>1089.0</td>\n",
       "      <td>-1.0</td>\n",
       "      <td>221.0</td>\n",
       "      <td>173.0</td>\n",
       "      <td>157.0</td>\n",
       "      <td>90.0</td>\n",
       "      <td>222.0</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6419</th>\n",
       "      <td>1550.0</td>\n",
       "      <td>62.6</td>\n",
       "      <td>350.0</td>\n",
       "      <td>821.0</td>\n",
       "      <td>26.056191</td>\n",
       "      <td>32.2</td>\n",
       "      <td>1287.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>227.0</td>\n",
       "      <td>174.0</td>\n",
       "      <td>154.0</td>\n",
       "      <td>96.0</td>\n",
       "      <td>228.0</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>6420 rows × 14 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           키    몸무게    목둘레    허리둘레        BMI  체지방율   기초대사량  기초대사량평가  머리수직길이  \\\n",
       "0     1736.0   72.1  385.0   819.0  23.924112  18.9  1633.0     -1.0   224.0   \n",
       "1     1833.0  106.2  418.0  1036.0  31.608187  27.0  2044.0      0.0   247.0   \n",
       "2     1744.0   69.1  400.0   754.0  22.718784   8.1  1742.0     -1.0   251.0   \n",
       "3     1751.0   68.4  355.0   720.0  22.309190  10.3  1695.0     -1.0   234.0   \n",
       "4     1851.0   81.9  394.0   846.0  23.904027  14.1  1890.0     -1.0   223.0   \n",
       "...      ...    ...    ...     ...        ...   ...     ...      ...     ...   \n",
       "6415  1706.0   59.6  340.0   699.0  20.478031  30.9  1259.0      0.0   222.0   \n",
       "6416  1624.0   54.9  321.0   765.0  20.816145  27.7  1228.0     -1.0   243.0   \n",
       "6417  1508.0   50.1  325.0   782.0  22.031042  30.7  1120.0     -1.0   214.0   \n",
       "6418  1574.0   44.6  305.0   630.0  18.002193  25.4  1089.0     -1.0   221.0   \n",
       "6419  1550.0   62.6  350.0   821.0  26.056191  32.2  1287.0      0.0   227.0   \n",
       "\n",
       "       머리두께   머리너비   발너비  발직선길이  나이  \n",
       "0     188.0  159.0  88.0  228.0  25  \n",
       "1     201.0  161.0  98.0  269.0  28  \n",
       "2     201.0  162.0  90.0  244.0  19  \n",
       "3     199.0  154.0  88.0  231.0  20  \n",
       "4     195.0  158.0  86.0  265.0  22  \n",
       "...     ...    ...   ...    ...  ..  \n",
       "6415  175.0  161.0  89.0  242.0  20  \n",
       "6416  179.0  151.0  93.0  241.0  20  \n",
       "6417  170.0  142.0  95.0  228.0  44  \n",
       "6418  173.0  157.0  90.0  222.0  20  \n",
       "6419  174.0  154.0  96.0  228.0  37  \n",
       "\n",
       "[6420 rows x 14 columns]"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       0\n",
       "1       0\n",
       "2       0\n",
       "3       0\n",
       "4       0\n",
       "       ..\n",
       "6415    1\n",
       "6416    1\n",
       "6417    1\n",
       "6418    1\n",
       "6419    1\n",
       "Name: 성별, Length: 6420, dtype: int64"
      ]
     },
     "execution_count": 104,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:763: 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",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RFE(estimator=LogisticRegression(), n_features_to_select=5, step=20)"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfe_selector.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([False,  True,  True, False,  True,  True, False, False, False,\n",
       "       False, False,  True, False, False])"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfe_selector.get_support()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['몸무게', '목둘레', 'BMI', '체지방율', '발너비'], dtype='object')"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.columns[rfe_selector.get_support()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['몸무게', '목둘레', 'BMI', '체지방율', '발너비']\n",
      "[False  True  True False  True  True False False False False False  True\n",
      " False False]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:763: 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",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    }
   ],
   "source": [
    "rfe_selector = RFE(estimator=lr, n_features_to_select=5, step=20)\n",
    "rfe_selector.fit(X, y)\n",
    "rfe_support = rfe_selector.get_support()\n",
    "rfe_feature = X.columns[rfe_support].tolist()\n",
    "print(rfe_feature)\n",
    "print(rfe_support)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.pipeline import Pipeline\n",
    "clf = Pipeline(steps=[('feature_selector', rfe_selector),\n",
    "                      ('classifier', lr)])\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:763: 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",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model score: 0.977\n"
     ]
    }
   ],
   "source": [
    "clf.fit(X_train, y_train)\n",
    "print(f\"model score: {clf.score(X_test, y_test):.3f}\" )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['../working/clf.joblib']"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from joblib import dump, load\n",
    "dump(clf, '../working/clf.joblib')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 실습: Dataset loading utilities to practise\n",
    "https://scikit-learn.org/stable/datasets.html#datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'data': array([[5.1, 3.5, 1.4, 0.2],\n",
       "        [4.9, 3. , 1.4, 0.2],\n",
       "        [4.7, 3.2, 1.3, 0.2],\n",
       "        [4.6, 3.1, 1.5, 0.2],\n",
       "        [5. , 3.6, 1.4, 0.2],\n",
       "        [5.4, 3.9, 1.7, 0.4],\n",
       "        [4.6, 3.4, 1.4, 0.3],\n",
       "        [5. , 3.4, 1.5, 0.2],\n",
       "        [4.4, 2.9, 1.4, 0.2],\n",
       "        [4.9, 3.1, 1.5, 0.1],\n",
       "        [5.4, 3.7, 1.5, 0.2],\n",
       "        [4.8, 3.4, 1.6, 0.2],\n",
       "        [4.8, 3. , 1.4, 0.1],\n",
       "        [4.3, 3. , 1.1, 0.1],\n",
       "        [5.8, 4. , 1.2, 0.2],\n",
       "        [5.7, 4.4, 1.5, 0.4],\n",
       "        [5.4, 3.9, 1.3, 0.4],\n",
       "        [5.1, 3.5, 1.4, 0.3],\n",
       "        [5.7, 3.8, 1.7, 0.3],\n",
       "        [5.1, 3.8, 1.5, 0.3],\n",
       "        [5.4, 3.4, 1.7, 0.2],\n",
       "        [5.1, 3.7, 1.5, 0.4],\n",
       "        [4.6, 3.6, 1. , 0.2],\n",
       "        [5.1, 3.3, 1.7, 0.5],\n",
       "        [4.8, 3.4, 1.9, 0.2],\n",
       "        [5. , 3. , 1.6, 0.2],\n",
       "        [5. , 3.4, 1.6, 0.4],\n",
       "        [5.2, 3.5, 1.5, 0.2],\n",
       "        [5.2, 3.4, 1.4, 0.2],\n",
       "        [4.7, 3.2, 1.6, 0.2],\n",
       "        [4.8, 3.1, 1.6, 0.2],\n",
       "        [5.4, 3.4, 1.5, 0.4],\n",
       "        [5.2, 4.1, 1.5, 0.1],\n",
       "        [5.5, 4.2, 1.4, 0.2],\n",
       "        [4.9, 3.1, 1.5, 0.2],\n",
       "        [5. , 3.2, 1.2, 0.2],\n",
       "        [5.5, 3.5, 1.3, 0.2],\n",
       "        [4.9, 3.6, 1.4, 0.1],\n",
       "        [4.4, 3. , 1.3, 0.2],\n",
       "        [5.1, 3.4, 1.5, 0.2],\n",
       "        [5. , 3.5, 1.3, 0.3],\n",
       "        [4.5, 2.3, 1.3, 0.3],\n",
       "        [4.4, 3.2, 1.3, 0.2],\n",
       "        [5. , 3.5, 1.6, 0.6],\n",
       "        [5.1, 3.8, 1.9, 0.4],\n",
       "        [4.8, 3. , 1.4, 0.3],\n",
       "        [5.1, 3.8, 1.6, 0.2],\n",
       "        [4.6, 3.2, 1.4, 0.2],\n",
       "        [5.3, 3.7, 1.5, 0.2],\n",
       "        [5. , 3.3, 1.4, 0.2],\n",
       "        [7. , 3.2, 4.7, 1.4],\n",
       "        [6.4, 3.2, 4.5, 1.5],\n",
       "        [6.9, 3.1, 4.9, 1.5],\n",
       "        [5.5, 2.3, 4. , 1.3],\n",
       "        [6.5, 2.8, 4.6, 1.5],\n",
       "        [5.7, 2.8, 4.5, 1.3],\n",
       "        [6.3, 3.3, 4.7, 1.6],\n",
       "        [4.9, 2.4, 3.3, 1. ],\n",
       "        [6.6, 2.9, 4.6, 1.3],\n",
       "        [5.2, 2.7, 3.9, 1.4],\n",
       "        [5. , 2. , 3.5, 1. ],\n",
       "        [5.9, 3. , 4.2, 1.5],\n",
       "        [6. , 2.2, 4. , 1. ],\n",
       "        [6.1, 2.9, 4.7, 1.4],\n",
       "        [5.6, 2.9, 3.6, 1.3],\n",
       "        [6.7, 3.1, 4.4, 1.4],\n",
       "        [5.6, 3. , 4.5, 1.5],\n",
       "        [5.8, 2.7, 4.1, 1. ],\n",
       "        [6.2, 2.2, 4.5, 1.5],\n",
       "        [5.6, 2.5, 3.9, 1.1],\n",
       "        [5.9, 3.2, 4.8, 1.8],\n",
       "        [6.1, 2.8, 4. , 1.3],\n",
       "        [6.3, 2.5, 4.9, 1.5],\n",
       "        [6.1, 2.8, 4.7, 1.2],\n",
       "        [6.4, 2.9, 4.3, 1.3],\n",
       "        [6.6, 3. , 4.4, 1.4],\n",
       "        [6.8, 2.8, 4.8, 1.4],\n",
       "        [6.7, 3. , 5. , 1.7],\n",
       "        [6. , 2.9, 4.5, 1.5],\n",
       "        [5.7, 2.6, 3.5, 1. ],\n",
       "        [5.5, 2.4, 3.8, 1.1],\n",
       "        [5.5, 2.4, 3.7, 1. ],\n",
       "        [5.8, 2.7, 3.9, 1.2],\n",
       "        [6. , 2.7, 5.1, 1.6],\n",
       "        [5.4, 3. , 4.5, 1.5],\n",
       "        [6. , 3.4, 4.5, 1.6],\n",
       "        [6.7, 3.1, 4.7, 1.5],\n",
       "        [6.3, 2.3, 4.4, 1.3],\n",
       "        [5.6, 3. , 4.1, 1.3],\n",
       "        [5.5, 2.5, 4. , 1.3],\n",
       "        [5.5, 2.6, 4.4, 1.2],\n",
       "        [6.1, 3. , 4.6, 1.4],\n",
       "        [5.8, 2.6, 4. , 1.2],\n",
       "        [5. , 2.3, 3.3, 1. ],\n",
       "        [5.6, 2.7, 4.2, 1.3],\n",
       "        [5.7, 3. , 4.2, 1.2],\n",
       "        [5.7, 2.9, 4.2, 1.3],\n",
       "        [6.2, 2.9, 4.3, 1.3],\n",
       "        [5.1, 2.5, 3. , 1.1],\n",
       "        [5.7, 2.8, 4.1, 1.3],\n",
       "        [6.3, 3.3, 6. , 2.5],\n",
       "        [5.8, 2.7, 5.1, 1.9],\n",
       "        [7.1, 3. , 5.9, 2.1],\n",
       "        [6.3, 2.9, 5.6, 1.8],\n",
       "        [6.5, 3. , 5.8, 2.2],\n",
       "        [7.6, 3. , 6.6, 2.1],\n",
       "        [4.9, 2.5, 4.5, 1.7],\n",
       "        [7.3, 2.9, 6.3, 1.8],\n",
       "        [6.7, 2.5, 5.8, 1.8],\n",
       "        [7.2, 3.6, 6.1, 2.5],\n",
       "        [6.5, 3.2, 5.1, 2. ],\n",
       "        [6.4, 2.7, 5.3, 1.9],\n",
       "        [6.8, 3. , 5.5, 2.1],\n",
       "        [5.7, 2.5, 5. , 2. ],\n",
       "        [5.8, 2.8, 5.1, 2.4],\n",
       "        [6.4, 3.2, 5.3, 2.3],\n",
       "        [6.5, 3. , 5.5, 1.8],\n",
       "        [7.7, 3.8, 6.7, 2.2],\n",
       "        [7.7, 2.6, 6.9, 2.3],\n",
       "        [6. , 2.2, 5. , 1.5],\n",
       "        [6.9, 3.2, 5.7, 2.3],\n",
       "        [5.6, 2.8, 4.9, 2. ],\n",
       "        [7.7, 2.8, 6.7, 2. ],\n",
       "        [6.3, 2.7, 4.9, 1.8],\n",
       "        [6.7, 3.3, 5.7, 2.1],\n",
       "        [7.2, 3.2, 6. , 1.8],\n",
       "        [6.2, 2.8, 4.8, 1.8],\n",
       "        [6.1, 3. , 4.9, 1.8],\n",
       "        [6.4, 2.8, 5.6, 2.1],\n",
       "        [7.2, 3. , 5.8, 1.6],\n",
       "        [7.4, 2.8, 6.1, 1.9],\n",
       "        [7.9, 3.8, 6.4, 2. ],\n",
       "        [6.4, 2.8, 5.6, 2.2],\n",
       "        [6.3, 2.8, 5.1, 1.5],\n",
       "        [6.1, 2.6, 5.6, 1.4],\n",
       "        [7.7, 3. , 6.1, 2.3],\n",
       "        [6.3, 3.4, 5.6, 2.4],\n",
       "        [6.4, 3.1, 5.5, 1.8],\n",
       "        [6. , 3. , 4.8, 1.8],\n",
       "        [6.9, 3.1, 5.4, 2.1],\n",
       "        [6.7, 3.1, 5.6, 2.4],\n",
       "        [6.9, 3.1, 5.1, 2.3],\n",
       "        [5.8, 2.7, 5.1, 1.9],\n",
       "        [6.8, 3.2, 5.9, 2.3],\n",
       "        [6.7, 3.3, 5.7, 2.5],\n",
       "        [6.7, 3. , 5.2, 2.3],\n",
       "        [6.3, 2.5, 5. , 1.9],\n",
       "        [6.5, 3. , 5.2, 2. ],\n",
       "        [6.2, 3.4, 5.4, 2.3],\n",
       "        [5.9, 3. , 5.1, 1.8]]),\n",
       " 'target': array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "        0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "        0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "        1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "        1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
       "        2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
       "        2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2]),\n",
       " 'frame': None,\n",
       " 'target_names': array(['setosa', 'versicolor', 'virginica'], dtype='<U10'),\n",
       " 'DESCR': '.. _iris_dataset:\\n\\nIris plants dataset\\n--------------------\\n\\n**Data Set Characteristics:**\\n\\n    :Number of Instances: 150 (50 in each of three classes)\\n    :Number of Attributes: 4 numeric, predictive attributes and the class\\n    :Attribute Information:\\n        - sepal length in cm\\n        - sepal width in cm\\n        - petal length in cm\\n        - petal width in cm\\n        - class:\\n                - Iris-Setosa\\n                - Iris-Versicolour\\n                - Iris-Virginica\\n                \\n    :Summary Statistics:\\n\\n    ============== ==== ==== ======= ===== ====================\\n                    Min  Max   Mean    SD   Class Correlation\\n    ============== ==== ==== ======= ===== ====================\\n    sepal length:   4.3  7.9   5.84   0.83    0.7826\\n    sepal width:    2.0  4.4   3.05   0.43   -0.4194\\n    petal length:   1.0  6.9   3.76   1.76    0.9490  (high!)\\n    petal width:    0.1  2.5   1.20   0.76    0.9565  (high!)\\n    ============== ==== ==== ======= ===== ====================\\n\\n    :Missing Attribute Values: None\\n    :Class Distribution: 33.3% for each of 3 classes.\\n    :Creator: R.A. Fisher\\n    :Donor: Michael Marshall (MARSHALL%PLU@io.arc.nasa.gov)\\n    :Date: July, 1988\\n\\nThe famous Iris database, first used by Sir R.A. Fisher. The dataset is taken\\nfrom Fisher\\'s paper. Note that it\\'s the same as in R, but not as in the UCI\\nMachine Learning Repository, which has two wrong data points.\\n\\nThis is perhaps the best known database to be found in the\\npattern recognition literature.  Fisher\\'s paper is a classic in the field and\\nis referenced frequently to this day.  (See Duda & Hart, for example.)  The\\ndata set contains 3 classes of 50 instances each, where each class refers to a\\ntype of iris plant.  One class is linearly separable from the other 2; the\\nlatter are NOT linearly separable from each other.\\n\\n.. topic:: References\\n\\n   - Fisher, R.A. \"The use of multiple measurements in taxonomic problems\"\\n     Annual Eugenics, 7, Part II, 179-188 (1936); also in \"Contributions to\\n     Mathematical Statistics\" (John Wiley, NY, 1950).\\n   - Duda, R.O., & Hart, P.E. (1973) Pattern Classification and Scene Analysis.\\n     (Q327.D83) John Wiley & Sons.  ISBN 0-471-22361-1.  See page 218.\\n   - Dasarathy, B.V. (1980) \"Nosing Around the Neighborhood: A New System\\n     Structure and Classification Rule for Recognition in Partially Exposed\\n     Environments\".  IEEE Transactions on Pattern Analysis and Machine\\n     Intelligence, Vol. PAMI-2, No. 1, 67-71.\\n   - Gates, G.W. (1972) \"The Reduced Nearest Neighbor Rule\".  IEEE Transactions\\n     on Information Theory, May 1972, 431-433.\\n   - See also: 1988 MLC Proceedings, 54-64.  Cheeseman et al\"s AUTOCLASS II\\n     conceptual clustering system finds 3 classes in the data.\\n   - Many, many more ...',\n",
       " 'feature_names': ['sepal length (cm)',\n",
       "  'sepal width (cm)',\n",
       "  'petal length (cm)',\n",
       "  'petal width (cm)'],\n",
       " 'filename': 'C:\\\\ProgramData\\\\Anaconda3\\\\lib\\\\site-packages\\\\sklearn\\\\datasets\\\\data\\\\iris.csv'}"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.datasets import load_iris\n",
    "data =load_iris()\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['data', 'target', 'frame', 'target_names', 'DESCR', 'feature_names', 'filename'])"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['sepal length (cm)',\n",
       " 'sepal width (cm)',\n",
       " 'petal length (cm)',\n",
       " 'petal width (cm)']"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.feature_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ".. _iris_dataset:\n",
      "\n",
      "Iris plants dataset\n",
      "--------------------\n",
      "\n",
      "**Data Set Characteristics:**\n",
      "\n",
      "    :Number of Instances: 150 (50 in each of three classes)\n",
      "    :Number of Attributes: 4 numeric, predictive attributes and the class\n",
      "    :Attribute Information:\n",
      "        - sepal length in cm\n",
      "        - sepal width in cm\n",
      "        - petal length in cm\n",
      "        - petal width in cm\n",
      "        - class:\n",
      "                - Iris-Setosa\n",
      "                - Iris-Versicolour\n",
      "                - Iris-Virginica\n",
      "                \n",
      "    :Summary Statistics:\n",
      "\n",
      "    ============== ==== ==== ======= ===== ====================\n",
      "                    Min  Max   Mean    SD   Class Correlation\n",
      "    ============== ==== ==== ======= ===== ====================\n",
      "    sepal length:   4.3  7.9   5.84   0.83    0.7826\n",
      "    sepal width:    2.0  4.4   3.05   0.43   -0.4194\n",
      "    petal length:   1.0  6.9   3.76   1.76    0.9490  (high!)\n",
      "    petal width:    0.1  2.5   1.20   0.76    0.9565  (high!)\n",
      "    ============== ==== ==== ======= ===== ====================\n",
      "\n",
      "    :Missing Attribute Values: None\n",
      "    :Class Distribution: 33.3% for each of 3 classes.\n",
      "    :Creator: R.A. Fisher\n",
      "    :Donor: Michael Marshall (MARSHALL%PLU@io.arc.nasa.gov)\n",
      "    :Date: July, 1988\n",
      "\n",
      "The famous Iris database, first used by Sir R.A. Fisher. The dataset is taken\n",
      "from Fisher's paper. Note that it's the same as in R, but not as in the UCI\n",
      "Machine Learning Repository, which has two wrong data points.\n",
      "\n",
      "This is perhaps the best known database to be found in the\n",
      "pattern recognition literature.  Fisher's paper is a classic in the field and\n",
      "is referenced frequently to this day.  (See Duda & Hart, for example.)  The\n",
      "data set contains 3 classes of 50 instances each, where each class refers to a\n",
      "type of iris plant.  One class is linearly separable from the other 2; the\n",
      "latter are NOT linearly separable from each other.\n",
      "\n",
      ".. topic:: References\n",
      "\n",
      "   - Fisher, R.A. \"The use of multiple measurements in taxonomic problems\"\n",
      "     Annual Eugenics, 7, Part II, 179-188 (1936); also in \"Contributions to\n",
      "     Mathematical Statistics\" (John Wiley, NY, 1950).\n",
      "   - Duda, R.O., & Hart, P.E. (1973) Pattern Classification and Scene Analysis.\n",
      "     (Q327.D83) John Wiley & Sons.  ISBN 0-471-22361-1.  See page 218.\n",
      "   - Dasarathy, B.V. (1980) \"Nosing Around the Neighborhood: A New System\n",
      "     Structure and Classification Rule for Recognition in Partially Exposed\n",
      "     Environments\".  IEEE Transactions on Pattern Analysis and Machine\n",
      "     Intelligence, Vol. PAMI-2, No. 1, 67-71.\n",
      "   - Gates, G.W. (1972) \"The Reduced Nearest Neighbor Rule\".  IEEE Transactions\n",
      "     on Information Theory, May 1972, 431-433.\n",
      "   - See also: 1988 MLC Proceedings, 54-64.  Cheeseman et al\"s AUTOCLASS II\n",
      "     conceptual clustering system finds 3 classes in the data.\n",
      "   - Many, many more ...\n"
     ]
    }
   ],
   "source": [
    "print(data.DESCR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[5.1, 3.5, 1.4, 0.2],\n",
       "       [4.9, 3. , 1.4, 0.2],\n",
       "       [4.7, 3.2, 1.3, 0.2],\n",
       "       [4.6, 3.1, 1.5, 0.2],\n",
       "       [5. , 3.6, 1.4, 0.2],\n",
       "       [5.4, 3.9, 1.7, 0.4],\n",
       "       [4.6, 3.4, 1.4, 0.3],\n",
       "       [5. , 3.4, 1.5, 0.2],\n",
       "       [4.4, 2.9, 1.4, 0.2],\n",
       "       [4.9, 3.1, 1.5, 0.1],\n",
       "       [5.4, 3.7, 1.5, 0.2],\n",
       "       [4.8, 3.4, 1.6, 0.2],\n",
       "       [4.8, 3. , 1.4, 0.1],\n",
       "       [4.3, 3. , 1.1, 0.1],\n",
       "       [5.8, 4. , 1.2, 0.2],\n",
       "       [5.7, 4.4, 1.5, 0.4],\n",
       "       [5.4, 3.9, 1.3, 0.4],\n",
       "       [5.1, 3.5, 1.4, 0.3],\n",
       "       [5.7, 3.8, 1.7, 0.3],\n",
       "       [5.1, 3.8, 1.5, 0.3],\n",
       "       [5.4, 3.4, 1.7, 0.2],\n",
       "       [5.1, 3.7, 1.5, 0.4],\n",
       "       [4.6, 3.6, 1. , 0.2],\n",
       "       [5.1, 3.3, 1.7, 0.5],\n",
       "       [4.8, 3.4, 1.9, 0.2],\n",
       "       [5. , 3. , 1.6, 0.2],\n",
       "       [5. , 3.4, 1.6, 0.4],\n",
       "       [5.2, 3.5, 1.5, 0.2],\n",
       "       [5.2, 3.4, 1.4, 0.2],\n",
       "       [4.7, 3.2, 1.6, 0.2],\n",
       "       [4.8, 3.1, 1.6, 0.2],\n",
       "       [5.4, 3.4, 1.5, 0.4],\n",
       "       [5.2, 4.1, 1.5, 0.1],\n",
       "       [5.5, 4.2, 1.4, 0.2],\n",
       "       [4.9, 3.1, 1.5, 0.2],\n",
       "       [5. , 3.2, 1.2, 0.2],\n",
       "       [5.5, 3.5, 1.3, 0.2],\n",
       "       [4.9, 3.6, 1.4, 0.1],\n",
       "       [4.4, 3. , 1.3, 0.2],\n",
       "       [5.1, 3.4, 1.5, 0.2],\n",
       "       [5. , 3.5, 1.3, 0.3],\n",
       "       [4.5, 2.3, 1.3, 0.3],\n",
       "       [4.4, 3.2, 1.3, 0.2],\n",
       "       [5. , 3.5, 1.6, 0.6],\n",
       "       [5.1, 3.8, 1.9, 0.4],\n",
       "       [4.8, 3. , 1.4, 0.3],\n",
       "       [5.1, 3.8, 1.6, 0.2],\n",
       "       [4.6, 3.2, 1.4, 0.2],\n",
       "       [5.3, 3.7, 1.5, 0.2],\n",
       "       [5. , 3.3, 1.4, 0.2],\n",
       "       [7. , 3.2, 4.7, 1.4],\n",
       "       [6.4, 3.2, 4.5, 1.5],\n",
       "       [6.9, 3.1, 4.9, 1.5],\n",
       "       [5.5, 2.3, 4. , 1.3],\n",
       "       [6.5, 2.8, 4.6, 1.5],\n",
       "       [5.7, 2.8, 4.5, 1.3],\n",
       "       [6.3, 3.3, 4.7, 1.6],\n",
       "       [4.9, 2.4, 3.3, 1. ],\n",
       "       [6.6, 2.9, 4.6, 1.3],\n",
       "       [5.2, 2.7, 3.9, 1.4],\n",
       "       [5. , 2. , 3.5, 1. ],\n",
       "       [5.9, 3. , 4.2, 1.5],\n",
       "       [6. , 2.2, 4. , 1. ],\n",
       "       [6.1, 2.9, 4.7, 1.4],\n",
       "       [5.6, 2.9, 3.6, 1.3],\n",
       "       [6.7, 3.1, 4.4, 1.4],\n",
       "       [5.6, 3. , 4.5, 1.5],\n",
       "       [5.8, 2.7, 4.1, 1. ],\n",
       "       [6.2, 2.2, 4.5, 1.5],\n",
       "       [5.6, 2.5, 3.9, 1.1],\n",
       "       [5.9, 3.2, 4.8, 1.8],\n",
       "       [6.1, 2.8, 4. , 1.3],\n",
       "       [6.3, 2.5, 4.9, 1.5],\n",
       "       [6.1, 2.8, 4.7, 1.2],\n",
       "       [6.4, 2.9, 4.3, 1.3],\n",
       "       [6.6, 3. , 4.4, 1.4],\n",
       "       [6.8, 2.8, 4.8, 1.4],\n",
       "       [6.7, 3. , 5. , 1.7],\n",
       "       [6. , 2.9, 4.5, 1.5],\n",
       "       [5.7, 2.6, 3.5, 1. ],\n",
       "       [5.5, 2.4, 3.8, 1.1],\n",
       "       [5.5, 2.4, 3.7, 1. ],\n",
       "       [5.8, 2.7, 3.9, 1.2],\n",
       "       [6. , 2.7, 5.1, 1.6],\n",
       "       [5.4, 3. , 4.5, 1.5],\n",
       "       [6. , 3.4, 4.5, 1.6],\n",
       "       [6.7, 3.1, 4.7, 1.5],\n",
       "       [6.3, 2.3, 4.4, 1.3],\n",
       "       [5.6, 3. , 4.1, 1.3],\n",
       "       [5.5, 2.5, 4. , 1.3],\n",
       "       [5.5, 2.6, 4.4, 1.2],\n",
       "       [6.1, 3. , 4.6, 1.4],\n",
       "       [5.8, 2.6, 4. , 1.2],\n",
       "       [5. , 2.3, 3.3, 1. ],\n",
       "       [5.6, 2.7, 4.2, 1.3],\n",
       "       [5.7, 3. , 4.2, 1.2],\n",
       "       [5.7, 2.9, 4.2, 1.3],\n",
       "       [6.2, 2.9, 4.3, 1.3],\n",
       "       [5.1, 2.5, 3. , 1.1],\n",
       "       [5.7, 2.8, 4.1, 1.3],\n",
       "       [6.3, 3.3, 6. , 2.5],\n",
       "       [5.8, 2.7, 5.1, 1.9],\n",
       "       [7.1, 3. , 5.9, 2.1],\n",
       "       [6.3, 2.9, 5.6, 1.8],\n",
       "       [6.5, 3. , 5.8, 2.2],\n",
       "       [7.6, 3. , 6.6, 2.1],\n",
       "       [4.9, 2.5, 4.5, 1.7],\n",
       "       [7.3, 2.9, 6.3, 1.8],\n",
       "       [6.7, 2.5, 5.8, 1.8],\n",
       "       [7.2, 3.6, 6.1, 2.5],\n",
       "       [6.5, 3.2, 5.1, 2. ],\n",
       "       [6.4, 2.7, 5.3, 1.9],\n",
       "       [6.8, 3. , 5.5, 2.1],\n",
       "       [5.7, 2.5, 5. , 2. ],\n",
       "       [5.8, 2.8, 5.1, 2.4],\n",
       "       [6.4, 3.2, 5.3, 2.3],\n",
       "       [6.5, 3. , 5.5, 1.8],\n",
       "       [7.7, 3.8, 6.7, 2.2],\n",
       "       [7.7, 2.6, 6.9, 2.3],\n",
       "       [6. , 2.2, 5. , 1.5],\n",
       "       [6.9, 3.2, 5.7, 2.3],\n",
       "       [5.6, 2.8, 4.9, 2. ],\n",
       "       [7.7, 2.8, 6.7, 2. ],\n",
       "       [6.3, 2.7, 4.9, 1.8],\n",
       "       [6.7, 3.3, 5.7, 2.1],\n",
       "       [7.2, 3.2, 6. , 1.8],\n",
       "       [6.2, 2.8, 4.8, 1.8],\n",
       "       [6.1, 3. , 4.9, 1.8],\n",
       "       [6.4, 2.8, 5.6, 2.1],\n",
       "       [7.2, 3. , 5.8, 1.6],\n",
       "       [7.4, 2.8, 6.1, 1.9],\n",
       "       [7.9, 3.8, 6.4, 2. ],\n",
       "       [6.4, 2.8, 5.6, 2.2],\n",
       "       [6.3, 2.8, 5.1, 1.5],\n",
       "       [6.1, 2.6, 5.6, 1.4],\n",
       "       [7.7, 3. , 6.1, 2.3],\n",
       "       [6.3, 3.4, 5.6, 2.4],\n",
       "       [6.4, 3.1, 5.5, 1.8],\n",
       "       [6. , 3. , 4.8, 1.8],\n",
       "       [6.9, 3.1, 5.4, 2.1],\n",
       "       [6.7, 3.1, 5.6, 2.4],\n",
       "       [6.9, 3.1, 5.1, 2.3],\n",
       "       [5.8, 2.7, 5.1, 1.9],\n",
       "       [6.8, 3.2, 5.9, 2.3],\n",
       "       [6.7, 3.3, 5.7, 2.5],\n",
       "       [6.7, 3. , 5.2, 2.3],\n",
       "       [6.3, 2.5, 5. , 1.9],\n",
       "       [6.5, 3. , 5.2, 2. ],\n",
       "       [6.2, 3.4, 5.4, 2.3],\n",
       "       [5.9, 3. , 5.1, 1.8]])"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.data # data['data']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2])"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>6.7</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>6.3</td>\n",
       "      <td>2.5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147</th>\n",
       "      <td>6.5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>6.2</td>\n",
       "      <td>3.4</td>\n",
       "      <td>5.4</td>\n",
       "      <td>2.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>5.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.1</td>\n",
       "      <td>1.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>150 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)\n",
       "0                  5.1               3.5                1.4               0.2\n",
       "1                  4.9               3.0                1.4               0.2\n",
       "2                  4.7               3.2                1.3               0.2\n",
       "3                  4.6               3.1                1.5               0.2\n",
       "4                  5.0               3.6                1.4               0.2\n",
       "..                 ...               ...                ...               ...\n",
       "145                6.7               3.0                5.2               2.3\n",
       "146                6.3               2.5                5.0               1.9\n",
       "147                6.5               3.0                5.2               2.0\n",
       "148                6.2               3.4                5.4               2.3\n",
       "149                5.9               3.0                5.1               1.8\n",
       "\n",
       "[150 rows x 4 columns]"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(data.data, columns = data.feature_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.1</td>\n",
       "      <td>1.5</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>6.7</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>6.3</td>\n",
       "      <td>2.5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>147</th>\n",
       "      <td>6.5</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>6.2</td>\n",
       "      <td>3.4</td>\n",
       "      <td>5.4</td>\n",
       "      <td>2.3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149</th>\n",
       "      <td>5.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.1</td>\n",
       "      <td>1.8</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>150 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  \\\n",
       "0                  5.1               3.5                1.4               0.2   \n",
       "1                  4.9               3.0                1.4               0.2   \n",
       "2                  4.7               3.2                1.3               0.2   \n",
       "3                  4.6               3.1                1.5               0.2   \n",
       "4                  5.0               3.6                1.4               0.2   \n",
       "..                 ...               ...                ...               ...   \n",
       "145                6.7               3.0                5.2               2.3   \n",
       "146                6.3               2.5                5.0               1.9   \n",
       "147                6.5               3.0                5.2               2.0   \n",
       "148                6.2               3.4                5.4               2.3   \n",
       "149                5.9               3.0                5.1               1.8   \n",
       "\n",
       "     target  \n",
       "0         0  \n",
       "1         0  \n",
       "2         0  \n",
       "3         0  \n",
       "4         0  \n",
       "..      ...  \n",
       "145       2  \n",
       "146       2  \n",
       "147       2  \n",
       "148       2  \n",
       "149       2  \n",
       "\n",
       "[150 rows x 5 columns]"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.DataFrame(data.data, columns = data.feature_names)\n",
    "df['target'] = data.target\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "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>sepal length (cm)</th>\n",
       "      <th>sepal width (cm)</th>\n",
       "      <th>petal length (cm)</th>\n",
       "      <th>petal width (cm)</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>5.1</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.7</td>\n",
       "      <td>3.2</td>\n",
       "      <td>1.3</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.0</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>4.6</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>4.4</td>\n",
       "      <td>2.9</td>\n",
       "      <td>1.4</td>\n",
       "      <td>0.2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>6.7</td>\n",
       "      <td>3.1</td>\n",
       "      <td>5.6</td>\n",
       "      <td>2.4</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>142</th>\n",
       "      <td>5.8</td>\n",
       "      <td>2.7</td>\n",
       "      <td>5.1</td>\n",
       "      <td>1.9</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>144</th>\n",
       "      <td>6.7</td>\n",
       "      <td>3.3</td>\n",
       "      <td>5.7</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>6.3</td>\n",
       "      <td>2.5</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>6.2</td>\n",
       "      <td>3.4</td>\n",
       "      <td>5.4</td>\n",
       "      <td>2.3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>75 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  \\\n",
       "0                  5.1               3.5                1.4               0.2   \n",
       "2                  4.7               3.2                1.3               0.2   \n",
       "4                  5.0               3.6                1.4               0.2   \n",
       "6                  4.6               3.4                1.4               0.3   \n",
       "8                  4.4               2.9                1.4               0.2   \n",
       "..                 ...               ...                ...               ...   \n",
       "140                6.7               3.1                5.6               2.4   \n",
       "142                5.8               2.7                5.1               1.9   \n",
       "144                6.7               3.3                5.7               2.5   \n",
       "146                6.3               2.5                5.0               1.9   \n",
       "148                6.2               3.4                5.4               2.3   \n",
       "\n",
       "     target  \n",
       "0         0  \n",
       "2         0  \n",
       "4         0  \n",
       "6         0  \n",
       "8         0  \n",
       "..      ...  \n",
       "140       2  \n",
       "142       2  \n",
       "144       2  \n",
       "146       2  \n",
       "148       2  \n",
       "\n",
       "[75 rows x 5 columns]"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[::2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    25\n",
       "1    25\n",
       "2    25\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train = df.drop('target', axis=1)[::2]\n",
    "y_train = df.target[::2]\n",
    "X_train.shape, y_train.shape\n",
    "y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('feature_selector',\n",
       "                 RFE(estimator=LogisticRegression(), n_features_to_select=5,\n",
       "                     step=20)),\n",
       "                ('classifier', LogisticRegression())])"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,\n",
       "       2, 2, 2, 2, 2, 2, 2, 2, 2])"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test = df.drop('target', axis=1)[1::2]\n",
    "y_test = df.target[1::2]\n",
    "clf.predict(X_test) \n",
    "y_test.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.96"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(X_test) == y_train).count()\n",
    "(clf.predict(X_test) == y_train).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "82     False\n",
       "118    False\n",
       "132    False\n",
       "Name: target, dtype: bool"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(clf.predict(X_test) == y_train).loc[lambda x: x==False]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### train_test_split 활용"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.drop('target', axis=1)\n",
    "y = df.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:763: 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",
      "  n_iter_i = _check_optimize_result(\n",
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\sklearn\\linear_model\\_logistic.py:763: 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",
      "  n_iter_i = _check_optimize_result(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clf.fit(X_train, y_train)\n",
    "clf.predict(X_test) == y_test.values\n",
    "(clf.predict(X_test) == y_test.values).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 실습: iris"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.datasets import load_iris\n",
    "\n",
    "# load the iris dataset and split it into train and test sets\n",
    "X, y = load_iris(return_X_y=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### train_test_split 연습"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    0.421053\n",
       "0    0.342105\n",
       "2    0.236842\n",
       "dtype: float64"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)\n",
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape\n",
    "\n",
    "pd.Series(y_train).value_counts(normalize=True)\n",
    "pd.Series(y_test).value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    0.342105\n",
       "1    0.342105\n",
       "2    0.315789\n",
       "dtype: float64"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, stratify=y, random_state=0)\n",
    "pd.Series(y_train).value_counts(normalize=True)\n",
    "pd.Series(y_test).value_counts(normalize=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### pipeline 연습 \n",
    "- make_pipeline과 Pipeline은 동일한 역할\n",
    "- 사용하는 형식이 다름 --> Pipeline은 해당 단계의 이름을 부여"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('standardscaler', StandardScaler()),\n",
       "                ('logisticregression', LogisticRegression())])"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "# create a pipeline object\n",
    "pipe = make_pipeline(\n",
    "    StandardScaler(),\n",
    "    LogisticRegression()\n",
    ")\n",
    "\n",
    "# fit the whole pipeline\n",
    "pipe.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 1, 1, 1, 0, 1, 2, 2, 2, 1, 2, 1, 0, 0, 2, 0, 1, 2, 1,\n",
       "       1, 0, 2, 0, 0, 1, 2, 1, 0, 1, 2, 2, 0, 1, 2, 2])"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pipe.predict(X_test)\n",
    "y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9736842105263158"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# we can now use it like any other estimator\n",
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(pipe.predict(X_test), y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 실습: cross-validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "model = RandomForestClassifier(random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(data.data, columns = data.feature_names)\n",
    "df['target'] = data.target\n",
    "df.head()\n",
    "\n",
    "X = df.drop('target', axis=1)\n",
    "y = df.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9533333333333334"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "kf = KFold(n_splits=5, shuffle=True, random_state=11)\n",
    "\n",
    "accuracy_list = []\n",
    "for train_index, test_index in kf.split(X):\n",
    "    train_index[0:10], test_index[0:10]\n",
    "    X_train = X.loc[train_index]; y_train = y.loc[train_index]\n",
    "    X_test = X.loc[test_index]; y_test = y.loc[test_index]    \n",
    "    y_predict = model.fit(X_train, y_train).predict(X_test)\n",
    "    accuracy_list.append(accuracy_score(y_test, y_predict))\n",
    "    \n",
    "accuracy_list\n",
    "sum(accuracy_list)/len(accuracy_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'fit_time': array([0.34662986, 0.32533121, 0.32334852, 0.32440472, 0.30594993]),\n",
       " 'score_time': array([0.02840853, 0.0286355 , 0.02474475, 0.02772927, 0.02455354]),\n",
       " 'test_score': array([0.93333333, 0.9       , 0.96666667, 1.        , 0.96666667])}"
      ]
     },
     "execution_count": 138,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result = cross_validate(model, X, y, cv=kf)  # defaults to 5-fold CV\n",
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 139,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9533333333333334"
      ]
     },
     "execution_count": 139,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result['test_score'] # 이건 안됨, result.'test_score'\n",
    "np.mean(result['test_score'] )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 실습 size_korea"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((6401, 2), (6401,))"
      ]
     },
     "execution_count": 140,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = size_korea[['키','몸무게']].values\n",
    "y = size_korea.체지방율\n",
    "X.shape, y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.59514149, 0.54009813, 0.4276149 , 0.42368942, 0.01218939])"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.model_selection import cross_validate\n",
    "\n",
    "lr = LinearRegression()\n",
    "result = cross_validate(lr, X, y)  \n",
    "result['test_score']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 실습 make_regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1000,)"
      ]
     },
     "execution_count": 142,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# https://scikit-learn.org/stable/datasets/sample_generators.html\n",
    "from sklearn.datasets import make_regression\n",
    "X, y = make_regression(n_samples=1000, random_state=0)\n",
    "X.shape\n",
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1., 1., 1., 1., 1.])"
      ]
     },
     "execution_count": 143,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.model_selection import cross_validate\n",
    "\n",
    "lr = LinearRegression()\n",
    "result = cross_validate(lr, X, y)  # defaults to 5-fold CV\n",
    "result['test_score']  # r_squared score is high because dataset is easy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 실습 california_housing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.datasets import fetch_california_housing\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from sklearn.model_selection import train_test_split\n",
    "from scipy.stats import randint"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "https://scikit-learn.org/stable/datasets/real_world.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ".. _california_housing_dataset:\n",
      "\n",
      "California Housing dataset\n",
      "--------------------------\n",
      "\n",
      "**Data Set Characteristics:**\n",
      "\n",
      "    :Number of Instances: 20640\n",
      "\n",
      "    :Number of Attributes: 8 numeric, predictive attributes and the target\n",
      "\n",
      "    :Attribute Information:\n",
      "        - MedInc        median income in block\n",
      "        - HouseAge      median house age in block\n",
      "        - AveRooms      average number of rooms\n",
      "        - AveBedrms     average number of bedrooms\n",
      "        - Population    block population\n",
      "        - AveOccup      average house occupancy\n",
      "        - Latitude      house block latitude\n",
      "        - Longitude     house block longitude\n",
      "\n",
      "    :Missing Attribute Values: None\n",
      "\n",
      "This dataset was obtained from the StatLib repository.\n",
      "http://lib.stat.cmu.edu/datasets/\n",
      "\n",
      "The target variable is the median house value for California districts.\n",
      "\n",
      "This dataset was derived from the 1990 U.S. census, using one row per census\n",
      "block group. A block group is the smallest geographical unit for which the U.S.\n",
      "Census Bureau publishes sample data (a block group typically has a population\n",
      "of 600 to 3,000 people).\n",
      "\n",
      "It can be downloaded/loaded using the\n",
      ":func:`sklearn.datasets.fetch_california_housing` function.\n",
      "\n",
      ".. topic:: References\n",
      "\n",
      "    - Pace, R. Kelley and Ronald Barry, Sparse Spatial Autoregressions,\n",
      "      Statistics and Probability Letters, 33 (1997) 291-297\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(fetch_california_housing().DESCR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, y = fetch_california_housing(return_X_y=True)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.735363411343253"
      ]
     },
     "execution_count": 147,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# define the parameter space that will be searched over\n",
    "param_distributions = {'n_estimators': randint(1, 5),\n",
    "                       'max_depth': randint(5, 10)}\n",
    "\n",
    "# now create a searchCV object and fit it to the data\n",
    "search = RandomizedSearchCV(estimator=RandomForestRegressor(random_state=0),\n",
    "                            n_iter=5,\n",
    "                            param_distributions=param_distributions,\n",
    "                            random_state=0)\n",
    "search.fit(X_train, y_train)\n",
    "\n",
    "search.best_params_\n",
    "\n",
    "# the search object now acts like a normal random forest estimator\n",
    "# with max_depth=9 and n_estimators=4\n",
    "search.score(X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1.68495047, 2.64131467, 1.62184929, ..., 1.99436275, 2.31629552,\n",
       "       2.44966756])"
      ]
     },
     "execution_count": 148,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "search.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# End"
   ]
  },
  {
   "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": 2
}