{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**2장 – 머신러닝 프로젝트 처음부터 끝까지**\n",
    "\n",
    "*머신러닝 부동산 회사에 오신 것을 환영합니다! 여러분이 할 작업은 캘리포니아 지역 주택의 여러 특성을 사용해 중간 가격을 예측하는 것입니다.*\n",
    "\n",
    "*이 노트북은 2장의 모든 샘플 코드와 연습 문제 정답을 담고 있습니다.*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<table align=\"left\">\n",
    "  <td>\n",
    "    <a target=\"_blank\" href=\"https://colab.research.google.com/github/rickiepark/handson-ml2/blob/master/02_end_to_end_machine_learning_project.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />구글 코랩에서 실행하기</a>\n",
    "  </td>\n",
    "</table>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 설정"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "먼저 몇 개의 모듈을 임포트합니다. 맷플롯립 그래프를 인라인으로 출력하도록 만들고 그림을 저장하는 함수를 준비합니다. 또한 파이썬 버전이 3.5 이상인지 확인합니다(파이썬 2.x에서도 동작하지만 곧 지원이 중단되므로 파이썬 3을 사용하는 것이 좋습니다). 사이킷런 버전이 0.20 이상인지도 확인합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:11.219363Z",
     "iopub.status.busy": "2021-11-03T03:33:11.218691Z",
     "iopub.status.idle": "2021-11-03T03:33:11.942839Z",
     "shell.execute_reply": "2021-11-03T03:33:11.941785Z"
    }
   },
   "outputs": [],
   "source": [
    "# 파이썬 ≥3.5 필수\n",
    "import sys\n",
    "assert sys.version_info >= (3, 5)\n",
    "\n",
    "# 사이킷런 ≥0.20 필수\n",
    "import sklearn\n",
    "assert sklearn.__version__ >= \"0.20\"\n",
    "\n",
    "# 공통 모듈 임포트\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "# 깔금한 그래프 출력을 위해\n",
    "%matplotlib inline\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "mpl.rc('axes', labelsize=14)\n",
    "mpl.rc('xtick', labelsize=12)\n",
    "mpl.rc('ytick', labelsize=12)\n",
    "\n",
    "# 그림을 저장할 위치\n",
    "PROJECT_ROOT_DIR = \".\"\n",
    "CHAPTER_ID = \"end_to_end_project\"\n",
    "IMAGES_PATH = os.path.join(PROJECT_ROOT_DIR, \"images\", CHAPTER_ID)\n",
    "os.makedirs(IMAGES_PATH, exist_ok=True)\n",
    "\n",
    "def save_fig(fig_id, tight_layout=True, fig_extension=\"png\", resolution=300):\n",
    "    path = os.path.join(IMAGES_PATH, fig_id + \".\" + fig_extension)\n",
    "    print(\"그림 저장:\", fig_id)\n",
    "    if tight_layout:\n",
    "        plt.tight_layout()\n",
    "    plt.savefig(path, format=fig_extension, dpi=resolution)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 데이터 가져오기"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 데이터 다운로드하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:11.949935Z",
     "iopub.status.busy": "2021-11-03T03:33:11.948263Z",
     "iopub.status.idle": "2021-11-03T03:33:11.952719Z",
     "shell.execute_reply": "2021-11-03T03:33:11.952179Z"
    }
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import tarfile\n",
    "import urllib.request\n",
    "\n",
    "DOWNLOAD_ROOT = \"https://raw.githubusercontent.com/rickiepark/handson-ml2/master/\"\n",
    "HOUSING_PATH = os.path.join(\"datasets\", \"housing\")\n",
    "HOUSING_URL = DOWNLOAD_ROOT + \"datasets/housing/housing.tgz\"\n",
    "\n",
    "def fetch_housing_data(housing_url=HOUSING_URL, housing_path=HOUSING_PATH):\n",
    "    if not os.path.isdir(housing_path):\n",
    "        os.makedirs(housing_path)\n",
    "    tgz_path = os.path.join(housing_path, \"housing.tgz\")\n",
    "    urllib.request.urlretrieve(housing_url, tgz_path)\n",
    "    housing_tgz = tarfile.open(tgz_path)\n",
    "    housing_tgz.extractall(path=housing_path)\n",
    "    housing_tgz.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:11.957509Z",
     "iopub.status.busy": "2021-11-03T03:33:11.956858Z",
     "iopub.status.idle": "2021-11-03T03:33:13.579164Z",
     "shell.execute_reply": "2021-11-03T03:33:13.580002Z"
    }
   },
   "outputs": [],
   "source": [
    "fetch_housing_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:13.583831Z",
     "iopub.status.busy": "2021-11-03T03:33:13.582631Z",
     "iopub.status.idle": "2021-11-03T03:33:14.164232Z",
     "shell.execute_reply": "2021-11-03T03:33:14.165056Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "def load_housing_data(housing_path=HOUSING_PATH):\n",
    "    csv_path = os.path.join(housing_path, \"housing.csv\")\n",
    "    return pd.read_csv(csv_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 데이터 구조 훑어 보기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:14.173583Z",
     "iopub.status.busy": "2021-11-03T03:33:14.172612Z",
     "iopub.status.idle": "2021-11-03T03:33:14.220780Z",
     "shell.execute_reply": "2021-11-03T03:33:14.220191Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-122.23</td>\n",
       "      <td>37.88</td>\n",
       "      <td>41.0</td>\n",
       "      <td>880.0</td>\n",
       "      <td>129.0</td>\n",
       "      <td>322.0</td>\n",
       "      <td>126.0</td>\n",
       "      <td>8.3252</td>\n",
       "      <td>452600.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-122.22</td>\n",
       "      <td>37.86</td>\n",
       "      <td>21.0</td>\n",
       "      <td>7099.0</td>\n",
       "      <td>1106.0</td>\n",
       "      <td>2401.0</td>\n",
       "      <td>1138.0</td>\n",
       "      <td>8.3014</td>\n",
       "      <td>358500.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-122.24</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1467.0</td>\n",
       "      <td>190.0</td>\n",
       "      <td>496.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>7.2574</td>\n",
       "      <td>352100.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-122.25</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1274.0</td>\n",
       "      <td>235.0</td>\n",
       "      <td>558.0</td>\n",
       "      <td>219.0</td>\n",
       "      <td>5.6431</td>\n",
       "      <td>341300.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-122.25</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1627.0</td>\n",
       "      <td>280.0</td>\n",
       "      <td>565.0</td>\n",
       "      <td>259.0</td>\n",
       "      <td>3.8462</td>\n",
       "      <td>342200.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "0    -122.23     37.88                41.0        880.0           129.0   \n",
       "1    -122.22     37.86                21.0       7099.0          1106.0   \n",
       "2    -122.24     37.85                52.0       1467.0           190.0   \n",
       "3    -122.25     37.85                52.0       1274.0           235.0   \n",
       "4    -122.25     37.85                52.0       1627.0           280.0   \n",
       "\n",
       "   population  households  median_income  median_house_value ocean_proximity  \n",
       "0       322.0       126.0         8.3252            452600.0        NEAR BAY  \n",
       "1      2401.0      1138.0         8.3014            358500.0        NEAR BAY  \n",
       "2       496.0       177.0         7.2574            352100.0        NEAR BAY  \n",
       "3       558.0       219.0         5.6431            341300.0        NEAR BAY  \n",
       "4       565.0       259.0         3.8462            342200.0        NEAR BAY  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing = load_housing_data()\n",
    "housing.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:14.226550Z",
     "iopub.status.busy": "2021-11-03T03:33:14.225717Z",
     "iopub.status.idle": "2021-11-03T03:33:14.244807Z",
     "shell.execute_reply": "2021-11-03T03:33:14.245286Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 20640 entries, 0 to 20639\n",
      "Data columns (total 10 columns):\n",
      " #   Column              Non-Null Count  Dtype  \n",
      "---  ------              --------------  -----  \n",
      " 0   longitude           20640 non-null  float64\n",
      " 1   latitude            20640 non-null  float64\n",
      " 2   housing_median_age  20640 non-null  float64\n",
      " 3   total_rooms         20640 non-null  float64\n",
      " 4   total_bedrooms      20433 non-null  float64\n",
      " 5   population          20640 non-null  float64\n",
      " 6   households          20640 non-null  float64\n",
      " 7   median_income       20640 non-null  float64\n",
      " 8   median_house_value  20640 non-null  float64\n",
      " 9   ocean_proximity     20640 non-null  object \n",
      "dtypes: float64(9), object(1)\n",
      "memory usage: 1.6+ MB\n"
     ]
    }
   ],
   "source": [
    "housing.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:14.248698Z",
     "iopub.status.busy": "2021-11-03T03:33:14.247808Z",
     "iopub.status.idle": "2021-11-03T03:33:14.254831Z",
     "shell.execute_reply": "2021-11-03T03:33:14.255301Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<1H OCEAN     9136\n",
       "INLAND        6551\n",
       "NEAR OCEAN    2658\n",
       "NEAR BAY      2290\n",
       "ISLAND           5\n",
       "Name: ocean_proximity, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing[\"ocean_proximity\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:14.258668Z",
     "iopub.status.busy": "2021-11-03T03:33:14.257790Z",
     "iopub.status.idle": "2021-11-03T03:33:14.296180Z",
     "shell.execute_reply": "2021-11-03T03:33:14.296614Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20433.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-119.569704</td>\n",
       "      <td>35.631861</td>\n",
       "      <td>28.639486</td>\n",
       "      <td>2635.763081</td>\n",
       "      <td>537.870553</td>\n",
       "      <td>1425.476744</td>\n",
       "      <td>499.539680</td>\n",
       "      <td>3.870671</td>\n",
       "      <td>206855.816909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.003532</td>\n",
       "      <td>2.135952</td>\n",
       "      <td>12.585558</td>\n",
       "      <td>2181.615252</td>\n",
       "      <td>421.385070</td>\n",
       "      <td>1132.462122</td>\n",
       "      <td>382.329753</td>\n",
       "      <td>1.899822</td>\n",
       "      <td>115395.615874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-124.350000</td>\n",
       "      <td>32.540000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.499900</td>\n",
       "      <td>14999.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-121.800000</td>\n",
       "      <td>33.930000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>1447.750000</td>\n",
       "      <td>296.000000</td>\n",
       "      <td>787.000000</td>\n",
       "      <td>280.000000</td>\n",
       "      <td>2.563400</td>\n",
       "      <td>119600.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-118.490000</td>\n",
       "      <td>34.260000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>2127.000000</td>\n",
       "      <td>435.000000</td>\n",
       "      <td>1166.000000</td>\n",
       "      <td>409.000000</td>\n",
       "      <td>3.534800</td>\n",
       "      <td>179700.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>-118.010000</td>\n",
       "      <td>37.710000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>3148.000000</td>\n",
       "      <td>647.000000</td>\n",
       "      <td>1725.000000</td>\n",
       "      <td>605.000000</td>\n",
       "      <td>4.743250</td>\n",
       "      <td>264725.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>-114.310000</td>\n",
       "      <td>41.950000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>39320.000000</td>\n",
       "      <td>6445.000000</td>\n",
       "      <td>35682.000000</td>\n",
       "      <td>6082.000000</td>\n",
       "      <td>15.000100</td>\n",
       "      <td>500001.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          longitude      latitude  housing_median_age   total_rooms  \\\n",
       "count  20640.000000  20640.000000        20640.000000  20640.000000   \n",
       "mean    -119.569704     35.631861           28.639486   2635.763081   \n",
       "std        2.003532      2.135952           12.585558   2181.615252   \n",
       "min     -124.350000     32.540000            1.000000      2.000000   \n",
       "25%     -121.800000     33.930000           18.000000   1447.750000   \n",
       "50%     -118.490000     34.260000           29.000000   2127.000000   \n",
       "75%     -118.010000     37.710000           37.000000   3148.000000   \n",
       "max     -114.310000     41.950000           52.000000  39320.000000   \n",
       "\n",
       "       total_bedrooms    population    households  median_income  \\\n",
       "count    20433.000000  20640.000000  20640.000000   20640.000000   \n",
       "mean       537.870553   1425.476744    499.539680       3.870671   \n",
       "std        421.385070   1132.462122    382.329753       1.899822   \n",
       "min          1.000000      3.000000      1.000000       0.499900   \n",
       "25%        296.000000    787.000000    280.000000       2.563400   \n",
       "50%        435.000000   1166.000000    409.000000       3.534800   \n",
       "75%        647.000000   1725.000000    605.000000       4.743250   \n",
       "max       6445.000000  35682.000000   6082.000000      15.000100   \n",
       "\n",
       "       median_house_value  \n",
       "count        20640.000000  \n",
       "mean        206855.816909  \n",
       "std         115395.615874  \n",
       "min          14999.000000  \n",
       "25%         119600.000000  \n",
       "50%         179700.000000  \n",
       "75%         264725.000000  \n",
       "max         500001.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:14.300168Z",
     "iopub.status.busy": "2021-11-03T03:33:14.299229Z",
     "iopub.status.idle": "2021-11-03T03:33:18.312132Z",
     "shell.execute_reply": "2021-11-03T03:33:18.312586Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "그림 저장: attribute_histogram_plots\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1080 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "housing.hist(bins=50, figsize=(20,15))\n",
    "save_fig(\"attribute_histogram_plots\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 테스트 세트 만들기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.315755Z",
     "iopub.status.busy": "2021-11-03T03:33:18.315140Z",
     "iopub.status.idle": "2021-11-03T03:33:18.317754Z",
     "shell.execute_reply": "2021-11-03T03:33:18.318170Z"
    }
   },
   "outputs": [],
   "source": [
    "# 노트북의 실행 결과가 동일하도록\n",
    "np.random.seed(42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.322672Z",
     "iopub.status.busy": "2021-11-03T03:33:18.320727Z",
     "iopub.status.idle": "2021-11-03T03:33:18.324858Z",
     "shell.execute_reply": "2021-11-03T03:33:18.324345Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "# 예시로 만든 것입니다. 실전에서는 사이킷런의 train_test_split()를 사용하세요.\n",
    "def split_train_test(data, test_ratio):\n",
    "    shuffled_indices = np.random.permutation(len(data))\n",
    "    test_set_size = int(len(data) * test_ratio)\n",
    "    test_indices = shuffled_indices[:test_set_size]\n",
    "    train_indices = shuffled_indices[test_set_size:]\n",
    "    return data.iloc[train_indices], data.iloc[test_indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.327803Z",
     "iopub.status.busy": "2021-11-03T03:33:18.327214Z",
     "iopub.status.idle": "2021-11-03T03:33:18.333390Z",
     "shell.execute_reply": "2021-11-03T03:33:18.333812Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16512"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_set, test_set = split_train_test(housing, 0.2)\n",
    "len(train_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.337352Z",
     "iopub.status.busy": "2021-11-03T03:33:18.336007Z",
     "iopub.status.idle": "2021-11-03T03:33:18.339416Z",
     "shell.execute_reply": "2021-11-03T03:33:18.339833Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4128"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.342589Z",
     "iopub.status.busy": "2021-11-03T03:33:18.341987Z",
     "iopub.status.idle": "2021-11-03T03:33:18.345834Z",
     "shell.execute_reply": "2021-11-03T03:33:18.346285Z"
    }
   },
   "outputs": [],
   "source": [
    "from zlib import crc32\n",
    "\n",
    "def test_set_check(identifier, test_ratio):\n",
    "    return crc32(np.int64(identifier)) & 0xffffffff < test_ratio * 2**32\n",
    "\n",
    "def split_train_test_by_id(data, test_ratio, id_column):\n",
    "    ids = data[id_column]\n",
    "    in_test_set = ids.apply(lambda id_: test_set_check(id_, test_ratio))\n",
    "    return data.loc[~in_test_set], data.loc[in_test_set]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "위의 `test_set_check()` 함수가 파이썬 2와 파이썬 3에서 모두 잘 동작합니다. 초판에서는 모든 해시 함수를 지원하는 다음 방식을 제안했지만 느리고 파이썬 2를 지원하지 않습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.349199Z",
     "iopub.status.busy": "2021-11-03T03:33:18.348603Z",
     "iopub.status.idle": "2021-11-03T03:33:18.351671Z",
     "shell.execute_reply": "2021-11-03T03:33:18.352093Z"
    }
   },
   "outputs": [],
   "source": [
    "import hashlib\n",
    "\n",
    "def test_set_check(identifier, test_ratio, hash=hashlib.md5):\n",
    "    return hash(np.int64(identifier)).digest()[-1] < 256 * test_ratio"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "모든 해시 함수를 지원하고 파이썬 2와 파이썬 3에서 사용할 수 있는 함수를 원한다면 다음을 사용하세요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.355053Z",
     "iopub.status.busy": "2021-11-03T03:33:18.354443Z",
     "iopub.status.idle": "2021-11-03T03:33:18.357390Z",
     "shell.execute_reply": "2021-11-03T03:33:18.357833Z"
    }
   },
   "outputs": [],
   "source": [
    "def test_set_check(identifier, test_ratio, hash=hashlib.md5):\n",
    "    return bytearray(hash(np.int64(identifier)).digest())[-1] < 256 * test_ratio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.360615Z",
     "iopub.status.busy": "2021-11-03T03:33:18.360003Z",
     "iopub.status.idle": "2021-11-03T03:33:18.401559Z",
     "shell.execute_reply": "2021-11-03T03:33:18.402030Z"
    }
   },
   "outputs": [],
   "source": [
    "housing_with_id = housing.reset_index()   # `index` 열이 추가된 데이터프레임을 반환합니다\n",
    "train_set, test_set = split_train_test_by_id(housing_with_id, 0.2, \"index\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.405108Z",
     "iopub.status.busy": "2021-11-03T03:33:18.404492Z",
     "iopub.status.idle": "2021-11-03T03:33:18.446507Z",
     "shell.execute_reply": "2021-11-03T03:33:18.446996Z"
    }
   },
   "outputs": [],
   "source": [
    "housing_with_id[\"id\"] = housing[\"longitude\"] * 1000 + housing[\"latitude\"]\n",
    "train_set, test_set = split_train_test_by_id(housing_with_id, 0.2, \"id\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.450075Z",
     "iopub.status.busy": "2021-11-03T03:33:18.449475Z",
     "iopub.status.idle": "2021-11-03T03:33:18.466287Z",
     "shell.execute_reply": "2021-11-03T03:33:18.466728Z"
    }
   },
   "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>index</th>\n",
       "      <th>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>ocean_proximity</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>-122.26</td>\n",
       "      <td>37.84</td>\n",
       "      <td>42.0</td>\n",
       "      <td>2555.0</td>\n",
       "      <td>665.0</td>\n",
       "      <td>1206.0</td>\n",
       "      <td>595.0</td>\n",
       "      <td>2.0804</td>\n",
       "      <td>226700.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "      <td>-122222.16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>10</td>\n",
       "      <td>-122.26</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>2202.0</td>\n",
       "      <td>434.0</td>\n",
       "      <td>910.0</td>\n",
       "      <td>402.0</td>\n",
       "      <td>3.2031</td>\n",
       "      <td>281500.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "      <td>-122222.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>11</td>\n",
       "      <td>-122.26</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>3503.0</td>\n",
       "      <td>752.0</td>\n",
       "      <td>1504.0</td>\n",
       "      <td>734.0</td>\n",
       "      <td>3.2705</td>\n",
       "      <td>241800.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "      <td>-122222.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>12</td>\n",
       "      <td>-122.26</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>2491.0</td>\n",
       "      <td>474.0</td>\n",
       "      <td>1098.0</td>\n",
       "      <td>468.0</td>\n",
       "      <td>3.0750</td>\n",
       "      <td>213500.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "      <td>-122222.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>13</td>\n",
       "      <td>-122.26</td>\n",
       "      <td>37.84</td>\n",
       "      <td>52.0</td>\n",
       "      <td>696.0</td>\n",
       "      <td>191.0</td>\n",
       "      <td>345.0</td>\n",
       "      <td>174.0</td>\n",
       "      <td>2.6736</td>\n",
       "      <td>191300.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "      <td>-122222.16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    index  longitude  latitude  housing_median_age  total_rooms  \\\n",
       "8       8    -122.26     37.84                42.0       2555.0   \n",
       "10     10    -122.26     37.85                52.0       2202.0   \n",
       "11     11    -122.26     37.85                52.0       3503.0   \n",
       "12     12    -122.26     37.85                52.0       2491.0   \n",
       "13     13    -122.26     37.84                52.0        696.0   \n",
       "\n",
       "    total_bedrooms  population  households  median_income  median_house_value  \\\n",
       "8            665.0      1206.0       595.0         2.0804            226700.0   \n",
       "10           434.0       910.0       402.0         3.2031            281500.0   \n",
       "11           752.0      1504.0       734.0         3.2705            241800.0   \n",
       "12           474.0      1098.0       468.0         3.0750            213500.0   \n",
       "13           191.0       345.0       174.0         2.6736            191300.0   \n",
       "\n",
       "   ocean_proximity         id  \n",
       "8         NEAR BAY -122222.16  \n",
       "10        NEAR BAY -122222.15  \n",
       "11        NEAR BAY -122222.15  \n",
       "12        NEAR BAY -122222.15  \n",
       "13        NEAR BAY -122222.16  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.469951Z",
     "iopub.status.busy": "2021-11-03T03:33:18.469351Z",
     "iopub.status.idle": "2021-11-03T03:33:18.489933Z",
     "shell.execute_reply": "2021-11-03T03:33:18.490402Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "train_set, test_set = train_test_split(housing, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.493295Z",
     "iopub.status.busy": "2021-11-03T03:33:18.492678Z",
     "iopub.status.idle": "2021-11-03T03:33:18.505507Z",
     "shell.execute_reply": "2021-11-03T03:33:18.506015Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>20046</th>\n",
       "      <td>-119.01</td>\n",
       "      <td>36.06</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1505.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1392.0</td>\n",
       "      <td>359.0</td>\n",
       "      <td>1.6812</td>\n",
       "      <td>47700.0</td>\n",
       "      <td>INLAND</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3024</th>\n",
       "      <td>-119.46</td>\n",
       "      <td>35.14</td>\n",
       "      <td>30.0</td>\n",
       "      <td>2943.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1565.0</td>\n",
       "      <td>584.0</td>\n",
       "      <td>2.5313</td>\n",
       "      <td>45800.0</td>\n",
       "      <td>INLAND</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15663</th>\n",
       "      <td>-122.44</td>\n",
       "      <td>37.80</td>\n",
       "      <td>52.0</td>\n",
       "      <td>3830.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1310.0</td>\n",
       "      <td>963.0</td>\n",
       "      <td>3.4801</td>\n",
       "      <td>500001.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20484</th>\n",
       "      <td>-118.72</td>\n",
       "      <td>34.28</td>\n",
       "      <td>17.0</td>\n",
       "      <td>3051.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1705.0</td>\n",
       "      <td>495.0</td>\n",
       "      <td>5.7376</td>\n",
       "      <td>218600.0</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9814</th>\n",
       "      <td>-121.93</td>\n",
       "      <td>36.62</td>\n",
       "      <td>34.0</td>\n",
       "      <td>2351.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1063.0</td>\n",
       "      <td>428.0</td>\n",
       "      <td>3.7250</td>\n",
       "      <td>278000.0</td>\n",
       "      <td>NEAR OCEAN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "20046    -119.01     36.06                25.0       1505.0             NaN   \n",
       "3024     -119.46     35.14                30.0       2943.0             NaN   \n",
       "15663    -122.44     37.80                52.0       3830.0             NaN   \n",
       "20484    -118.72     34.28                17.0       3051.0             NaN   \n",
       "9814     -121.93     36.62                34.0       2351.0             NaN   \n",
       "\n",
       "       population  households  median_income  median_house_value  \\\n",
       "20046      1392.0       359.0         1.6812             47700.0   \n",
       "3024       1565.0       584.0         2.5313             45800.0   \n",
       "15663      1310.0       963.0         3.4801            500001.0   \n",
       "20484      1705.0       495.0         5.7376            218600.0   \n",
       "9814       1063.0       428.0         3.7250            278000.0   \n",
       "\n",
       "      ocean_proximity  \n",
       "20046          INLAND  \n",
       "3024           INLAND  \n",
       "15663        NEAR BAY  \n",
       "20484       <1H OCEAN  \n",
       "9814       NEAR OCEAN  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.509017Z",
     "iopub.status.busy": "2021-11-03T03:33:18.508427Z",
     "iopub.status.idle": "2021-11-03T03:33:18.745183Z",
     "shell.execute_reply": "2021-11-03T03:33:18.745655Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 22,
     "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": [
    "housing[\"median_income\"].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.751557Z",
     "iopub.status.busy": "2021-11-03T03:33:18.750953Z",
     "iopub.status.idle": "2021-11-03T03:33:18.756874Z",
     "shell.execute_reply": "2021-11-03T03:33:18.756268Z"
    }
   },
   "outputs": [],
   "source": [
    "housing[\"income_cat\"] = pd.cut(housing[\"median_income\"],\n",
    "                               bins=[0., 1.5, 3.0, 4.5, 6., np.inf],\n",
    "                               labels=[1, 2, 3, 4, 5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.760898Z",
     "iopub.status.busy": "2021-11-03T03:33:18.760319Z",
     "iopub.status.idle": "2021-11-03T03:33:18.765863Z",
     "shell.execute_reply": "2021-11-03T03:33:18.766260Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3    7236\n",
       "2    6581\n",
       "4    3639\n",
       "5    2362\n",
       "1     822\n",
       "Name: income_cat, dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing[\"income_cat\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.788121Z",
     "iopub.status.busy": "2021-11-03T03:33:18.787509Z",
     "iopub.status.idle": "2021-11-03T03:33:18.942090Z",
     "shell.execute_reply": "2021-11-03T03:33:18.941511Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 25,
     "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": [
    "housing[\"income_cat\"].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.945448Z",
     "iopub.status.busy": "2021-11-03T03:33:18.944607Z",
     "iopub.status.idle": "2021-11-03T03:33:18.959420Z",
     "shell.execute_reply": "2021-11-03T03:33:18.959841Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "\n",
    "split = StratifiedShuffleSplit(n_splits=1, test_size=0.2, random_state=42)\n",
    "for train_index, test_index in split.split(housing, housing[\"income_cat\"]):\n",
    "    strat_train_set = housing.loc[train_index]\n",
    "    strat_test_set = housing.loc[test_index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.967214Z",
     "iopub.status.busy": "2021-11-03T03:33:18.966607Z",
     "iopub.status.idle": "2021-11-03T03:33:18.971343Z",
     "shell.execute_reply": "2021-11-03T03:33:18.971849Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3    0.350533\n",
       "2    0.318798\n",
       "4    0.176357\n",
       "5    0.114341\n",
       "1    0.039971\n",
       "Name: income_cat, dtype: float64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "strat_test_set[\"income_cat\"].value_counts() / len(strat_test_set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.978197Z",
     "iopub.status.busy": "2021-11-03T03:33:18.975370Z",
     "iopub.status.idle": "2021-11-03T03:33:18.980900Z",
     "shell.execute_reply": "2021-11-03T03:33:18.981377Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3    0.350581\n",
       "2    0.318847\n",
       "4    0.176308\n",
       "5    0.114438\n",
       "1    0.039826\n",
       "Name: income_cat, dtype: float64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing[\"income_cat\"].value_counts() / len(housing)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:18.988066Z",
     "iopub.status.busy": "2021-11-03T03:33:18.984094Z",
     "iopub.status.idle": "2021-11-03T03:33:18.999541Z",
     "shell.execute_reply": "2021-11-03T03:33:18.999980Z"
    }
   },
   "outputs": [],
   "source": [
    "def income_cat_proportions(data):\n",
    "    return data[\"income_cat\"].value_counts() / len(data)\n",
    "\n",
    "train_set, test_set = train_test_split(housing, test_size=0.2, random_state=42)\n",
    "\n",
    "compare_props = pd.DataFrame({\n",
    "    \"Overall\": income_cat_proportions(housing),\n",
    "    \"Stratified\": income_cat_proportions(strat_test_set),\n",
    "    \"Random\": income_cat_proportions(test_set),\n",
    "}).sort_index()\n",
    "compare_props[\"Rand. %error\"] = 100 * compare_props[\"Random\"] / compare_props[\"Overall\"] - 100\n",
    "compare_props[\"Strat. %error\"] = 100 * compare_props[\"Stratified\"] / compare_props[\"Overall\"] - 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:19.010365Z",
     "iopub.status.busy": "2021-11-03T03:33:19.009197Z",
     "iopub.status.idle": "2021-11-03T03:33:19.013153Z",
     "shell.execute_reply": "2021-11-03T03:33:19.012615Z"
    }
   },
   "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>Overall</th>\n",
       "      <th>Stratified</th>\n",
       "      <th>Random</th>\n",
       "      <th>Rand. %error</th>\n",
       "      <th>Strat. %error</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.039826</td>\n",
       "      <td>0.039971</td>\n",
       "      <td>0.040213</td>\n",
       "      <td>0.973236</td>\n",
       "      <td>0.364964</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.318847</td>\n",
       "      <td>0.318798</td>\n",
       "      <td>0.324370</td>\n",
       "      <td>1.732260</td>\n",
       "      <td>-0.015195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.350581</td>\n",
       "      <td>0.350533</td>\n",
       "      <td>0.358527</td>\n",
       "      <td>2.266446</td>\n",
       "      <td>-0.013820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.176308</td>\n",
       "      <td>0.176357</td>\n",
       "      <td>0.167393</td>\n",
       "      <td>-5.056334</td>\n",
       "      <td>0.027480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.114438</td>\n",
       "      <td>0.114341</td>\n",
       "      <td>0.109496</td>\n",
       "      <td>-4.318374</td>\n",
       "      <td>-0.084674</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Overall  Stratified    Random  Rand. %error  Strat. %error\n",
       "1  0.039826    0.039971  0.040213      0.973236       0.364964\n",
       "2  0.318847    0.318798  0.324370      1.732260      -0.015195\n",
       "3  0.350581    0.350533  0.358527      2.266446      -0.013820\n",
       "4  0.176308    0.176357  0.167393     -5.056334       0.027480\n",
       "5  0.114438    0.114341  0.109496     -4.318374      -0.084674"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compare_props"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:19.018592Z",
     "iopub.status.busy": "2021-11-03T03:33:19.017163Z",
     "iopub.status.idle": "2021-11-03T03:33:19.020268Z",
     "shell.execute_reply": "2021-11-03T03:33:19.020674Z"
    }
   },
   "outputs": [],
   "source": [
    "for set_ in (strat_train_set, strat_test_set):\n",
    "    set_.drop(\"income_cat\", axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 데이터 이해를 위한 탐색과 시각화"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:19.025342Z",
     "iopub.status.busy": "2021-11-03T03:33:19.024069Z",
     "iopub.status.idle": "2021-11-03T03:33:19.025995Z",
     "shell.execute_reply": "2021-11-03T03:33:19.026418Z"
    }
   },
   "outputs": [],
   "source": [
    "housing = strat_train_set.copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 지리적 데이터 시각화"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:19.030230Z",
     "iopub.status.busy": "2021-11-03T03:33:19.029597Z",
     "iopub.status.idle": "2021-11-03T03:33:19.537331Z",
     "shell.execute_reply": "2021-11-03T03:33:19.536739Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "그림 저장: bad_visualization_plot\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\")\n",
    "save_fig(\"bad_visualization_plot\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:19.541408Z",
     "iopub.status.busy": "2021-11-03T03:33:19.540778Z",
     "iopub.status.idle": "2021-11-03T03:33:20.146810Z",
     "shell.execute_reply": "2021-11-03T03:33:20.147255Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "그림 저장: better_visualization_plot\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", alpha=0.1)\n",
    "save_fig(\"better_visualization_plot\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`sharex=False` 매개변수는 x-축의 값과 범례를 표시하지 못하는 버그를 수정합니다. 이는 임시 방편입니다(https://github.com/pandas-dev/pandas/issues/10611 참조). 수정 사항을 알려준 Wilmer Arellano에게 감사합니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:20.153472Z",
     "iopub.status.busy": "2021-11-03T03:33:20.152803Z",
     "iopub.status.idle": "2021-11-03T03:33:22.119603Z",
     "shell.execute_reply": "2021-11-03T03:33:22.120040Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "그림 저장: housing_prices_scatterplot\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", alpha=0.4,\n",
    "             s=housing[\"population\"]/100, label=\"population\", figsize=(10,7),\n",
    "             c=\"median_house_value\", cmap=plt.get_cmap(\"jet\"), colorbar=True,\n",
    "             sharex=False)\n",
    "plt.legend()\n",
    "save_fig(\"housing_prices_scatterplot\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:22.125328Z",
     "iopub.status.busy": "2021-11-03T03:33:22.124304Z",
     "iopub.status.idle": "2021-11-03T03:33:22.921494Z",
     "shell.execute_reply": "2021-11-03T03:33:22.921959Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading california.png\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "('./images/end_to_end_project/california.png',\n",
       " <http.client.HTTPMessage at 0x7f707866c3c8>)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Download the California image\n",
    "images_path = os.path.join(PROJECT_ROOT_DIR, \"images\", \"end_to_end_project\")\n",
    "os.makedirs(images_path, exist_ok=True)\n",
    "DOWNLOAD_ROOT = \"https://raw.githubusercontent.com/ageron/handson-ml2/master/\"\n",
    "filename = \"california.png\"\n",
    "print(\"Downloading\", filename)\n",
    "url = DOWNLOAD_ROOT + \"images/end_to_end_project/\" + filename\n",
    "urllib.request.urlretrieve(url, os.path.join(images_path, filename))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:22.930537Z",
     "iopub.status.busy": "2021-11-03T03:33:22.929650Z",
     "iopub.status.idle": "2021-11-03T03:33:25.718845Z",
     "shell.execute_reply": "2021-11-03T03:33:25.717811Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "그림 저장: california_housing_prices_plot\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.image as mpimg\n",
    "california_img=mpimg.imread(os.path.join(images_path, filename))\n",
    "ax = housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", figsize=(10,7),\n",
    "                  s=housing['population']/100, label=\"Population\",\n",
    "                  c=\"median_house_value\", cmap=plt.get_cmap(\"jet\"),\n",
    "                  colorbar=False, alpha=0.4)\n",
    "plt.imshow(california_img, extent=[-124.55, -113.80, 32.45, 42.05], alpha=0.5,\n",
    "           cmap=plt.get_cmap(\"jet\"))\n",
    "plt.ylabel(\"Latitude\", fontsize=14)\n",
    "plt.xlabel(\"Longitude\", fontsize=14)\n",
    "\n",
    "prices = housing[\"median_house_value\"]\n",
    "tick_values = np.linspace(prices.min(), prices.max(), 11)\n",
    "cbar = plt.colorbar(ticks=tick_values/prices.max())\n",
    "cbar.ax.set_yticklabels([\"$%dk\"%(round(v/1000)) for v in tick_values], fontsize=14)\n",
    "cbar.set_label('Median House Value', fontsize=16)\n",
    "\n",
    "plt.legend(fontsize=16)\n",
    "save_fig(\"california_housing_prices_plot\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 상관관계 조사"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:25.727426Z",
     "iopub.status.busy": "2021-11-03T03:33:25.726070Z",
     "iopub.status.idle": "2021-11-03T03:33:25.749740Z",
     "shell.execute_reply": "2021-11-03T03:33:25.750515Z"
    }
   },
   "outputs": [],
   "source": [
    "corr_matrix = housing.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:25.757201Z",
     "iopub.status.busy": "2021-11-03T03:33:25.756484Z",
     "iopub.status.idle": "2021-11-03T03:33:25.759283Z",
     "shell.execute_reply": "2021-11-03T03:33:25.759723Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "median_house_value    1.000000\n",
       "median_income         0.687151\n",
       "total_rooms           0.135140\n",
       "housing_median_age    0.114146\n",
       "households            0.064590\n",
       "total_bedrooms        0.047781\n",
       "population           -0.026882\n",
       "longitude            -0.047466\n",
       "latitude             -0.142673\n",
       "Name: median_house_value, dtype: float64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_matrix[\"median_house_value\"].sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "피어슨의 상관 계수(위키백과):\n",
    "![Pearson correlation coefficient](https://upload.wikimedia.org/wikipedia/commons/d/d4/Correlation_examples2.svg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:25.856882Z",
     "iopub.status.busy": "2021-11-03T03:33:25.822855Z",
     "iopub.status.idle": "2021-11-03T03:33:30.654572Z",
     "shell.execute_reply": "2021-11-03T03:33:30.655032Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "그림 저장: scatter_matrix_plot\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# from pandas.tools.plotting import scatter_matrix # 옛날 버전의 판다스에서는\n",
    "from pandas.plotting import scatter_matrix\n",
    "\n",
    "attributes = [\"median_house_value\", \"median_income\", \"total_rooms\",\n",
    "              \"housing_median_age\"]\n",
    "scatter_matrix(housing[attributes], figsize=(12, 8))\n",
    "save_fig(\"scatter_matrix_plot\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:30.660366Z",
     "iopub.status.busy": "2021-11-03T03:33:30.659743Z",
     "iopub.status.idle": "2021-11-03T03:33:31.367943Z",
     "shell.execute_reply": "2021-11-03T03:33:31.368396Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "그림 저장: income_vs_house_value_scatterplot\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "housing.plot(kind=\"scatter\", x=\"median_income\", y=\"median_house_value\",\n",
    "             alpha=0.1)\n",
    "plt.axis([0, 16, 0, 550000])\n",
    "save_fig(\"income_vs_house_value_scatterplot\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 특성 조합으로 실험"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.376116Z",
     "iopub.status.busy": "2021-11-03T03:33:31.375258Z",
     "iopub.status.idle": "2021-11-03T03:33:31.378026Z",
     "shell.execute_reply": "2021-11-03T03:33:31.377343Z"
    }
   },
   "outputs": [],
   "source": [
    "housing[\"rooms_per_household\"] = housing[\"total_rooms\"]/housing[\"households\"]\n",
    "housing[\"bedrooms_per_room\"] = housing[\"total_bedrooms\"]/housing[\"total_rooms\"]\n",
    "housing[\"population_per_household\"]=housing[\"population\"]/housing[\"households\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.382493Z",
     "iopub.status.busy": "2021-11-03T03:33:31.381875Z",
     "iopub.status.idle": "2021-11-03T03:33:31.395067Z",
     "shell.execute_reply": "2021-11-03T03:33:31.395505Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "median_house_value          1.000000\n",
       "median_income               0.687151\n",
       "rooms_per_household         0.146255\n",
       "total_rooms                 0.135140\n",
       "housing_median_age          0.114146\n",
       "households                  0.064590\n",
       "total_bedrooms              0.047781\n",
       "population_per_household   -0.021991\n",
       "population                 -0.026882\n",
       "longitude                  -0.047466\n",
       "latitude                   -0.142673\n",
       "bedrooms_per_room          -0.259952\n",
       "Name: median_house_value, dtype: float64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_matrix = housing.corr()\n",
    "corr_matrix[\"median_house_value\"].sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.399652Z",
     "iopub.status.busy": "2021-11-03T03:33:31.398973Z",
     "iopub.status.idle": "2021-11-03T03:33:31.550008Z",
     "shell.execute_reply": "2021-11-03T03:33:31.550440Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "housing.plot(kind=\"scatter\", x=\"rooms_per_household\", y=\"median_house_value\",\n",
    "             alpha=0.2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.555341Z",
     "iopub.status.busy": "2021-11-03T03:33:31.554761Z",
     "iopub.status.idle": "2021-11-03T03:33:31.599075Z",
     "shell.execute_reply": "2021-11-03T03:33:31.599521Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>rooms_per_household</th>\n",
       "      <th>bedrooms_per_room</th>\n",
       "      <th>population_per_household</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16354.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "      <td>16354.000000</td>\n",
       "      <td>16512.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-119.575635</td>\n",
       "      <td>35.639314</td>\n",
       "      <td>28.653404</td>\n",
       "      <td>2622.539789</td>\n",
       "      <td>534.914639</td>\n",
       "      <td>1419.687379</td>\n",
       "      <td>497.011810</td>\n",
       "      <td>3.875884</td>\n",
       "      <td>207005.322372</td>\n",
       "      <td>5.440406</td>\n",
       "      <td>0.212873</td>\n",
       "      <td>3.096469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.001828</td>\n",
       "      <td>2.137963</td>\n",
       "      <td>12.574819</td>\n",
       "      <td>2138.417080</td>\n",
       "      <td>412.665649</td>\n",
       "      <td>1115.663036</td>\n",
       "      <td>375.696156</td>\n",
       "      <td>1.904931</td>\n",
       "      <td>115701.297250</td>\n",
       "      <td>2.611696</td>\n",
       "      <td>0.057378</td>\n",
       "      <td>11.584825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-124.350000</td>\n",
       "      <td>32.540000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.499900</td>\n",
       "      <td>14999.000000</td>\n",
       "      <td>1.130435</td>\n",
       "      <td>0.100000</td>\n",
       "      <td>0.692308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-121.800000</td>\n",
       "      <td>33.940000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>1443.000000</td>\n",
       "      <td>295.000000</td>\n",
       "      <td>784.000000</td>\n",
       "      <td>279.000000</td>\n",
       "      <td>2.566950</td>\n",
       "      <td>119800.000000</td>\n",
       "      <td>4.442168</td>\n",
       "      <td>0.175304</td>\n",
       "      <td>2.431352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-118.510000</td>\n",
       "      <td>34.260000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>2119.000000</td>\n",
       "      <td>433.000000</td>\n",
       "      <td>1164.000000</td>\n",
       "      <td>408.000000</td>\n",
       "      <td>3.541550</td>\n",
       "      <td>179500.000000</td>\n",
       "      <td>5.232342</td>\n",
       "      <td>0.203027</td>\n",
       "      <td>2.817661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>-118.010000</td>\n",
       "      <td>37.720000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>3141.000000</td>\n",
       "      <td>644.000000</td>\n",
       "      <td>1719.000000</td>\n",
       "      <td>602.000000</td>\n",
       "      <td>4.745325</td>\n",
       "      <td>263900.000000</td>\n",
       "      <td>6.056361</td>\n",
       "      <td>0.239816</td>\n",
       "      <td>3.281420</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>-114.310000</td>\n",
       "      <td>41.950000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>39320.000000</td>\n",
       "      <td>6210.000000</td>\n",
       "      <td>35682.000000</td>\n",
       "      <td>5358.000000</td>\n",
       "      <td>15.000100</td>\n",
       "      <td>500001.000000</td>\n",
       "      <td>141.909091</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1243.333333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          longitude      latitude  housing_median_age   total_rooms  \\\n",
       "count  16512.000000  16512.000000        16512.000000  16512.000000   \n",
       "mean    -119.575635     35.639314           28.653404   2622.539789   \n",
       "std        2.001828      2.137963           12.574819   2138.417080   \n",
       "min     -124.350000     32.540000            1.000000      6.000000   \n",
       "25%     -121.800000     33.940000           18.000000   1443.000000   \n",
       "50%     -118.510000     34.260000           29.000000   2119.000000   \n",
       "75%     -118.010000     37.720000           37.000000   3141.000000   \n",
       "max     -114.310000     41.950000           52.000000  39320.000000   \n",
       "\n",
       "       total_bedrooms    population    households  median_income  \\\n",
       "count    16354.000000  16512.000000  16512.000000   16512.000000   \n",
       "mean       534.914639   1419.687379    497.011810       3.875884   \n",
       "std        412.665649   1115.663036    375.696156       1.904931   \n",
       "min          2.000000      3.000000      2.000000       0.499900   \n",
       "25%        295.000000    784.000000    279.000000       2.566950   \n",
       "50%        433.000000   1164.000000    408.000000       3.541550   \n",
       "75%        644.000000   1719.000000    602.000000       4.745325   \n",
       "max       6210.000000  35682.000000   5358.000000      15.000100   \n",
       "\n",
       "       median_house_value  rooms_per_household  bedrooms_per_room  \\\n",
       "count        16512.000000         16512.000000       16354.000000   \n",
       "mean        207005.322372             5.440406           0.212873   \n",
       "std         115701.297250             2.611696           0.057378   \n",
       "min          14999.000000             1.130435           0.100000   \n",
       "25%         119800.000000             4.442168           0.175304   \n",
       "50%         179500.000000             5.232342           0.203027   \n",
       "75%         263900.000000             6.056361           0.239816   \n",
       "max         500001.000000           141.909091           1.000000   \n",
       "\n",
       "       population_per_household  \n",
       "count              16512.000000  \n",
       "mean                   3.096469  \n",
       "std                   11.584825  \n",
       "min                    0.692308  \n",
       "25%                    2.431352  \n",
       "50%                    2.817661  \n",
       "75%                    3.281420  \n",
       "max                 1243.333333  "
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 머신러닝 알고리즘을 위한 데이터 준비"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.605466Z",
     "iopub.status.busy": "2021-11-03T03:33:31.604439Z",
     "iopub.status.idle": "2021-11-03T03:33:31.607367Z",
     "shell.execute_reply": "2021-11-03T03:33:31.608010Z"
    }
   },
   "outputs": [],
   "source": [
    "housing = strat_train_set.drop(\"median_house_value\", axis=1) # 훈련 세트를 위해 레이블 삭제\n",
    "housing_labels = strat_train_set[\"median_house_value\"].copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 데이터 정제\n",
    "\n",
    "책에 소개된 세 개의 옵션은 다음과 같습니다:\n",
    "\n",
    "```\n",
    "housing.dropna(subset=[\"total_bedrooms\"])    # 옵션 1\n",
    "housing.drop(\"total_bedrooms\", axis=1)       # 옵션 2\n",
    "median = housing[\"total_bedrooms\"].median()  # 옵션 3\n",
    "housing[\"total_bedrooms\"].fillna(median, inplace=True)\n",
    "```\n",
    "\n",
    "각 옵션을 설명하기 위해 주택 데이터셋의 복사본을 만듭니다. 이 때 적어도 하나의 열이 비어 있는 행만 고릅니다. 이렇게 하면 각 옵션의 정확한 동작을 눈으로 쉽게 확인할 수 있습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.612997Z",
     "iopub.status.busy": "2021-11-03T03:33:31.611897Z",
     "iopub.status.idle": "2021-11-03T03:33:31.626155Z",
     "shell.execute_reply": "2021-11-03T03:33:31.626857Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1606</th>\n",
       "      <td>-122.08</td>\n",
       "      <td>37.88</td>\n",
       "      <td>26.0</td>\n",
       "      <td>2947.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>825.0</td>\n",
       "      <td>626.0</td>\n",
       "      <td>2.9330</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10915</th>\n",
       "      <td>-117.87</td>\n",
       "      <td>33.73</td>\n",
       "      <td>45.0</td>\n",
       "      <td>2264.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1970.0</td>\n",
       "      <td>499.0</td>\n",
       "      <td>3.4193</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19150</th>\n",
       "      <td>-122.70</td>\n",
       "      <td>38.35</td>\n",
       "      <td>14.0</td>\n",
       "      <td>2313.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>954.0</td>\n",
       "      <td>397.0</td>\n",
       "      <td>3.7813</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4186</th>\n",
       "      <td>-118.23</td>\n",
       "      <td>34.13</td>\n",
       "      <td>48.0</td>\n",
       "      <td>1308.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>835.0</td>\n",
       "      <td>294.0</td>\n",
       "      <td>4.2891</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16885</th>\n",
       "      <td>-122.40</td>\n",
       "      <td>37.58</td>\n",
       "      <td>26.0</td>\n",
       "      <td>3281.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1145.0</td>\n",
       "      <td>480.0</td>\n",
       "      <td>6.3580</td>\n",
       "      <td>NEAR OCEAN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "1606     -122.08     37.88                26.0       2947.0             NaN   \n",
       "10915    -117.87     33.73                45.0       2264.0             NaN   \n",
       "19150    -122.70     38.35                14.0       2313.0             NaN   \n",
       "4186     -118.23     34.13                48.0       1308.0             NaN   \n",
       "16885    -122.40     37.58                26.0       3281.0             NaN   \n",
       "\n",
       "       population  households  median_income ocean_proximity  \n",
       "1606        825.0       626.0         2.9330        NEAR BAY  \n",
       "10915      1970.0       499.0         3.4193       <1H OCEAN  \n",
       "19150       954.0       397.0         3.7813       <1H OCEAN  \n",
       "4186        835.0       294.0         4.2891       <1H OCEAN  \n",
       "16885      1145.0       480.0         6.3580      NEAR OCEAN  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_incomplete_rows = housing[housing.isnull().any(axis=1)].head()\n",
    "sample_incomplete_rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.636329Z",
     "iopub.status.busy": "2021-11-03T03:33:31.635372Z",
     "iopub.status.idle": "2021-11-03T03:33:31.639684Z",
     "shell.execute_reply": "2021-11-03T03:33:31.640389Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [longitude, latitude, housing_median_age, total_rooms, total_bedrooms, population, households, median_income, ocean_proximity]\n",
       "Index: []"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_incomplete_rows.dropna(subset=[\"total_bedrooms\"])    # 옵션 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.653883Z",
     "iopub.status.busy": "2021-11-03T03:33:31.652890Z",
     "iopub.status.idle": "2021-11-03T03:33:31.657314Z",
     "shell.execute_reply": "2021-11-03T03:33:31.657986Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1606</th>\n",
       "      <td>-122.08</td>\n",
       "      <td>37.88</td>\n",
       "      <td>26.0</td>\n",
       "      <td>2947.0</td>\n",
       "      <td>825.0</td>\n",
       "      <td>626.0</td>\n",
       "      <td>2.9330</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10915</th>\n",
       "      <td>-117.87</td>\n",
       "      <td>33.73</td>\n",
       "      <td>45.0</td>\n",
       "      <td>2264.0</td>\n",
       "      <td>1970.0</td>\n",
       "      <td>499.0</td>\n",
       "      <td>3.4193</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19150</th>\n",
       "      <td>-122.70</td>\n",
       "      <td>38.35</td>\n",
       "      <td>14.0</td>\n",
       "      <td>2313.0</td>\n",
       "      <td>954.0</td>\n",
       "      <td>397.0</td>\n",
       "      <td>3.7813</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4186</th>\n",
       "      <td>-118.23</td>\n",
       "      <td>34.13</td>\n",
       "      <td>48.0</td>\n",
       "      <td>1308.0</td>\n",
       "      <td>835.0</td>\n",
       "      <td>294.0</td>\n",
       "      <td>4.2891</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16885</th>\n",
       "      <td>-122.40</td>\n",
       "      <td>37.58</td>\n",
       "      <td>26.0</td>\n",
       "      <td>3281.0</td>\n",
       "      <td>1145.0</td>\n",
       "      <td>480.0</td>\n",
       "      <td>6.3580</td>\n",
       "      <td>NEAR OCEAN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  population  \\\n",
       "1606     -122.08     37.88                26.0       2947.0       825.0   \n",
       "10915    -117.87     33.73                45.0       2264.0      1970.0   \n",
       "19150    -122.70     38.35                14.0       2313.0       954.0   \n",
       "4186     -118.23     34.13                48.0       1308.0       835.0   \n",
       "16885    -122.40     37.58                26.0       3281.0      1145.0   \n",
       "\n",
       "       households  median_income ocean_proximity  \n",
       "1606        626.0         2.9330        NEAR BAY  \n",
       "10915       499.0         3.4193       <1H OCEAN  \n",
       "19150       397.0         3.7813       <1H OCEAN  \n",
       "4186        294.0         4.2891       <1H OCEAN  \n",
       "16885       480.0         6.3580      NEAR OCEAN  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_incomplete_rows.drop(\"total_bedrooms\", axis=1)       # 옵션 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.663142Z",
     "iopub.status.busy": "2021-11-03T03:33:31.662195Z",
     "iopub.status.idle": "2021-11-03T03:33:31.665569Z",
     "shell.execute_reply": "2021-11-03T03:33:31.666231Z"
    }
   },
   "outputs": [],
   "source": [
    "median = housing[\"total_bedrooms\"].median()\n",
    "sample_incomplete_rows[\"total_bedrooms\"].fillna(median, inplace=True) # 옵션 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.679587Z",
     "iopub.status.busy": "2021-11-03T03:33:31.678620Z",
     "iopub.status.idle": "2021-11-03T03:33:31.683061Z",
     "shell.execute_reply": "2021-11-03T03:33:31.683748Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1606</th>\n",
       "      <td>-122.08</td>\n",
       "      <td>37.88</td>\n",
       "      <td>26.0</td>\n",
       "      <td>2947.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>825.0</td>\n",
       "      <td>626.0</td>\n",
       "      <td>2.9330</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10915</th>\n",
       "      <td>-117.87</td>\n",
       "      <td>33.73</td>\n",
       "      <td>45.0</td>\n",
       "      <td>2264.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>1970.0</td>\n",
       "      <td>499.0</td>\n",
       "      <td>3.4193</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19150</th>\n",
       "      <td>-122.70</td>\n",
       "      <td>38.35</td>\n",
       "      <td>14.0</td>\n",
       "      <td>2313.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>954.0</td>\n",
       "      <td>397.0</td>\n",
       "      <td>3.7813</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4186</th>\n",
       "      <td>-118.23</td>\n",
       "      <td>34.13</td>\n",
       "      <td>48.0</td>\n",
       "      <td>1308.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>835.0</td>\n",
       "      <td>294.0</td>\n",
       "      <td>4.2891</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16885</th>\n",
       "      <td>-122.40</td>\n",
       "      <td>37.58</td>\n",
       "      <td>26.0</td>\n",
       "      <td>3281.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>1145.0</td>\n",
       "      <td>480.0</td>\n",
       "      <td>6.3580</td>\n",
       "      <td>NEAR OCEAN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "1606     -122.08     37.88                26.0       2947.0           433.0   \n",
       "10915    -117.87     33.73                45.0       2264.0           433.0   \n",
       "19150    -122.70     38.35                14.0       2313.0           433.0   \n",
       "4186     -118.23     34.13                48.0       1308.0           433.0   \n",
       "16885    -122.40     37.58                26.0       3281.0           433.0   \n",
       "\n",
       "       population  households  median_income ocean_proximity  \n",
       "1606        825.0       626.0         2.9330        NEAR BAY  \n",
       "10915      1970.0       499.0         3.4193       <1H OCEAN  \n",
       "19150       954.0       397.0         3.7813       <1H OCEAN  \n",
       "4186        835.0       294.0         4.2891       <1H OCEAN  \n",
       "16885      1145.0       480.0         6.3580      NEAR OCEAN  "
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_incomplete_rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.688401Z",
     "iopub.status.busy": "2021-11-03T03:33:31.687462Z",
     "iopub.status.idle": "2021-11-03T03:33:31.753378Z",
     "shell.execute_reply": "2021-11-03T03:33:31.752415Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.impute import SimpleImputer\n",
    "imputer = SimpleImputer(strategy=\"median\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "중간값이 수치형 특성에서만 계산될 수 있기 때문에 텍스트 특성을 삭제합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.759421Z",
     "iopub.status.busy": "2021-11-03T03:33:31.758453Z",
     "iopub.status.idle": "2021-11-03T03:33:31.760416Z",
     "shell.execute_reply": "2021-11-03T03:33:31.761074Z"
    }
   },
   "outputs": [],
   "source": [
    "housing_num = housing.drop(\"ocean_proximity\", axis=1)\n",
    "# 다른 방법: housing_num = housing.select_dtypes(include=[np.number])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.767192Z",
     "iopub.status.busy": "2021-11-03T03:33:31.766283Z",
     "iopub.status.idle": "2021-11-03T03:33:31.786509Z",
     "shell.execute_reply": "2021-11-03T03:33:31.787217Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SimpleImputer(strategy='median')"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputer.fit(housing_num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.792575Z",
     "iopub.status.busy": "2021-11-03T03:33:31.791639Z",
     "iopub.status.idle": "2021-11-03T03:33:31.795648Z",
     "shell.execute_reply": "2021-11-03T03:33:31.796355Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-118.51   ,   34.26   ,   29.     , 2119.     ,  433.     ,\n",
       "       1164.     ,  408.     ,    3.54155])"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputer.statistics_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "각 특성의 중간 값이 수동으로 계산한 것과 같은지 확인해 보세요:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.801158Z",
     "iopub.status.busy": "2021-11-03T03:33:31.800224Z",
     "iopub.status.idle": "2021-11-03T03:33:31.808233Z",
     "shell.execute_reply": "2021-11-03T03:33:31.807730Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-118.51   ,   34.26   ,   29.     , 2119.     ,  433.     ,\n",
       "       1164.     ,  408.     ,    3.54155])"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_num.median().values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "훈련 세트를 변환합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.813997Z",
     "iopub.status.busy": "2021-11-03T03:33:31.810555Z",
     "iopub.status.idle": "2021-11-03T03:33:31.815921Z",
     "shell.execute_reply": "2021-11-03T03:33:31.815343Z"
    }
   },
   "outputs": [],
   "source": [
    "X = imputer.transform(housing_num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.819964Z",
     "iopub.status.busy": "2021-11-03T03:33:31.819385Z",
     "iopub.status.idle": "2021-11-03T03:33:31.822555Z",
     "shell.execute_reply": "2021-11-03T03:33:31.822941Z"
    }
   },
   "outputs": [],
   "source": [
    "housing_tr = pd.DataFrame(X, columns=housing_num.columns,\n",
    "                          index=housing_num.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.830482Z",
     "iopub.status.busy": "2021-11-03T03:33:31.829885Z",
     "iopub.status.idle": "2021-11-03T03:33:31.839929Z",
     "shell.execute_reply": "2021-11-03T03:33:31.840363Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1606</th>\n",
       "      <td>-122.08</td>\n",
       "      <td>37.88</td>\n",
       "      <td>26.0</td>\n",
       "      <td>2947.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>825.0</td>\n",
       "      <td>626.0</td>\n",
       "      <td>2.9330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10915</th>\n",
       "      <td>-117.87</td>\n",
       "      <td>33.73</td>\n",
       "      <td>45.0</td>\n",
       "      <td>2264.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>1970.0</td>\n",
       "      <td>499.0</td>\n",
       "      <td>3.4193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19150</th>\n",
       "      <td>-122.70</td>\n",
       "      <td>38.35</td>\n",
       "      <td>14.0</td>\n",
       "      <td>2313.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>954.0</td>\n",
       "      <td>397.0</td>\n",
       "      <td>3.7813</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4186</th>\n",
       "      <td>-118.23</td>\n",
       "      <td>34.13</td>\n",
       "      <td>48.0</td>\n",
       "      <td>1308.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>835.0</td>\n",
       "      <td>294.0</td>\n",
       "      <td>4.2891</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16885</th>\n",
       "      <td>-122.40</td>\n",
       "      <td>37.58</td>\n",
       "      <td>26.0</td>\n",
       "      <td>3281.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>1145.0</td>\n",
       "      <td>480.0</td>\n",
       "      <td>6.3580</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "1606     -122.08     37.88                26.0       2947.0           433.0   \n",
       "10915    -117.87     33.73                45.0       2264.0           433.0   \n",
       "19150    -122.70     38.35                14.0       2313.0           433.0   \n",
       "4186     -118.23     34.13                48.0       1308.0           433.0   \n",
       "16885    -122.40     37.58                26.0       3281.0           433.0   \n",
       "\n",
       "       population  households  median_income  \n",
       "1606        825.0       626.0         2.9330  \n",
       "10915      1970.0       499.0         3.4193  \n",
       "19150       954.0       397.0         3.7813  \n",
       "4186        835.0       294.0         4.2891  \n",
       "16885      1145.0       480.0         6.3580  "
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_tr.loc[sample_incomplete_rows.index.values]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.845009Z",
     "iopub.status.busy": "2021-11-03T03:33:31.844165Z",
     "iopub.status.idle": "2021-11-03T03:33:31.850413Z",
     "shell.execute_reply": "2021-11-03T03:33:31.849855Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'median'"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imputer.strategy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.863144Z",
     "iopub.status.busy": "2021-11-03T03:33:31.862293Z",
     "iopub.status.idle": "2021-11-03T03:33:31.865904Z",
     "shell.execute_reply": "2021-11-03T03:33:31.866362Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12655</th>\n",
       "      <td>-121.46</td>\n",
       "      <td>38.52</td>\n",
       "      <td>29.0</td>\n",
       "      <td>3873.0</td>\n",
       "      <td>797.0</td>\n",
       "      <td>2237.0</td>\n",
       "      <td>706.0</td>\n",
       "      <td>2.1736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15502</th>\n",
       "      <td>-117.23</td>\n",
       "      <td>33.09</td>\n",
       "      <td>7.0</td>\n",
       "      <td>5320.0</td>\n",
       "      <td>855.0</td>\n",
       "      <td>2015.0</td>\n",
       "      <td>768.0</td>\n",
       "      <td>6.3373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2908</th>\n",
       "      <td>-119.04</td>\n",
       "      <td>35.37</td>\n",
       "      <td>44.0</td>\n",
       "      <td>1618.0</td>\n",
       "      <td>310.0</td>\n",
       "      <td>667.0</td>\n",
       "      <td>300.0</td>\n",
       "      <td>2.8750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14053</th>\n",
       "      <td>-117.13</td>\n",
       "      <td>32.75</td>\n",
       "      <td>24.0</td>\n",
       "      <td>1877.0</td>\n",
       "      <td>519.0</td>\n",
       "      <td>898.0</td>\n",
       "      <td>483.0</td>\n",
       "      <td>2.2264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20496</th>\n",
       "      <td>-118.70</td>\n",
       "      <td>34.28</td>\n",
       "      <td>27.0</td>\n",
       "      <td>3536.0</td>\n",
       "      <td>646.0</td>\n",
       "      <td>1837.0</td>\n",
       "      <td>580.0</td>\n",
       "      <td>4.4964</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "12655    -121.46     38.52                29.0       3873.0           797.0   \n",
       "15502    -117.23     33.09                 7.0       5320.0           855.0   \n",
       "2908     -119.04     35.37                44.0       1618.0           310.0   \n",
       "14053    -117.13     32.75                24.0       1877.0           519.0   \n",
       "20496    -118.70     34.28                27.0       3536.0           646.0   \n",
       "\n",
       "       population  households  median_income  \n",
       "12655      2237.0       706.0         2.1736  \n",
       "15502      2015.0       768.0         6.3373  \n",
       "2908        667.0       300.0         2.8750  \n",
       "14053       898.0       483.0         2.2264  \n",
       "20496      1837.0       580.0         4.4964  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_tr.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 텍스트와 범주형 특성 다루기\n",
    "\n",
    "이제 범주형 입력 특성인 `ocean_proximity`을 전처리합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.875408Z",
     "iopub.status.busy": "2021-11-03T03:33:31.874637Z",
     "iopub.status.idle": "2021-11-03T03:33:31.878104Z",
     "shell.execute_reply": "2021-11-03T03:33:31.878532Z"
    }
   },
   "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>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12655</th>\n",
       "      <td>INLAND</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15502</th>\n",
       "      <td>NEAR OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2908</th>\n",
       "      <td>INLAND</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14053</th>\n",
       "      <td>NEAR OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20496</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1481</th>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18125</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5830</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17989</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4861</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      ocean_proximity\n",
       "12655          INLAND\n",
       "15502      NEAR OCEAN\n",
       "2908           INLAND\n",
       "14053      NEAR OCEAN\n",
       "20496       <1H OCEAN\n",
       "1481         NEAR BAY\n",
       "18125       <1H OCEAN\n",
       "5830        <1H OCEAN\n",
       "17989       <1H OCEAN\n",
       "4861        <1H OCEAN"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_cat = housing[[\"ocean_proximity\"]]\n",
    "housing_cat.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.888039Z",
     "iopub.status.busy": "2021-11-03T03:33:31.886530Z",
     "iopub.status.idle": "2021-11-03T03:33:31.891432Z",
     "shell.execute_reply": "2021-11-03T03:33:31.891895Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.],\n",
       "       [4.],\n",
       "       [1.],\n",
       "       [4.],\n",
       "       [0.],\n",
       "       [3.],\n",
       "       [0.],\n",
       "       [0.],\n",
       "       [0.],\n",
       "       [0.]])"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import OrdinalEncoder\n",
    "\n",
    "ordinal_encoder = OrdinalEncoder()\n",
    "housing_cat_encoded = ordinal_encoder.fit_transform(housing_cat)\n",
    "housing_cat_encoded[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.897476Z",
     "iopub.status.busy": "2021-11-03T03:33:31.896792Z",
     "iopub.status.idle": "2021-11-03T03:33:31.900986Z",
     "shell.execute_reply": "2021-11-03T03:33:31.901681Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array(['<1H OCEAN', 'INLAND', 'ISLAND', 'NEAR BAY', 'NEAR OCEAN'],\n",
       "       dtype=object)]"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ordinal_encoder.categories_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.911522Z",
     "iopub.status.busy": "2021-11-03T03:33:31.910737Z",
     "iopub.status.idle": "2021-11-03T03:33:31.913942Z",
     "shell.execute_reply": "2021-11-03T03:33:31.914370Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<16512x5 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 16512 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import OneHotEncoder\n",
    "\n",
    "cat_encoder = OneHotEncoder()\n",
    "housing_cat_1hot = cat_encoder.fit_transform(housing_cat)\n",
    "housing_cat_1hot"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`OneHotEncoder`는 기본적으로 희소 행렬을 반환합니다. 필요하면 `toarray()` 메서드를 사용해 밀집 배열로 변환할 수 있습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.918205Z",
     "iopub.status.busy": "2021-11-03T03:33:31.917570Z",
     "iopub.status.idle": "2021-11-03T03:33:31.924202Z",
     "shell.execute_reply": "2021-11-03T03:33:31.923687Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 1., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 1.],\n",
       "       [0., 1., 0., 0., 0.],\n",
       "       ...,\n",
       "       [1., 0., 0., 0., 0.],\n",
       "       [1., 0., 0., 0., 0.],\n",
       "       [0., 1., 0., 0., 0.]])"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_cat_1hot.toarray()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "또는 `OneHotEncoder`를 만들 때 `sparse=False`로 지정할 수 있습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.932775Z",
     "iopub.status.busy": "2021-11-03T03:33:31.932174Z",
     "iopub.status.idle": "2021-11-03T03:33:31.937164Z",
     "shell.execute_reply": "2021-11-03T03:33:31.936598Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 1., 0., 0., 0.],\n",
       "       [0., 0., 0., 0., 1.],\n",
       "       [0., 1., 0., 0., 0.],\n",
       "       ...,\n",
       "       [1., 0., 0., 0., 0.],\n",
       "       [1., 0., 0., 0., 0.],\n",
       "       [0., 1., 0., 0., 0.]])"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_encoder = OneHotEncoder(sparse=False)\n",
    "housing_cat_1hot = cat_encoder.fit_transform(housing_cat)\n",
    "housing_cat_1hot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.941498Z",
     "iopub.status.busy": "2021-11-03T03:33:31.940851Z",
     "iopub.status.idle": "2021-11-03T03:33:31.944614Z",
     "shell.execute_reply": "2021-11-03T03:33:31.943908Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array(['<1H OCEAN', 'INLAND', 'ISLAND', 'NEAR BAY', 'NEAR OCEAN'],\n",
       "       dtype=object)]"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_encoder.categories_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 나만의 변환기\n",
    "\n",
    "추가 특성을 위해 사용자 정의 변환기를 만들어보죠:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.957988Z",
     "iopub.status.busy": "2021-11-03T03:33:31.951684Z",
     "iopub.status.idle": "2021-11-03T03:33:31.960613Z",
     "shell.execute_reply": "2021-11-03T03:33:31.959984Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "\n",
    "# 열 인덱스\n",
    "rooms_ix, bedrooms_ix, population_ix, households_ix = 3, 4, 5, 6\n",
    "\n",
    "class CombinedAttributesAdder(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, add_bedrooms_per_room=True): # *args 또는 **kargs 없음\n",
    "        self.add_bedrooms_per_room = add_bedrooms_per_room\n",
    "    def fit(self, X, y=None):\n",
    "        return self  # 아무것도 하지 않습니다\n",
    "    def transform(self, X):\n",
    "        rooms_per_household = X[:, rooms_ix] / X[:, households_ix]\n",
    "        population_per_household = X[:, population_ix] / X[:, households_ix]\n",
    "        if self.add_bedrooms_per_room:\n",
    "            bedrooms_per_room = X[:, bedrooms_ix] / X[:, rooms_ix]\n",
    "            return np.c_[X, rooms_per_household, population_per_household,\n",
    "                         bedrooms_per_room]\n",
    "        else:\n",
    "            return np.c_[X, rooms_per_household, population_per_household]\n",
    "\n",
    "attr_adder = CombinedAttributesAdder(add_bedrooms_per_room=False)\n",
    "housing_extra_attribs = attr_adder.transform(housing.to_numpy())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "책에서는 간단하게 인덱스 (3, 4, 5, 6)을 하드코딩했지만 다음처럼 동적으로 처리하는 것이 더 좋습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.965855Z",
     "iopub.status.busy": "2021-11-03T03:33:31.965186Z",
     "iopub.status.idle": "2021-11-03T03:33:31.967934Z",
     "shell.execute_reply": "2021-11-03T03:33:31.967416Z"
    }
   },
   "outputs": [],
   "source": [
    "col_names = \"total_rooms\", \"total_bedrooms\", \"population\", \"households\"\n",
    "rooms_ix, bedrooms_ix, population_ix, households_ix = [\n",
    "    housing.columns.get_loc(c) for c in col_names] # 열 인덱스 구하기"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "또한 `housing_extra_attribs`는 넘파이 배열이기 때문에 열 이름이 없습니다(안타깝지만 사이킷런을 사용할 때 생기는 문제입니다). `DataFrame`으로 복원하려면 다음과 같이 할 수 있습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.973307Z",
     "iopub.status.busy": "2021-11-03T03:33:31.972387Z",
     "iopub.status.idle": "2021-11-03T03:33:31.986470Z",
     "shell.execute_reply": "2021-11-03T03:33:31.986902Z"
    }
   },
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>ocean_proximity</th>\n",
       "      <th>rooms_per_household</th>\n",
       "      <th>population_per_household</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12655</th>\n",
       "      <td>-121.46</td>\n",
       "      <td>38.52</td>\n",
       "      <td>29.0</td>\n",
       "      <td>3873.0</td>\n",
       "      <td>797.0</td>\n",
       "      <td>2237.0</td>\n",
       "      <td>706.0</td>\n",
       "      <td>2.1736</td>\n",
       "      <td>INLAND</td>\n",
       "      <td>5.485836</td>\n",
       "      <td>3.168555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15502</th>\n",
       "      <td>-117.23</td>\n",
       "      <td>33.09</td>\n",
       "      <td>7.0</td>\n",
       "      <td>5320.0</td>\n",
       "      <td>855.0</td>\n",
       "      <td>2015.0</td>\n",
       "      <td>768.0</td>\n",
       "      <td>6.3373</td>\n",
       "      <td>NEAR OCEAN</td>\n",
       "      <td>6.927083</td>\n",
       "      <td>2.623698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2908</th>\n",
       "      <td>-119.04</td>\n",
       "      <td>35.37</td>\n",
       "      <td>44.0</td>\n",
       "      <td>1618.0</td>\n",
       "      <td>310.0</td>\n",
       "      <td>667.0</td>\n",
       "      <td>300.0</td>\n",
       "      <td>2.875</td>\n",
       "      <td>INLAND</td>\n",
       "      <td>5.393333</td>\n",
       "      <td>2.223333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14053</th>\n",
       "      <td>-117.13</td>\n",
       "      <td>32.75</td>\n",
       "      <td>24.0</td>\n",
       "      <td>1877.0</td>\n",
       "      <td>519.0</td>\n",
       "      <td>898.0</td>\n",
       "      <td>483.0</td>\n",
       "      <td>2.2264</td>\n",
       "      <td>NEAR OCEAN</td>\n",
       "      <td>3.886128</td>\n",
       "      <td>1.859213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20496</th>\n",
       "      <td>-118.7</td>\n",
       "      <td>34.28</td>\n",
       "      <td>27.0</td>\n",
       "      <td>3536.0</td>\n",
       "      <td>646.0</td>\n",
       "      <td>1837.0</td>\n",
       "      <td>580.0</td>\n",
       "      <td>4.4964</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "      <td>6.096552</td>\n",
       "      <td>3.167241</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      longitude latitude housing_median_age total_rooms total_bedrooms  \\\n",
       "12655   -121.46    38.52               29.0      3873.0          797.0   \n",
       "15502   -117.23    33.09                7.0      5320.0          855.0   \n",
       "2908    -119.04    35.37               44.0      1618.0          310.0   \n",
       "14053   -117.13    32.75               24.0      1877.0          519.0   \n",
       "20496    -118.7    34.28               27.0      3536.0          646.0   \n",
       "\n",
       "      population households median_income ocean_proximity rooms_per_household  \\\n",
       "12655     2237.0      706.0        2.1736          INLAND            5.485836   \n",
       "15502     2015.0      768.0        6.3373      NEAR OCEAN            6.927083   \n",
       "2908       667.0      300.0         2.875          INLAND            5.393333   \n",
       "14053      898.0      483.0        2.2264      NEAR OCEAN            3.886128   \n",
       "20496     1837.0      580.0        4.4964       <1H OCEAN            6.096552   \n",
       "\n",
       "      population_per_household  \n",
       "12655                 3.168555  \n",
       "15502                 2.623698  \n",
       "2908                  2.223333  \n",
       "14053                 1.859213  \n",
       "20496                 3.167241  "
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_extra_attribs = pd.DataFrame(\n",
    "    housing_extra_attribs,\n",
    "    columns=list(housing.columns)+[\"rooms_per_household\", \"population_per_household\"],\n",
    "    index=housing.index)\n",
    "housing_extra_attribs.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 변환 파이프라인\n",
    "\n",
    "수치형 특성을 전처리하기 위해 파이프라인을 만듭니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:31.992030Z",
     "iopub.status.busy": "2021-11-03T03:33:31.991391Z",
     "iopub.status.idle": "2021-11-03T03:33:32.019129Z",
     "shell.execute_reply": "2021-11-03T03:33:32.018479Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "num_pipeline = Pipeline([\n",
    "        ('imputer', SimpleImputer(strategy=\"median\")),\n",
    "        ('attribs_adder', CombinedAttributesAdder()),\n",
    "        ('std_scaler', StandardScaler()),\n",
    "    ])\n",
    "\n",
    "housing_num_tr = num_pipeline.fit_transform(housing_num)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.024369Z",
     "iopub.status.busy": "2021-11-03T03:33:32.023742Z",
     "iopub.status.idle": "2021-11-03T03:33:32.026880Z",
     "shell.execute_reply": "2021-11-03T03:33:32.027285Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.94135046,  1.34743822,  0.02756357, ...,  0.01739526,\n",
       "         0.00622264, -0.12112176],\n",
       "       [ 1.17178212, -1.19243966, -1.72201763, ...,  0.56925554,\n",
       "        -0.04081077, -0.81086696],\n",
       "       [ 0.26758118, -0.1259716 ,  1.22045984, ..., -0.01802432,\n",
       "        -0.07537122, -0.33827252],\n",
       "       ...,\n",
       "       [-1.5707942 ,  1.31001828,  1.53856552, ..., -0.5092404 ,\n",
       "        -0.03743619,  0.32286937],\n",
       "       [-1.56080303,  1.2492109 , -1.1653327 , ...,  0.32814891,\n",
       "        -0.05915604, -0.45702273],\n",
       "       [-1.28105026,  2.02567448, -0.13148926, ...,  0.01407228,\n",
       "         0.00657083, -0.12169672]])"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_num_tr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.031793Z",
     "iopub.status.busy": "2021-11-03T03:33:32.031145Z",
     "iopub.status.idle": "2021-11-03T03:33:32.067633Z",
     "shell.execute_reply": "2021-11-03T03:33:32.066989Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.compose import ColumnTransformer\n",
    "\n",
    "num_attribs = list(housing_num)\n",
    "cat_attribs = [\"ocean_proximity\"]\n",
    "\n",
    "full_pipeline = ColumnTransformer([\n",
    "        (\"num\", num_pipeline, num_attribs),\n",
    "        (\"cat\", OneHotEncoder(), cat_attribs),\n",
    "    ])\n",
    "\n",
    "housing_prepared = full_pipeline.fit_transform(housing)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.072860Z",
     "iopub.status.busy": "2021-11-03T03:33:32.072222Z",
     "iopub.status.idle": "2021-11-03T03:33:32.074847Z",
     "shell.execute_reply": "2021-11-03T03:33:32.075276Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.94135046,  1.34743822,  0.02756357, ...,  0.        ,\n",
       "         0.        ,  0.        ],\n",
       "       [ 1.17178212, -1.19243966, -1.72201763, ...,  0.        ,\n",
       "         0.        ,  1.        ],\n",
       "       [ 0.26758118, -0.1259716 ,  1.22045984, ...,  0.        ,\n",
       "         0.        ,  0.        ],\n",
       "       ...,\n",
       "       [-1.5707942 ,  1.31001828,  1.53856552, ...,  0.        ,\n",
       "         0.        ,  0.        ],\n",
       "       [-1.56080303,  1.2492109 , -1.1653327 , ...,  0.        ,\n",
       "         0.        ,  0.        ],\n",
       "       [-1.28105026,  2.02567448, -0.13148926, ...,  0.        ,\n",
       "         0.        ,  0.        ]])"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_prepared"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.079759Z",
     "iopub.status.busy": "2021-11-03T03:33:32.079164Z",
     "iopub.status.idle": "2021-11-03T03:33:32.083411Z",
     "shell.execute_reply": "2021-11-03T03:33:32.083967Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16512, 16)"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_prepared.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "다음은 (판다스 `DataFrame` 열의 일부를 선택하기 위해) `DataFrameSelector` 변환기와 `FeatureUnion`를 사용한 예전 방식입니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.089544Z",
     "iopub.status.busy": "2021-11-03T03:33:32.086720Z",
     "iopub.status.idle": "2021-11-03T03:33:32.091377Z",
     "shell.execute_reply": "2021-11-03T03:33:32.091776Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "\n",
    "# 수치형 열과 범주형 열을 선택하기 위한 클래스\n",
    "class OldDataFrameSelector(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, attribute_names):\n",
    "        self.attribute_names = attribute_names\n",
    "    def fit(self, X, y=None):\n",
    "        return self\n",
    "    def transform(self, X):\n",
    "        return X[self.attribute_names].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "하나의 큰 파이프라인에 이들을 모두 결합하여 수치형과 범주형 특성을 전처리합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.098850Z",
     "iopub.status.busy": "2021-11-03T03:33:32.097009Z",
     "iopub.status.idle": "2021-11-03T03:33:32.099570Z",
     "shell.execute_reply": "2021-11-03T03:33:32.100003Z"
    }
   },
   "outputs": [],
   "source": [
    "num_attribs = list(housing_num)\n",
    "cat_attribs = [\"ocean_proximity\"]\n",
    "\n",
    "old_num_pipeline = Pipeline([\n",
    "        ('selector', OldDataFrameSelector(num_attribs)),\n",
    "        ('imputer', SimpleImputer(strategy=\"median\")),\n",
    "        ('attribs_adder', CombinedAttributesAdder()),\n",
    "        ('std_scaler', StandardScaler()),\n",
    "    ])\n",
    "\n",
    "old_cat_pipeline = Pipeline([\n",
    "        ('selector', OldDataFrameSelector(cat_attribs)),\n",
    "        ('cat_encoder', OneHotEncoder(sparse=False)),\n",
    "    ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.104435Z",
     "iopub.status.busy": "2021-11-03T03:33:32.102677Z",
     "iopub.status.idle": "2021-11-03T03:33:32.107119Z",
     "shell.execute_reply": "2021-11-03T03:33:32.107524Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.pipeline import FeatureUnion\n",
    "\n",
    "old_full_pipeline = FeatureUnion(transformer_list=[\n",
    "        (\"num_pipeline\", old_num_pipeline),\n",
    "        (\"cat_pipeline\", old_cat_pipeline),\n",
    "    ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.115114Z",
     "iopub.status.busy": "2021-11-03T03:33:32.114174Z",
     "iopub.status.idle": "2021-11-03T03:33:32.142733Z",
     "shell.execute_reply": "2021-11-03T03:33:32.143170Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.94135046,  1.34743822,  0.02756357, ...,  0.        ,\n",
       "         0.        ,  0.        ],\n",
       "       [ 1.17178212, -1.19243966, -1.72201763, ...,  0.        ,\n",
       "         0.        ,  1.        ],\n",
       "       [ 0.26758118, -0.1259716 ,  1.22045984, ...,  0.        ,\n",
       "         0.        ,  0.        ],\n",
       "       ...,\n",
       "       [-1.5707942 ,  1.31001828,  1.53856552, ...,  0.        ,\n",
       "         0.        ,  0.        ],\n",
       "       [-1.56080303,  1.2492109 , -1.1653327 , ...,  0.        ,\n",
       "         0.        ,  0.        ],\n",
       "       [-1.28105026,  2.02567448, -0.13148926, ...,  0.        ,\n",
       "         0.        ,  0.        ]])"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "old_housing_prepared = old_full_pipeline.fit_transform(housing)\n",
    "old_housing_prepared"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`ColumnTransformer`의 결과와 동일합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.148301Z",
     "iopub.status.busy": "2021-11-03T03:33:32.147201Z",
     "iopub.status.idle": "2021-11-03T03:33:32.153478Z",
     "shell.execute_reply": "2021-11-03T03:33:32.152771Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.allclose(housing_prepared, old_housing_prepared)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 모델 선택과 훈련"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 훈련 세트에서 훈련하고 평가하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.158403Z",
     "iopub.status.busy": "2021-11-03T03:33:32.157450Z",
     "iopub.status.idle": "2021-11-03T03:33:32.180638Z",
     "shell.execute_reply": "2021-11-03T03:33:32.181444Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression()"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "lin_reg = LinearRegression()\n",
    "lin_reg.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.185209Z",
     "iopub.status.busy": "2021-11-03T03:33:32.184276Z",
     "iopub.status.idle": "2021-11-03T03:33:32.193697Z",
     "shell.execute_reply": "2021-11-03T03:33:32.194215Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "예측: [ 85657.90192014 305492.60737488 152056.46122456 186095.70946094\n",
      " 244550.67966089]\n"
     ]
    }
   ],
   "source": [
    "# 훈련 샘플 몇 개를 사용해 전체 파이프라인을 적용해 보겠습니다\n",
    "some_data = housing.iloc[:5]\n",
    "some_labels = housing_labels.iloc[:5]\n",
    "some_data_prepared = full_pipeline.transform(some_data)\n",
    "\n",
    "print(\"예측:\", lin_reg.predict(some_data_prepared))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "실제 값과 비교합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.197767Z",
     "iopub.status.busy": "2021-11-03T03:33:32.196822Z",
     "iopub.status.idle": "2021-11-03T03:33:32.201016Z",
     "shell.execute_reply": "2021-11-03T03:33:32.201552Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "레이블: [72100.0, 279600.0, 82700.0, 112500.0, 238300.0]\n"
     ]
    }
   ],
   "source": [
    "print(\"레이블:\", list(some_labels))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.204846Z",
     "iopub.status.busy": "2021-11-03T03:33:32.203957Z",
     "iopub.status.idle": "2021-11-03T03:33:32.209219Z",
     "shell.execute_reply": "2021-11-03T03:33:32.209756Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.94135046,  1.34743822,  0.02756357,  0.58477745,  0.64037127,\n",
       "         0.73260236,  0.55628602, -0.8936472 ,  0.01739526,  0.00622264,\n",
       "        -0.12112176,  0.        ,  1.        ,  0.        ,  0.        ,\n",
       "         0.        ],\n",
       "       [ 1.17178212, -1.19243966, -1.72201763,  1.26146668,  0.78156132,\n",
       "         0.53361152,  0.72131799,  1.292168  ,  0.56925554, -0.04081077,\n",
       "        -0.81086696,  0.        ,  0.        ,  0.        ,  0.        ,\n",
       "         1.        ],\n",
       "       [ 0.26758118, -0.1259716 ,  1.22045984, -0.46977281, -0.54513828,\n",
       "        -0.67467519, -0.52440722, -0.52543365, -0.01802432, -0.07537122,\n",
       "        -0.33827252,  0.        ,  1.        ,  0.        ,  0.        ,\n",
       "         0.        ],\n",
       "       [ 1.22173797, -1.35147437, -0.37006852, -0.34865152, -0.03636724,\n",
       "        -0.46761716, -0.03729672, -0.86592882, -0.59513997, -0.10680295,\n",
       "         0.96120521,  0.        ,  0.        ,  0.        ,  0.        ,\n",
       "         1.        ],\n",
       "       [ 0.43743108, -0.63581817, -0.13148926,  0.42717947,  0.27279028,\n",
       "         0.37406031,  0.22089846,  0.32575178,  0.2512412 ,  0.00610923,\n",
       "        -0.47451338,  1.        ,  0.        ,  0.        ,  0.        ,\n",
       "         0.        ]])"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "some_data_prepared"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.213065Z",
     "iopub.status.busy": "2021-11-03T03:33:32.212148Z",
     "iopub.status.idle": "2021-11-03T03:33:32.226935Z",
     "shell.execute_reply": "2021-11-03T03:33:32.227655Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "68627.87390018745"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "housing_predictions = lin_reg.predict(housing_prepared)\n",
    "lin_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "lin_rmse = np.sqrt(lin_mse)\n",
    "lin_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**노트**: 사이킷런 0.22 버전부터는 `squared=False` 매개변수로 `mean_squared_error()` 함수를 호출하면 RMSE를 바로 얻을 수 있습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.236093Z",
     "iopub.status.busy": "2021-11-03T03:33:32.233382Z",
     "iopub.status.idle": "2021-11-03T03:33:32.240377Z",
     "shell.execute_reply": "2021-11-03T03:33:32.241118Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "49438.66860915802"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import mean_absolute_error\n",
    "\n",
    "lin_mae = mean_absolute_error(housing_labels, housing_predictions)\n",
    "lin_mae"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.247772Z",
     "iopub.status.busy": "2021-11-03T03:33:32.246724Z",
     "iopub.status.idle": "2021-11-03T03:33:32.579845Z",
     "shell.execute_reply": "2021-11-03T03:33:32.579251Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(random_state=42)"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.tree import DecisionTreeRegressor\n",
    "\n",
    "tree_reg = DecisionTreeRegressor(random_state=42)\n",
    "tree_reg.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.584326Z",
     "iopub.status.busy": "2021-11-03T03:33:32.583709Z",
     "iopub.status.idle": "2021-11-03T03:33:32.594994Z",
     "shell.execute_reply": "2021-11-03T03:33:32.594476Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_predictions = tree_reg.predict(housing_prepared)\n",
    "tree_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "tree_rmse = np.sqrt(tree_mse)\n",
    "tree_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 교차 검증을 사용한 평가"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:32.599985Z",
     "iopub.status.busy": "2021-11-03T03:33:32.599343Z",
     "iopub.status.idle": "2021-11-03T03:33:35.262426Z",
     "shell.execute_reply": "2021-11-03T03:33:35.262873Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "scores = cross_val_score(tree_reg, housing_prepared, housing_labels,\n",
    "                         scoring=\"neg_mean_squared_error\", cv=10)\n",
    "tree_rmse_scores = np.sqrt(-scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:35.268947Z",
     "iopub.status.busy": "2021-11-03T03:33:35.268321Z",
     "iopub.status.idle": "2021-11-03T03:33:35.272864Z",
     "shell.execute_reply": "2021-11-03T03:33:35.272349Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "점수: [72831.45749112 69973.18438322 69528.56551415 72517.78229792\n",
      " 69145.50006909 79094.74123727 68960.045444   73344.50225684\n",
      " 69826.02473916 71077.09753998]\n",
      "평균: 71629.89009727491\n",
      "표준 편차: 2914.035468468928\n"
     ]
    }
   ],
   "source": [
    "def display_scores(scores):\n",
    "    print(\"점수:\", scores)\n",
    "    print(\"평균:\", scores.mean())\n",
    "    print(\"표준 편차:\", scores.std())\n",
    "\n",
    "display_scores(tree_rmse_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:35.279526Z",
     "iopub.status.busy": "2021-11-03T03:33:35.278577Z",
     "iopub.status.idle": "2021-11-03T03:33:35.484527Z",
     "shell.execute_reply": "2021-11-03T03:33:35.485571Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "점수: [71762.76364394 64114.99166359 67771.17124356 68635.19072082\n",
      " 66846.14089488 72528.03725385 73997.08050233 68802.33629334\n",
      " 66443.28836884 70139.79923956]\n",
      "평균: 69104.07998247063\n",
      "표준 편차: 2880.3282098180694\n"
     ]
    }
   ],
   "source": [
    "lin_scores = cross_val_score(lin_reg, housing_prepared, housing_labels,\n",
    "                             scoring=\"neg_mean_squared_error\", cv=10)\n",
    "lin_rmse_scores = np.sqrt(-lin_scores)\n",
    "display_scores(lin_rmse_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**노트**: 사이킷런 0.22 버전에서 `n_estimators`의 기본값이 100으로 바뀌기 때문에 향후를 위해 `n_estimators=100`로 지정합니다(책에는 등장하지 않습니다)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:35.490607Z",
     "iopub.status.busy": "2021-11-03T03:33:35.489134Z",
     "iopub.status.idle": "2021-11-03T03:33:52.173846Z",
     "shell.execute_reply": "2021-11-03T03:33:52.174343Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(random_state=42)"
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "forest_reg = RandomForestRegressor(n_estimators=100, random_state=42)\n",
    "forest_reg.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:52.178006Z",
     "iopub.status.busy": "2021-11-03T03:33:52.177154Z",
     "iopub.status.idle": "2021-11-03T03:33:52.610252Z",
     "shell.execute_reply": "2021-11-03T03:33:52.610733Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "18650.698705770003"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_predictions = forest_reg.predict(housing_prepared)\n",
    "forest_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "forest_rmse = np.sqrt(forest_mse)\n",
    "forest_rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:33:52.616572Z",
     "iopub.status.busy": "2021-11-03T03:33:52.615741Z",
     "iopub.status.idle": "2021-11-03T03:36:17.983159Z",
     "shell.execute_reply": "2021-11-03T03:36:17.983604Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "점수: [51559.63379638 48737.57100062 47210.51269766 51875.21247297\n",
      " 47577.50470123 51863.27467888 52746.34645573 50065.1762751\n",
      " 48664.66818196 54055.90894609]\n",
      "평균: 50435.58092066179\n",
      "표준 편차: 2203.3381412764606\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "forest_scores = cross_val_score(forest_reg, housing_prepared, housing_labels,\n",
    "                                scoring=\"neg_mean_squared_error\", cv=10)\n",
    "forest_rmse_scores = np.sqrt(-forest_scores)\n",
    "display_scores(forest_rmse_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:36:17.988661Z",
     "iopub.status.busy": "2021-11-03T03:36:17.987996Z",
     "iopub.status.idle": "2021-11-03T03:36:18.200577Z",
     "shell.execute_reply": "2021-11-03T03:36:18.201704Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count       10.000000\n",
       "mean     69104.079982\n",
       "std       3036.132517\n",
       "min      64114.991664\n",
       "25%      67077.398482\n",
       "50%      68718.763507\n",
       "75%      71357.022543\n",
       "max      73997.080502\n",
       "dtype: float64"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scores = cross_val_score(lin_reg, housing_prepared, housing_labels, scoring=\"neg_mean_squared_error\", cv=10)\n",
    "pd.Series(np.sqrt(-scores)).describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:36:18.207100Z",
     "iopub.status.busy": "2021-11-03T03:36:18.205458Z",
     "iopub.status.idle": "2021-11-03T03:36:35.411131Z",
     "shell.execute_reply": "2021-11-03T03:36:35.411547Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "111095.06635291968"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import SVR\n",
    "\n",
    "svm_reg = SVR(kernel=\"linear\")\n",
    "svm_reg.fit(housing_prepared, housing_labels)\n",
    "housing_predictions = svm_reg.predict(housing_prepared)\n",
    "svm_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "svm_rmse = np.sqrt(svm_mse)\n",
    "svm_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 모델 세부 튜닝"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 그리드 탐색"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:36:35.415030Z",
     "iopub.status.busy": "2021-11-03T03:36:35.414249Z",
     "iopub.status.idle": "2021-11-03T03:37:29.778818Z",
     "shell.execute_reply": "2021-11-03T03:37:29.779278Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, estimator=RandomForestRegressor(random_state=42),\n",
       "             param_grid=[{'max_features': [2, 4, 6, 8],\n",
       "                          'n_estimators': [3, 10, 30]},\n",
       "                         {'bootstrap': [False], 'max_features': [2, 3, 4],\n",
       "                          'n_estimators': [3, 10]}],\n",
       "             return_train_score=True, scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param_grid = [\n",
    "    # 12(=3×4)개의 하이퍼파라미터 조합을 시도합니다.\n",
    "    {'n_estimators': [3, 10, 30], 'max_features': [2, 4, 6, 8]},\n",
    "    # bootstrap은 False로 하고 6(=2×3)개의 조합을 시도합니다.\n",
    "    {'bootstrap': [False], 'n_estimators': [3, 10], 'max_features': [2, 3, 4]},\n",
    "  ]\n",
    "\n",
    "forest_reg = RandomForestRegressor(random_state=42)\n",
    "# 다섯 개의 폴드로 훈련하면 총 (12+6)*5=90번의 훈련이 일어납니다.\n",
    "grid_search = GridSearchCV(forest_reg, param_grid, cv=5,\n",
    "                           scoring='neg_mean_squared_error',\n",
    "                           return_train_score=True)\n",
    "grid_search.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "최상의 파라미터 조합은 다음과 같습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:37:29.782174Z",
     "iopub.status.busy": "2021-11-03T03:37:29.781584Z",
     "iopub.status.idle": "2021-11-03T03:37:29.785108Z",
     "shell.execute_reply": "2021-11-03T03:37:29.785584Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_features': 8, 'n_estimators': 30}"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:37:29.789607Z",
     "iopub.status.busy": "2021-11-03T03:37:29.788026Z",
     "iopub.status.idle": "2021-11-03T03:37:29.791938Z",
     "shell.execute_reply": "2021-11-03T03:37:29.792328Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(max_features=8, n_estimators=30, random_state=42)"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_search.best_estimator_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "그리드서치에서 테스트한 하이퍼파라미터 조합의 점수를 확인합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:37:29.796640Z",
     "iopub.status.busy": "2021-11-03T03:37:29.796040Z",
     "iopub.status.idle": "2021-11-03T03:37:29.801724Z",
     "shell.execute_reply": "2021-11-03T03:37:29.802133Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "63895.161577951665 {'max_features': 2, 'n_estimators': 3}\n",
      "54916.32386349543 {'max_features': 2, 'n_estimators': 10}\n",
      "52885.86715332332 {'max_features': 2, 'n_estimators': 30}\n",
      "60075.3680329983 {'max_features': 4, 'n_estimators': 3}\n",
      "52495.01284985185 {'max_features': 4, 'n_estimators': 10}\n",
      "50187.24324926565 {'max_features': 4, 'n_estimators': 30}\n",
      "58064.73529982314 {'max_features': 6, 'n_estimators': 3}\n",
      "51519.32062366315 {'max_features': 6, 'n_estimators': 10}\n",
      "49969.80441627874 {'max_features': 6, 'n_estimators': 30}\n",
      "58895.824998155826 {'max_features': 8, 'n_estimators': 3}\n",
      "52459.79624724529 {'max_features': 8, 'n_estimators': 10}\n",
      "49898.98913455217 {'max_features': 8, 'n_estimators': 30}\n",
      "62381.765106921855 {'bootstrap': False, 'max_features': 2, 'n_estimators': 3}\n",
      "54476.57050944266 {'bootstrap': False, 'max_features': 2, 'n_estimators': 10}\n",
      "59974.60028085155 {'bootstrap': False, 'max_features': 3, 'n_estimators': 3}\n",
      "52754.5632813202 {'bootstrap': False, 'max_features': 3, 'n_estimators': 10}\n",
      "57831.136061214274 {'bootstrap': False, 'max_features': 4, 'n_estimators': 3}\n",
      "51278.37877140253 {'bootstrap': False, 'max_features': 4, 'n_estimators': 10}\n"
     ]
    }
   ],
   "source": [
    "cvres = grid_search.cv_results_\n",
    "for mean_score, params in zip(cvres[\"mean_test_score\"], cvres[\"params\"]):\n",
    "    print(np.sqrt(-mean_score), params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:37:29.835776Z",
     "iopub.status.busy": "2021-11-03T03:37:29.835124Z",
     "iopub.status.idle": "2021-11-03T03:37:29.849492Z",
     "shell.execute_reply": "2021-11-03T03:37:29.849936Z"
    }
   },
   "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>mean_fit_time</th>\n",
       "      <th>std_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>param_max_features</th>\n",
       "      <th>param_n_estimators</th>\n",
       "      <th>param_bootstrap</th>\n",
       "      <th>params</th>\n",
       "      <th>split0_test_score</th>\n",
       "      <th>split1_test_score</th>\n",
       "      <th>...</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>rank_test_score</th>\n",
       "      <th>split0_train_score</th>\n",
       "      <th>split1_train_score</th>\n",
       "      <th>split2_train_score</th>\n",
       "      <th>split3_train_score</th>\n",
       "      <th>split4_train_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "      <th>std_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.067164</td>\n",
       "      <td>0.000877</td>\n",
       "      <td>0.004181</td>\n",
       "      <td>0.000047</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 2, 'n_estimators': 3}</td>\n",
       "      <td>-4.119912e+09</td>\n",
       "      <td>-3.723465e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-4.082592e+09</td>\n",
       "      <td>1.867375e+08</td>\n",
       "      <td>18</td>\n",
       "      <td>-1.155630e+09</td>\n",
       "      <td>-1.089726e+09</td>\n",
       "      <td>-1.153843e+09</td>\n",
       "      <td>-1.118149e+09</td>\n",
       "      <td>-1.093446e+09</td>\n",
       "      <td>-1.122159e+09</td>\n",
       "      <td>2.834288e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.219164</td>\n",
       "      <td>0.001392</td>\n",
       "      <td>0.011773</td>\n",
       "      <td>0.000110</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 2, 'n_estimators': 10}</td>\n",
       "      <td>-2.973521e+09</td>\n",
       "      <td>-2.810319e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.015803e+09</td>\n",
       "      <td>1.139808e+08</td>\n",
       "      <td>11</td>\n",
       "      <td>-5.982947e+08</td>\n",
       "      <td>-5.904781e+08</td>\n",
       "      <td>-6.123850e+08</td>\n",
       "      <td>-5.727681e+08</td>\n",
       "      <td>-5.905210e+08</td>\n",
       "      <td>-5.928894e+08</td>\n",
       "      <td>1.284978e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.657425</td>\n",
       "      <td>0.003093</td>\n",
       "      <td>0.033022</td>\n",
       "      <td>0.000531</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 2, 'n_estimators': 30}</td>\n",
       "      <td>-2.801229e+09</td>\n",
       "      <td>-2.671474e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.796915e+09</td>\n",
       "      <td>7.980892e+07</td>\n",
       "      <td>9</td>\n",
       "      <td>-4.412567e+08</td>\n",
       "      <td>-4.326398e+08</td>\n",
       "      <td>-4.553722e+08</td>\n",
       "      <td>-4.320746e+08</td>\n",
       "      <td>-4.311606e+08</td>\n",
       "      <td>-4.385008e+08</td>\n",
       "      <td>9.184397e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.113365</td>\n",
       "      <td>0.001084</td>\n",
       "      <td>0.004352</td>\n",
       "      <td>0.000104</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 4, 'n_estimators': 3}</td>\n",
       "      <td>-3.528743e+09</td>\n",
       "      <td>-3.490303e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.609050e+09</td>\n",
       "      <td>1.375683e+08</td>\n",
       "      <td>16</td>\n",
       "      <td>-9.782368e+08</td>\n",
       "      <td>-9.806455e+08</td>\n",
       "      <td>-1.003780e+09</td>\n",
       "      <td>-1.016515e+09</td>\n",
       "      <td>-1.011270e+09</td>\n",
       "      <td>-9.980896e+08</td>\n",
       "      <td>1.577372e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.364473</td>\n",
       "      <td>0.002683</td>\n",
       "      <td>0.011766</td>\n",
       "      <td>0.000099</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 4, 'n_estimators': 10}</td>\n",
       "      <td>-2.742620e+09</td>\n",
       "      <td>-2.609311e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.755726e+09</td>\n",
       "      <td>1.182604e+08</td>\n",
       "      <td>7</td>\n",
       "      <td>-5.063215e+08</td>\n",
       "      <td>-5.257983e+08</td>\n",
       "      <td>-5.081984e+08</td>\n",
       "      <td>-5.174405e+08</td>\n",
       "      <td>-5.282066e+08</td>\n",
       "      <td>-5.171931e+08</td>\n",
       "      <td>8.882622e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.094907</td>\n",
       "      <td>0.004319</td>\n",
       "      <td>0.032924</td>\n",
       "      <td>0.000188</td>\n",
       "      <td>4</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 4, 'n_estimators': 30}</td>\n",
       "      <td>-2.522176e+09</td>\n",
       "      <td>-2.440241e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.518759e+09</td>\n",
       "      <td>8.488084e+07</td>\n",
       "      <td>3</td>\n",
       "      <td>-3.776568e+08</td>\n",
       "      <td>-3.902106e+08</td>\n",
       "      <td>-3.885042e+08</td>\n",
       "      <td>-3.830866e+08</td>\n",
       "      <td>-3.894779e+08</td>\n",
       "      <td>-3.857872e+08</td>\n",
       "      <td>4.774229e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.149577</td>\n",
       "      <td>0.001430</td>\n",
       "      <td>0.004289</td>\n",
       "      <td>0.000041</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 6, 'n_estimators': 3}</td>\n",
       "      <td>-3.362127e+09</td>\n",
       "      <td>-3.311863e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.371513e+09</td>\n",
       "      <td>1.378086e+08</td>\n",
       "      <td>13</td>\n",
       "      <td>-8.909397e+08</td>\n",
       "      <td>-9.583733e+08</td>\n",
       "      <td>-9.000201e+08</td>\n",
       "      <td>-8.964731e+08</td>\n",
       "      <td>-9.151927e+08</td>\n",
       "      <td>-9.121998e+08</td>\n",
       "      <td>2.444837e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.502295</td>\n",
       "      <td>0.002231</td>\n",
       "      <td>0.011696</td>\n",
       "      <td>0.000079</td>\n",
       "      <td>6</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 6, 'n_estimators': 10}</td>\n",
       "      <td>-2.622099e+09</td>\n",
       "      <td>-2.669655e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.654240e+09</td>\n",
       "      <td>6.967978e+07</td>\n",
       "      <td>5</td>\n",
       "      <td>-4.939906e+08</td>\n",
       "      <td>-5.145996e+08</td>\n",
       "      <td>-5.023512e+08</td>\n",
       "      <td>-4.959467e+08</td>\n",
       "      <td>-5.147087e+08</td>\n",
       "      <td>-5.043194e+08</td>\n",
       "      <td>8.880106e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.533953</td>\n",
       "      <td>0.009125</td>\n",
       "      <td>0.033168</td>\n",
       "      <td>0.000290</td>\n",
       "      <td>6</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 6, 'n_estimators': 30}</td>\n",
       "      <td>-2.446142e+09</td>\n",
       "      <td>-2.446594e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.496981e+09</td>\n",
       "      <td>7.357046e+07</td>\n",
       "      <td>2</td>\n",
       "      <td>-3.760968e+08</td>\n",
       "      <td>-3.876636e+08</td>\n",
       "      <td>-3.875307e+08</td>\n",
       "      <td>-3.760938e+08</td>\n",
       "      <td>-3.861056e+08</td>\n",
       "      <td>-3.826981e+08</td>\n",
       "      <td>5.418747e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.200251</td>\n",
       "      <td>0.004116</td>\n",
       "      <td>0.004347</td>\n",
       "      <td>0.000070</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 8, 'n_estimators': 3}</td>\n",
       "      <td>-3.590333e+09</td>\n",
       "      <td>-3.232664e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.468718e+09</td>\n",
       "      <td>1.293758e+08</td>\n",
       "      <td>14</td>\n",
       "      <td>-9.505012e+08</td>\n",
       "      <td>-9.166119e+08</td>\n",
       "      <td>-9.033910e+08</td>\n",
       "      <td>-9.070642e+08</td>\n",
       "      <td>-9.459386e+08</td>\n",
       "      <td>-9.247014e+08</td>\n",
       "      <td>1.973471e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.659325</td>\n",
       "      <td>0.003863</td>\n",
       "      <td>0.011855</td>\n",
       "      <td>0.000197</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 8, 'n_estimators': 10}</td>\n",
       "      <td>-2.721311e+09</td>\n",
       "      <td>-2.675886e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.752030e+09</td>\n",
       "      <td>6.258030e+07</td>\n",
       "      <td>6</td>\n",
       "      <td>-4.998373e+08</td>\n",
       "      <td>-4.997970e+08</td>\n",
       "      <td>-5.099880e+08</td>\n",
       "      <td>-5.047868e+08</td>\n",
       "      <td>-5.348043e+08</td>\n",
       "      <td>-5.098427e+08</td>\n",
       "      <td>1.303601e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.977234</td>\n",
       "      <td>0.008652</td>\n",
       "      <td>0.032824</td>\n",
       "      <td>0.000151</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 8, 'n_estimators': 30}</td>\n",
       "      <td>-2.492636e+09</td>\n",
       "      <td>-2.444818e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.489909e+09</td>\n",
       "      <td>7.086483e+07</td>\n",
       "      <td>1</td>\n",
       "      <td>-3.801679e+08</td>\n",
       "      <td>-3.832972e+08</td>\n",
       "      <td>-3.823818e+08</td>\n",
       "      <td>-3.778452e+08</td>\n",
       "      <td>-3.817589e+08</td>\n",
       "      <td>-3.810902e+08</td>\n",
       "      <td>1.916605e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.104956</td>\n",
       "      <td>0.001028</td>\n",
       "      <td>0.005049</td>\n",
       "      <td>0.000138</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 2, 'n_est...</td>\n",
       "      <td>-4.020842e+09</td>\n",
       "      <td>-3.951861e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.891485e+09</td>\n",
       "      <td>8.648595e+07</td>\n",
       "      <td>17</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-4.306828e+01</td>\n",
       "      <td>-1.051392e+04</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-2.111398e+03</td>\n",
       "      <td>4.201294e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.348037</td>\n",
       "      <td>0.004065</td>\n",
       "      <td>0.014048</td>\n",
       "      <td>0.000118</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 2, 'n_est...</td>\n",
       "      <td>-2.901352e+09</td>\n",
       "      <td>-3.036875e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.967697e+09</td>\n",
       "      <td>4.582448e+07</td>\n",
       "      <td>10</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-3.876145e+00</td>\n",
       "      <td>-9.462528e+02</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-1.900258e+02</td>\n",
       "      <td>3.781165e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.141391</td>\n",
       "      <td>0.002086</td>\n",
       "      <td>0.005016</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 3, 'n_est...</td>\n",
       "      <td>-3.687132e+09</td>\n",
       "      <td>-3.446245e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.596953e+09</td>\n",
       "      <td>8.011960e+07</td>\n",
       "      <td>15</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.465390</td>\n",
       "      <td>0.003114</td>\n",
       "      <td>0.014140</td>\n",
       "      <td>0.000077</td>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 3, 'n_est...</td>\n",
       "      <td>-2.837028e+09</td>\n",
       "      <td>-2.619558e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.783044e+09</td>\n",
       "      <td>8.862580e+07</td>\n",
       "      <td>8</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.181031</td>\n",
       "      <td>0.001810</td>\n",
       "      <td>0.005012</td>\n",
       "      <td>0.000027</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 4, 'n_est...</td>\n",
       "      <td>-3.549428e+09</td>\n",
       "      <td>-3.318176e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.344440e+09</td>\n",
       "      <td>1.099355e+08</td>\n",
       "      <td>12</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.589645</td>\n",
       "      <td>0.005947</td>\n",
       "      <td>0.014093</td>\n",
       "      <td>0.000140</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 4, 'n_est...</td>\n",
       "      <td>-2.692499e+09</td>\n",
       "      <td>-2.542704e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.629472e+09</td>\n",
       "      <td>8.510266e+07</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>18 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    mean_fit_time  std_fit_time  mean_score_time  std_score_time  \\\n",
       "0        0.067164      0.000877         0.004181        0.000047   \n",
       "1        0.219164      0.001392         0.011773        0.000110   \n",
       "2        0.657425      0.003093         0.033022        0.000531   \n",
       "3        0.113365      0.001084         0.004352        0.000104   \n",
       "4        0.364473      0.002683         0.011766        0.000099   \n",
       "5        1.094907      0.004319         0.032924        0.000188   \n",
       "6        0.149577      0.001430         0.004289        0.000041   \n",
       "7        0.502295      0.002231         0.011696        0.000079   \n",
       "8        1.533953      0.009125         0.033168        0.000290   \n",
       "9        0.200251      0.004116         0.004347        0.000070   \n",
       "10       0.659325      0.003863         0.011855        0.000197   \n",
       "11       1.977234      0.008652         0.032824        0.000151   \n",
       "12       0.104956      0.001028         0.005049        0.000138   \n",
       "13       0.348037      0.004065         0.014048        0.000118   \n",
       "14       0.141391      0.002086         0.005016        0.000045   \n",
       "15       0.465390      0.003114         0.014140        0.000077   \n",
       "16       0.181031      0.001810         0.005012        0.000027   \n",
       "17       0.589645      0.005947         0.014093        0.000140   \n",
       "\n",
       "   param_max_features param_n_estimators param_bootstrap  \\\n",
       "0                   2                  3             NaN   \n",
       "1                   2                 10             NaN   \n",
       "2                   2                 30             NaN   \n",
       "3                   4                  3             NaN   \n",
       "4                   4                 10             NaN   \n",
       "5                   4                 30             NaN   \n",
       "6                   6                  3             NaN   \n",
       "7                   6                 10             NaN   \n",
       "8                   6                 30             NaN   \n",
       "9                   8                  3             NaN   \n",
       "10                  8                 10             NaN   \n",
       "11                  8                 30             NaN   \n",
       "12                  2                  3           False   \n",
       "13                  2                 10           False   \n",
       "14                  3                  3           False   \n",
       "15                  3                 10           False   \n",
       "16                  4                  3           False   \n",
       "17                  4                 10           False   \n",
       "\n",
       "                                               params  split0_test_score  \\\n",
       "0              {'max_features': 2, 'n_estimators': 3}      -4.119912e+09   \n",
       "1             {'max_features': 2, 'n_estimators': 10}      -2.973521e+09   \n",
       "2             {'max_features': 2, 'n_estimators': 30}      -2.801229e+09   \n",
       "3              {'max_features': 4, 'n_estimators': 3}      -3.528743e+09   \n",
       "4             {'max_features': 4, 'n_estimators': 10}      -2.742620e+09   \n",
       "5             {'max_features': 4, 'n_estimators': 30}      -2.522176e+09   \n",
       "6              {'max_features': 6, 'n_estimators': 3}      -3.362127e+09   \n",
       "7             {'max_features': 6, 'n_estimators': 10}      -2.622099e+09   \n",
       "8             {'max_features': 6, 'n_estimators': 30}      -2.446142e+09   \n",
       "9              {'max_features': 8, 'n_estimators': 3}      -3.590333e+09   \n",
       "10            {'max_features': 8, 'n_estimators': 10}      -2.721311e+09   \n",
       "11            {'max_features': 8, 'n_estimators': 30}      -2.492636e+09   \n",
       "12  {'bootstrap': False, 'max_features': 2, 'n_est...      -4.020842e+09   \n",
       "13  {'bootstrap': False, 'max_features': 2, 'n_est...      -2.901352e+09   \n",
       "14  {'bootstrap': False, 'max_features': 3, 'n_est...      -3.687132e+09   \n",
       "15  {'bootstrap': False, 'max_features': 3, 'n_est...      -2.837028e+09   \n",
       "16  {'bootstrap': False, 'max_features': 4, 'n_est...      -3.549428e+09   \n",
       "17  {'bootstrap': False, 'max_features': 4, 'n_est...      -2.692499e+09   \n",
       "\n",
       "    split1_test_score  ...  mean_test_score  std_test_score  rank_test_score  \\\n",
       "0       -3.723465e+09  ...    -4.082592e+09    1.867375e+08               18   \n",
       "1       -2.810319e+09  ...    -3.015803e+09    1.139808e+08               11   \n",
       "2       -2.671474e+09  ...    -2.796915e+09    7.980892e+07                9   \n",
       "3       -3.490303e+09  ...    -3.609050e+09    1.375683e+08               16   \n",
       "4       -2.609311e+09  ...    -2.755726e+09    1.182604e+08                7   \n",
       "5       -2.440241e+09  ...    -2.518759e+09    8.488084e+07                3   \n",
       "6       -3.311863e+09  ...    -3.371513e+09    1.378086e+08               13   \n",
       "7       -2.669655e+09  ...    -2.654240e+09    6.967978e+07                5   \n",
       "8       -2.446594e+09  ...    -2.496981e+09    7.357046e+07                2   \n",
       "9       -3.232664e+09  ...    -3.468718e+09    1.293758e+08               14   \n",
       "10      -2.675886e+09  ...    -2.752030e+09    6.258030e+07                6   \n",
       "11      -2.444818e+09  ...    -2.489909e+09    7.086483e+07                1   \n",
       "12      -3.951861e+09  ...    -3.891485e+09    8.648595e+07               17   \n",
       "13      -3.036875e+09  ...    -2.967697e+09    4.582448e+07               10   \n",
       "14      -3.446245e+09  ...    -3.596953e+09    8.011960e+07               15   \n",
       "15      -2.619558e+09  ...    -2.783044e+09    8.862580e+07                8   \n",
       "16      -3.318176e+09  ...    -3.344440e+09    1.099355e+08               12   \n",
       "17      -2.542704e+09  ...    -2.629472e+09    8.510266e+07                4   \n",
       "\n",
       "    split0_train_score  split1_train_score  split2_train_score  \\\n",
       "0        -1.155630e+09       -1.089726e+09       -1.153843e+09   \n",
       "1        -5.982947e+08       -5.904781e+08       -6.123850e+08   \n",
       "2        -4.412567e+08       -4.326398e+08       -4.553722e+08   \n",
       "3        -9.782368e+08       -9.806455e+08       -1.003780e+09   \n",
       "4        -5.063215e+08       -5.257983e+08       -5.081984e+08   \n",
       "5        -3.776568e+08       -3.902106e+08       -3.885042e+08   \n",
       "6        -8.909397e+08       -9.583733e+08       -9.000201e+08   \n",
       "7        -4.939906e+08       -5.145996e+08       -5.023512e+08   \n",
       "8        -3.760968e+08       -3.876636e+08       -3.875307e+08   \n",
       "9        -9.505012e+08       -9.166119e+08       -9.033910e+08   \n",
       "10       -4.998373e+08       -4.997970e+08       -5.099880e+08   \n",
       "11       -3.801679e+08       -3.832972e+08       -3.823818e+08   \n",
       "12       -0.000000e+00       -4.306828e+01       -1.051392e+04   \n",
       "13       -0.000000e+00       -3.876145e+00       -9.462528e+02   \n",
       "14       -0.000000e+00       -0.000000e+00       -0.000000e+00   \n",
       "15       -0.000000e+00       -0.000000e+00       -0.000000e+00   \n",
       "16       -0.000000e+00       -0.000000e+00       -0.000000e+00   \n",
       "17       -0.000000e+00       -0.000000e+00       -0.000000e+00   \n",
       "\n",
       "    split3_train_score  split4_train_score  mean_train_score  std_train_score  \n",
       "0        -1.118149e+09       -1.093446e+09     -1.122159e+09     2.834288e+07  \n",
       "1        -5.727681e+08       -5.905210e+08     -5.928894e+08     1.284978e+07  \n",
       "2        -4.320746e+08       -4.311606e+08     -4.385008e+08     9.184397e+06  \n",
       "3        -1.016515e+09       -1.011270e+09     -9.980896e+08     1.577372e+07  \n",
       "4        -5.174405e+08       -5.282066e+08     -5.171931e+08     8.882622e+06  \n",
       "5        -3.830866e+08       -3.894779e+08     -3.857872e+08     4.774229e+06  \n",
       "6        -8.964731e+08       -9.151927e+08     -9.121998e+08     2.444837e+07  \n",
       "7        -4.959467e+08       -5.147087e+08     -5.043194e+08     8.880106e+06  \n",
       "8        -3.760938e+08       -3.861056e+08     -3.826981e+08     5.418747e+06  \n",
       "9        -9.070642e+08       -9.459386e+08     -9.247014e+08     1.973471e+07  \n",
       "10       -5.047868e+08       -5.348043e+08     -5.098427e+08     1.303601e+07  \n",
       "11       -3.778452e+08       -3.817589e+08     -3.810902e+08     1.916605e+06  \n",
       "12       -0.000000e+00       -0.000000e+00     -2.111398e+03     4.201294e+03  \n",
       "13       -0.000000e+00       -0.000000e+00     -1.900258e+02     3.781165e+02  \n",
       "14       -0.000000e+00       -0.000000e+00      0.000000e+00     0.000000e+00  \n",
       "15       -0.000000e+00       -0.000000e+00      0.000000e+00     0.000000e+00  \n",
       "16       -0.000000e+00       -0.000000e+00      0.000000e+00     0.000000e+00  \n",
       "17       -0.000000e+00       -0.000000e+00      0.000000e+00     0.000000e+00  \n",
       "\n",
       "[18 rows x 23 columns]"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(grid_search.cv_results_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 랜덤 탐색"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:37:29.852787Z",
     "iopub.status.busy": "2021-11-03T03:37:29.852190Z",
     "iopub.status.idle": "2021-11-03T03:40:42.380554Z",
     "shell.execute_reply": "2021-11-03T03:40:42.379953Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, estimator=RandomForestRegressor(random_state=42),\n",
       "                   param_distributions={'max_features': <scipy.stats._distn_infrastructure.rv_frozen object at 0x7f7071466e48>,\n",
       "                                        'n_estimators': <scipy.stats._distn_infrastructure.rv_frozen object at 0x7f7071e76780>},\n",
       "                   random_state=42, scoring='neg_mean_squared_error')"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from scipy.stats import randint\n",
    "\n",
    "param_distribs = {\n",
    "        'n_estimators': randint(low=1, high=200),\n",
    "        'max_features': randint(low=1, high=8),\n",
    "    }\n",
    "\n",
    "forest_reg = RandomForestRegressor(random_state=42)\n",
    "rnd_search = RandomizedSearchCV(forest_reg, param_distributions=param_distribs,\n",
    "                                n_iter=10, cv=5, scoring='neg_mean_squared_error', random_state=42)\n",
    "rnd_search.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.387876Z",
     "iopub.status.busy": "2021-11-03T03:40:42.386678Z",
     "iopub.status.idle": "2021-11-03T03:40:42.390005Z",
     "shell.execute_reply": "2021-11-03T03:40:42.389475Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "49117.55344336652 {'max_features': 7, 'n_estimators': 180}\n",
      "51450.63202856348 {'max_features': 5, 'n_estimators': 15}\n",
      "50692.53588182537 {'max_features': 3, 'n_estimators': 72}\n",
      "50783.614493515 {'max_features': 5, 'n_estimators': 21}\n",
      "49162.89877456354 {'max_features': 7, 'n_estimators': 122}\n",
      "50655.798471042704 {'max_features': 3, 'n_estimators': 75}\n",
      "50513.856319990606 {'max_features': 3, 'n_estimators': 88}\n",
      "49521.17201976928 {'max_features': 5, 'n_estimators': 100}\n",
      "50302.90440763418 {'max_features': 3, 'n_estimators': 150}\n",
      "65167.02018649492 {'max_features': 5, 'n_estimators': 2}\n"
     ]
    }
   ],
   "source": [
    "cvres = rnd_search.cv_results_\n",
    "for mean_score, params in zip(cvres[\"mean_test_score\"], cvres[\"params\"]):\n",
    "    print(np.sqrt(-mean_score), params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 최상의 모델과 오차 분석"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.393844Z",
     "iopub.status.busy": "2021-11-03T03:40:42.393219Z",
     "iopub.status.idle": "2021-11-03T03:40:42.407846Z",
     "shell.execute_reply": "2021-11-03T03:40:42.408273Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6.96542523e-02, 6.04213840e-02, 4.21882202e-02, 1.52450557e-02,\n",
       "       1.55545295e-02, 1.58491147e-02, 1.49346552e-02, 3.79009225e-01,\n",
       "       5.47789150e-02, 1.07031322e-01, 4.82031213e-02, 6.79266007e-03,\n",
       "       1.65706303e-01, 7.83480660e-05, 1.52473276e-03, 3.02816106e-03])"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_importances = grid_search.best_estimator_.feature_importances_\n",
    "feature_importances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.414163Z",
     "iopub.status.busy": "2021-11-03T03:40:42.413552Z",
     "iopub.status.idle": "2021-11-03T03:40:42.417100Z",
     "shell.execute_reply": "2021-11-03T03:40:42.416553Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0.3790092248170967, 'median_income'),\n",
       " (0.16570630316895876, 'INLAND'),\n",
       " (0.10703132208204354, 'pop_per_hhold'),\n",
       " (0.06965425227942929, 'longitude'),\n",
       " (0.0604213840080722, 'latitude'),\n",
       " (0.054778915018283726, 'rooms_per_hhold'),\n",
       " (0.048203121338269206, 'bedrooms_per_room'),\n",
       " (0.04218822024391753, 'housing_median_age'),\n",
       " (0.015849114744428634, 'population'),\n",
       " (0.015554529490469328, 'total_bedrooms'),\n",
       " (0.01524505568840977, 'total_rooms'),\n",
       " (0.014934655161887776, 'households'),\n",
       " (0.006792660074259966, '<1H OCEAN'),\n",
       " (0.0030281610628962747, 'NEAR OCEAN'),\n",
       " (0.0015247327555504937, 'NEAR BAY'),\n",
       " (7.834806602687504e-05, 'ISLAND')]"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "extra_attribs = [\"rooms_per_hhold\", \"pop_per_hhold\", \"bedrooms_per_room\"]\n",
    "#cat_encoder = cat_pipeline.named_steps[\"cat_encoder\"] # 예전 방식\n",
    "cat_encoder = full_pipeline.named_transformers_[\"cat\"]\n",
    "cat_one_hot_attribs = list(cat_encoder.categories_[0])\n",
    "attributes = num_attribs + extra_attribs + cat_one_hot_attribs\n",
    "sorted(zip(feature_importances, attributes), reverse=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 테스트 세트로 시스템 평가하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.422941Z",
     "iopub.status.busy": "2021-11-03T03:40:42.422046Z",
     "iopub.status.idle": "2021-11-03T03:40:42.474412Z",
     "shell.execute_reply": "2021-11-03T03:40:42.473784Z"
    }
   },
   "outputs": [],
   "source": [
    "final_model = grid_search.best_estimator_\n",
    "\n",
    "X_test = strat_test_set.drop(\"median_house_value\", axis=1)\n",
    "y_test = strat_test_set[\"median_house_value\"].copy()\n",
    "\n",
    "X_test_prepared = full_pipeline.transform(X_test)\n",
    "final_predictions = final_model.predict(X_test_prepared)\n",
    "\n",
    "final_mse = mean_squared_error(y_test, final_predictions)\n",
    "final_rmse = np.sqrt(final_mse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.478652Z",
     "iopub.status.busy": "2021-11-03T03:40:42.478015Z",
     "iopub.status.idle": "2021-11-03T03:40:42.480772Z",
     "shell.execute_reply": "2021-11-03T03:40:42.481159Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "47873.26095812988"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "테스트 RMSE에 대한 95% 신뢰 구간을 계산할 수 있습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.490181Z",
     "iopub.status.busy": "2021-11-03T03:40:42.489570Z",
     "iopub.status.idle": "2021-11-03T03:40:42.492715Z",
     "shell.execute_reply": "2021-11-03T03:40:42.492098Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([45893.36082829, 49774.46796717])"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from scipy import stats\n",
    "\n",
    "confidence = 0.95\n",
    "squared_errors = (final_predictions - y_test) ** 2\n",
    "np.sqrt(stats.t.interval(confidence, len(squared_errors) - 1,\n",
    "                         loc=squared_errors.mean(),\n",
    "                         scale=stats.sem(squared_errors)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "다음과 같이 수동으로 계산할 수도 있습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.499378Z",
     "iopub.status.busy": "2021-11-03T03:40:42.498421Z",
     "iopub.status.idle": "2021-11-03T03:40:42.505922Z",
     "shell.execute_reply": "2021-11-03T03:40:42.506325Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(45893.360828285535, 49774.46796717361)"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m = len(squared_errors)\n",
    "mean = squared_errors.mean()\n",
    "tscore = stats.t.ppf((1 + confidence) / 2, df=m - 1)\n",
    "tmargin = tscore * squared_errors.std(ddof=1) / np.sqrt(m)\n",
    "np.sqrt(mean - tmargin), np.sqrt(mean + tmargin)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "또는 t-점수 대신 z-점수를 사용할 수도 있습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.512135Z",
     "iopub.status.busy": "2021-11-03T03:40:42.511534Z",
     "iopub.status.idle": "2021-11-03T03:40:42.516020Z",
     "shell.execute_reply": "2021-11-03T03:40:42.516442Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(45893.9540110131, 49773.921030650374)"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "zscore = stats.norm.ppf((1 + confidence) / 2)\n",
    "zmargin = zscore * squared_errors.std(ddof=1) / np.sqrt(m)\n",
    "np.sqrt(mean - zmargin), np.sqrt(mean + zmargin)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 추가 내용"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 전처리와 예측을 포함한 전체 파이프라인"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.525738Z",
     "iopub.status.busy": "2021-11-03T03:40:42.525096Z",
     "iopub.status.idle": "2021-11-03T03:40:42.578427Z",
     "shell.execute_reply": "2021-11-03T03:40:42.579210Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 85657.90192014, 305492.60737488, 152056.46122456, 186095.70946094,\n",
       "       244550.67966089])"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_pipeline_with_predictor = Pipeline([\n",
    "        (\"preparation\", full_pipeline),\n",
    "        (\"linear\", LinearRegression())\n",
    "    ])\n",
    "\n",
    "full_pipeline_with_predictor.fit(housing, housing_labels)\n",
    "full_pipeline_with_predictor.predict(some_data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## joblib를 사용한 모델 저장"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.583410Z",
     "iopub.status.busy": "2021-11-03T03:40:42.582331Z",
     "iopub.status.idle": "2021-11-03T03:40:42.584989Z",
     "shell.execute_reply": "2021-11-03T03:40:42.585819Z"
    }
   },
   "outputs": [],
   "source": [
    "my_model = full_pipeline_with_predictor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.589576Z",
     "iopub.status.busy": "2021-11-03T03:40:42.588500Z",
     "iopub.status.idle": "2021-11-03T03:40:42.595991Z",
     "shell.execute_reply": "2021-11-03T03:40:42.596734Z"
    }
   },
   "outputs": [],
   "source": [
    "import joblib\n",
    "joblib.dump(my_model, \"my_model.pkl\") # DIFF\n",
    "#...\n",
    "my_model_loaded = joblib.load(\"my_model.pkl\") # DIFF"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## `RandomizedSearchCV`를 위한 Scipy 분포 함수"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.600677Z",
     "iopub.status.busy": "2021-11-03T03:40:42.599616Z",
     "iopub.status.idle": "2021-11-03T03:40:42.941007Z",
     "shell.execute_reply": "2021-11-03T03:40:42.941507Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAD+CAYAAAA3bQKHAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAAAsTAAALEwEAmpwYAAAXZElEQVR4nO3dfZBddZ3n8ffHRImS9BqWHgZxSUYGFzcsyZRdqzuWiuOzlitlnCrWJ5CRsFgZd8UpxaqAEXBHxtKxFFHjCAHRWWErwCjK1jA+jbPr7DTjJEwvLCuDMDykbDDmgYeA+N0/zmm9XBNyQ9/0bTjvV9Wte+/ve07f7+1O+tO/c849J1WFJKm7njLqBiRJo2UQSFLHGQSS1HEGgSR1nEEgSR1nEEhSxxkEktRxAwVBkrVJJpPsTrKxr/aMJBcmuSfJ9iTf66klyflJ7m1v5ydJT31VkuuT3N/erxrWG5MkDWbhgMvdBZwHvBp4el9tQ/t1ngf8FFjVU1sDnACsBAr4S+BW4HNJngZcDXwSuBA4Dbg6ydFV9dDeGjn00ENr+fLlA7YtSQK4/vrr76mq8T3Vsj+fLE5yHvDsqjq5fX4M8L/bsR17WP5/AhurakP7/A+AU6vqhUleBVzcrltt/XZgTVVdu7ceJiYmanJycuCeJUmQ5PqqmthTbbb7CP4dcBvw4XbT0A1JVvfUVwCbe55vbsdmalvq0Um0pacuSZoDsw2CZwPHAtuBZwFrgUuSPK+tL25rM7YDi9v9BP21mfqS/hdJsqbdRzE5PT09y5YlSb1mGwQPAA8D51XVQ1X1XeDbwKva+i5grGf5MWBXOwvor83Ud/a/SFVtqKqJqpoYH9/jJi5J0uM02yDYsoex3k09UzQ7imesbMdmasf1HkUEHNdTlyTNgUEPH12YZBGwAFiQZFGShcD3gNuBD7bLvAh4GfA/2lUvBc5IckSSZwHvAza2te8AjwDvSXJQkrXt+LeG8L4kSQMadEawjmYz0JnA29rH66rqYeCNwOtotu9/AXhHVd3Urvd54GvADcA/Ate0Y7SHiJ4AvAP4GXAKcMJjHToqSRq+/Tp8dD7w8FFJ2n8H8vBRSdITnEEgSR036CkmnjSWn3nNHsd//NHXz3EnkjQ/OCOQpI4zCCSp4wwCSeo4g0CSOs4gkKSOMwgkqeMMAknqOINAkjrOIJCkjjMIJKnjDAJJ6jiDQJI6ziCQpI4zCCSp4wa9ZvHaJJNJdifZuJdlzk5SSV7RM3ZQkouS7EiyNckZfeu8PMlNSe5P8u0ky2b1biRJ+23QGcFdwHnARXsqJjkK+H3g7r7SeuBoYBnNRe3fn+Q17TqHApuAs4BDgEngq/vXviRptgYKgqraVFVXAffuZZHPAB8A+i88fxJwblVtq6obaS5uf3JbexMwVVVXVNWDNKGxMskx+/UOJEmzMut9BEl+H9hdVd/oG18KHA5s7hneDKxoH6/orVXVfcAtPXVJ0hyY1aUqkywB/ivwyj2UF7f323vGtgNLeurTfev01ntfZw2wBuDII4+cRceSpH6znRGsB75UVT/eQ21Xez/WMzYG7Oypj/FovfVfqqoNVTVRVRPj4+OzaliS9GizDYKXA+9pjwjaCvwr4PIkH6iqbTQ7j1f2LL8SmGofT/XWkhwMHNVTlyTNgUEPH12YZBGwAFiQZFGShTRBcCywqr3dBZxGs/MY4FJgXZKl7U7gU4GNbe1K4Ngkq9uvfTawpapuGsYbkyQNZtAZwTrgAeBM4G3t43VVdW9VbZ25AY8A26pqZrPQh2h2AN8GfBf4WFVdC1BV08Bq4CPANuAFwInDeVuSpEENtLO4qtbT7A/Y13LL+57vBk5pb3ta/jrAw0UlaYQ8xYQkdZxBIEkdZxBIUscZBJLUcQaBJHWcQSBJHTercw09mSw/85o9jv/4o6+f404kaW45I5CkjjMIJKnjDAJJ6jiDQJI6ziCQpI4zCCSp4wwCSeo4g0CSOs4gkKSOMwgkqeMGvWbx2iSTSXYn2dgz/sIkf5nkp0mmk1yR5PCeepKcn+Te9nZ+kvTUVyW5Psn97f2qYb45SdK+DTojuAs4D7iob3wpsAFYDiwDdgIX99TXACcAK4HjgDfQXNyeJE8DrgYua7/OJcDV7bgkaY4MFARVtamqrgLu7Rv/ZlVdUVU7qup+4ALgRT2LnAR8vKruqKo7gY8DJ7e142lOevfJqtpdVZ8CAvzeLN6PJGk/DXsfwUuAqZ7nK4DNPc83t2MztS1VVT31LT11SdIcGNppqJMcB5wNvLFneDGwvef5dmBxu5+gvzZTX7KHr72GZjMTRx555LBaliQxpBlBkt8Gvgn856r6657SLmCs5/kYsKudBfTXZuo7+79+VW2oqomqmhgfHx9Gy5Kk1qyDIMky4Drg3Kr6Ul95imZH8YyV/GrT0RRwXO9RRDQ7lHs3LUmSDrBBDx9dmGQRsABYkGRRO3YE8C3ggqr63B5WvRQ4I8kRSZ4FvA/Y2Na+AzwCvCfJQUnWtuPfevxvR5K0vwbdR7AO+FDP87cBHwYKeA6wPsn6mWJVLW4ffr6t39A+/7N2jKp6KMkJ7dhHgRuBE6rqocfzRiRJj89AQVBV64H1eyl/+DHWK+D97W1P9R8Czx+kB0nSgeEpJiSp4wwCSeo4g0CSOs4gkKSOMwgkqeMMAknqOINAkjrOIJCkjjMIJKnjDAJJ6rihXY/gyWr5mdfscfzHH339HHciSQeGMwJJ6jiDQJI6ziCQpI4zCCSp4wwCSeo4g0CSOm7QaxavTTKZZHeSjX21lye5Kcn9Sb7dXsx+pnZQkouS7EiyNckZg64rSZobg84I7gLOAy7qHUxyKLAJOAs4BJgEvtqzyHrgaGAZ8DLg/UleM+C6kqQ5MFAQVNWmqroKuLev9CZgqqquqKoHaX7xr0xyTFs/CTi3qrZV1Y3AF4CTB1xXkjQHZruPYAWweeZJVd0H3AKsSLIUOLy33j5esa91+18kyZp209Tk9PT0LFuWJPWabRAsBrb3jW0HlrQ1+uoztX2t+yhVtaGqJqpqYnx8fJYtS5J6zTYIdgFjfWNjwM62Rl99pravdSVJc2S2QTAFrJx5kuRg4Ciabf/bgLt76+3jqX2tO8ueJEn7YdDDRxcmWQQsABYkWZRkIXAlcGyS1W39bGBLVd3UrnopsC7J0nYn8KnAxra2r3UlSXNg0BnBOuAB4Ezgbe3jdVU1DawGPgJsA14AnNiz3ododgDfBnwX+FhVXQswwLqSpDkw0PUIqmo9zeGde6pdB+zxkM+q2g2c0t72a11J0tzwFBOS1HFeoexx8splkp4snBFIUscZBJLUcQaBJHWcQSBJHWcQSFLHGQSS1HEGgSR1nEEgSR1nEEhSxxkEktRxBoEkdZxBIEkdZxBIUscZBJLUcQaBJHXcUK5HkGQ5cCHw74HdwH8H/ktV/TzJKuCLwPOAG4E/qKp/aNcL8FHgXe2X+jPgzKqqYfQ1Cl6nQNITzbBmBBcCPwEOB1YBLwXeneRpwNXAZcBS4BLg6nYcYA1wArASOA54A3DakHqSJA1gWEHwW8DlVfVgVW0FrgVWAMfTzDo+WVW7q+pTQIDfa9c7Cfh4Vd1RVXcCHwdOHlJPkqQBDCsIPgmcmOQZSY4AXsuvwmBL36aeLe047f3mntrmntovJVmTZDLJ5PT09JBaliTB8ILgezS/wHcAdwCTwFXAYmB737LbgSXt4/76dmBxu+/gl6pqQ1VNVNXE+Pj4kFqWJMEQgiDJU2j++t8EHAwcSrM/4HxgFzDWt8oYsLN93F8fA3Y9kXcWS9ITzTBmBIcARwIXtPsB7gUuBl4HTAHH9f2Ff1w7Tnu/sqe2sqcmSZoDsw6CqroHuBU4PcnCJM+k2Qm8BfgO8AjwniQHJVnbrvat9v5S4IwkRyR5FvA+YONse5IkDW5Y+wjeBLwGmAZ+BDwMvLeqHqI5PPQdwM+AU4AT2nGAzwNfA24A/hG4ph2TJM2RoXygrP2A2PF7qf0QeP5eagW8v71JkkbAU0xIUscZBJLUcQaBJHWcQSBJHWcQSFLHGQSS1HEGgSR1nEEgSR1nEEhSxxkEktRxBoEkddxQzjWkffOi9pLmK2cEktRxBoEkdZxBIEkdZxBIUscZBJLUcUMLgiQnJrkxyX1Jbkny4nb85UluSnJ/km8nWdazzkFJLkqyI8nWJGcMqx9J0mCGEgRJXgmcD7wTWAK8BPinJIcCm4CzgEOASeCrPauuB44GlgEvA96f5DXD6EmSNJhhzQg+DJxTVT+oql9U1Z1VdSfNRe2nquqKqnqQ5hf/yiTHtOudBJxbVduq6kbgC8DJQ+pJkjSAWQdBkgXABDCe5EdJ7khyQZKnAyuAzTPLVtV9wC3AiiRLgcN76+3jFXt4jTVJJpNMTk9Pz7ZlSVKPYcwIDgOeCrwZeDGwCvgdYB2wGNjet/x2ms1Hi3ue99cepao2VNVEVU2Mj48PoWVJ0oxhBMED7f2nq+ruqroH+ATwOmAXMNa3/Biws63RV5+pSZLmyKyDoKq2AXcA1Tvc3k8BK2cGkxwMHEWz32AbcHdvvX08NdueJEmDG9ZJ5y4G/jDJtcDDwHuBrwNXAh9Lshq4Bjgb2FJVN7XrXQqsSzJJs4npVJojjzrDk9FJGrVhHTV0LvB3wM3AjcAPgY9U1TSwGvgIsA14AXBiz3ofotl5fBvwXeBjVXXtkHqSJA1gKDOCqnoYeHd7669dBxzzays1td3AKe1NkjQCnmJCkjrOIJCkjjMIJKnjDAJJ6jiDQJI6ziCQpI4zCCSp4wwCSeo4g0CSOs4gkKSOMwgkqeOGdfZRDZlnJZU0V5wRSFLHGQSS1HEGgSR1nEEgSR031CBIcnSSB5Nc1jP2liS3JbkvyVVJDumpHZLkyrZ2W5K3DLMfSdK+DXtG8BmaS1YCkGQF8Hng7TTXJL4fuLBv+Yfa2luBz7brSJLmyNCCIMmJwM+Av+oZfivwtar6XlXtAs4C3pRkSZKDaa5nfFZV7aqq7wN/QRMakqQ5MpQgSDIGnAOc0VdaAWyeeVJVt9DMAJ7b3n5eVTf3LL+5XUeSNEeGNSM4F/hiVd3RN74Y2N43th1Y0tZ27KX2KEnWJJlMMjk9PT2kliVJMIQgSLIKeAXwp3so7wLG+sbGgJ37qD1KVW2oqomqmhgfH59ty5KkHsM4xcTxwHLg9iTQ/KW/IMm/Aa4FVs4smOQ5wEHAzcAvgIVJjq6q/9cushKYGkJPT1qeekLSsA0jCDYA/63n+R/RBMPpwG8A/yvJi4G/p9mPsKmqdgIk2QSck+RdwCrgjcDvDqEnSdKAZh0EVXU/zWGhACTZBTxYVdPAdJL/BHwZ+JfAdcA7e1Z/N3AR8BPgXuD0qnJGIElzaOhnH62q9X3PvwJ8ZS/L/hQ4Ydg9SJIG5ykmJKnjDAJJ6jiDQJI6ziCQpI7zUpVPEnv7fAH4GQNJj80ZgSR1nDOCDvDTyJIeizMCSeo4g0CSOs4gkKSOMwgkqeMMAknqOINAkjrOIJCkjjMIJKnjDAJJ6jiDQJI6btanmEhyEHAh8ArgEOAW4INV9c22/nLgM8CRwN8CJ1fVbT3rfhZ4M83lLv+kqj4x2540GE89IQmGc66hhcA/Ay8FbgdeB1ye5N8Cu4BNwLuArwHnAl8FXtiuux44GlgG/Cbw7ST/p6quHUJfepwMCKlbhnHx+vtofqHP+HqSW4Hn01ywfqqqrgBIsh64J8kxVXUTcBLNDGEbsC3JF4CTAYNAkubI0PcRJDkMeC4wBawANs/U2tC4BViRZClweG+9fbxi2D1JkvZuqEGQ5KnAl4FL2r/4FwPb+xbbDixpa/TVZ2r9X3dNkskkk9PT08NsWZI6b2hBkOQpwJeAh4C17fAuYKxv0TFgZ1ujrz5Te5Sq2lBVE1U1MT4+PqyWJUkMKQiSBPgicBiwuqoebktTwMqe5Q4GjqLZb7ANuLu33j6eGkZPkqTBDGtG8FngecAbquqBnvErgWOTrE6yCDgb2NJuNgK4FFiXZGmSY4BTgY1D6kmSNIBhfI5gGXAasBvY2kwOADitqr6cZDVwAXAZzecITuxZ/UM0IXIb8ABwvoeOzl8eVio9OQ3j8NHbgDxG/TrgmL3UdgOntDdJ0gh4iglJ6jiDQJI6bhinmFDHue9AemJzRiBJHeeMQHPOGYQ0vzgjkKSOc0agecOZgjQaBoEOmL39Ypc0v7hpSJI6ziCQpI5z05DmPfcdSAeWMwJJ6jhnBHrCejw7o51FSL/OGYEkdZwzAnWK+xukX2cQSBgQ6jY3DUlSx418RpDkEJoL378KuAf4YFV9ZbRdSY393SG9txmEMw7NZyMPAuAzwEPAYcAq4Jokm6tqaqRdSY+Dp9XQE9FIgyDJwcBq4Niq2gV8P8lfAG8Hzhxlb9JcGFZwOLPQbIx6RvBc4OdVdXPP2GbgpSPqR3pCejLMRIYVZm6G23+jDoLFwI6+se3Akt6BJGuANe3TXUn+7yxe81CafRHziT0NZj72BPOzrydcTzn/wL74Xr7+fPw+wYHpa9neCqMOgl3AWN/YGLCzd6CqNgAbhvGCSSaramIYX2tY7Gkw87EnmJ992dNg5mNPMPd9jfrw0ZuBhUmO7hlbCbijWJLmyEiDoKruAzYB5yQ5OMmLgDcCXxplX5LUJaOeEQC8G3g68BPgz4HTD/Cho0PZxDRk9jSY+dgTzM++7Gkw87EnmOO+UlVz+XqSpHlmPswIJEkjZBBIUsd1JgiSHJLkyiT3JbktyVtG3M/aJJNJdifZOMpeZiQ5KMkX2+/PziT/kOS186Cvy5LcnWRHkpuTvGvUPc1IcnSSB5NcNupeAJJ8p+1nV3ubzWduhibJiUlubP//3ZLkxSPuZ1ff7ZEknx5lT21fy5N8I8m2JFuTXJDkgB/m35kg4NHnNHor8NkkK0bYz13AecBFI+yh30Lgn2k+2f0vgHXA5UmWj7Ip4I+B5VU1BvwH4Lwkzx9xTzM+A/zdqJvos7aqFre3fz3qZpK8EjgfeCfNh0VfAvzTKHvq+f4sBn4TeAC4YpQ9tS6kOXDmcJpzr72U5oCaA6oTQdBzTqOzqmpXVX0fmDmn0UhU1aaqugq4d1Q99Kuq+6pqfVX9uKp+UVVfB24FRvpLt6qmqmr3zNP2dtQIWwKav3KBnwF/NeJW5rsPA+dU1Q/af1d3VtWdo26qx2qaX75/PepGgN8CLq+qB6tqK3AtcMD/YO1EELD3cxqNckYw7yU5jOZ7N/IP+CW5MMn9wE3A3cA3RtzPGHAOcMYo+9iLP05yT5K/SXL8KBtJsgCYAMaT/CjJHe3mjqePsq8+JwGX1vw4hPKTwIlJnpHkCOC1NGFwQHUlCAY6p5F+JclTgS8Dl1TVTaPup6reTfPzejHNhxB3P/YaB9y5wBer6o4R99HvA8BzgCNojkX/WpJRzp4OA54KvJnmZ7cK+B2azY4jl2QZzeaXS0bdS+t7NH+g7gDuACaBqw70i3YlCAY6p5EaSZ5C8+nuh4C1I27nl6rqkXaz3rOB00fVR5JVwCuAPx1VD3tTVX9bVTurandVXQL8DfC6Ebb0QHv/6aq6u6ruAT4x4p56vR34flXdOupG2v9319L8oXMwzYnnltLsXzmguhIEntNoQElCc8W4w4DVVfXwiFvak4WMdh/B8cBy4PYkW4E/AlYn+fsR9rQ3BWRkL161jeYv297NLvNhE8yMdzB/ZgOHAEcCF7RBfi9wMXMQmp0Igvl4TqMkC5MsAhYAC5IsmovDxAbwWeB5wBuq6oF9LXygJfmN9tDDxUkWJHk18B8Z7Q7aDTRBtKq9fQ64Bnj16FqCJM9M8uqZf0tJ3kpzhM4B38a8DxcDf9j+LJcC7wW+PuKeSPK7NJvQ5sPRQrSzpVuB09uf3zNp9l9smYsX78SNJm2vAu4DbgfeMuJ+1vOrI2BmbutH3NOyto8HaTanzdzeOsKexoHv0hydswO4ATh11P+e9vCzvGwe9DFOcyjrzvb79QPglfOgr6fSHBb5M2Ar8Clg0Tzo6/PAl0bdR19Pq4DvANtorkdwOXDYgX5dzzUkSR3XiU1DkqS9MwgkqeMMAknqOINAkjrOIJCkjjMIJKnjDAJJ6jiDQJI6ziCQpI77/36P9D+CwsF7AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from scipy.stats import geom, expon\n",
    "geom_distrib=geom(0.5).rvs(10000, random_state=42)\n",
    "expon_distrib=expon(scale=1).rvs(10000, random_state=42)\n",
    "plt.hist(geom_distrib, bins=50)\n",
    "plt.show()\n",
    "plt.hist(expon_distrib, bins=50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 연습문제 해답"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "질문: 서포트 벡터 머신 회귀(`sklearn.svm.SVR`)를 `kernel=“linear”`(하이퍼파라미터 `C`를 바꿔가며)나 `kernel=“rbf”`(하이퍼파라미터 `C`와 `gamma`를 바꿔가며) 등의 다양한 하이퍼파라미터 설정으로 시도해보세요. 지금은 이 하이퍼파라미터가 무엇을 의미하는지 너무 신경 쓰지 마세요. 최상의 `SVR` 모델은 무엇인가요?\n",
    "\n",
    "**경고**: 사용하는 하드웨어에 따라 다음 셀을 실행하는데 30분 또는 그 이상 걸릴 수 있습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T03:40:42.944739Z",
     "iopub.status.busy": "2021-11-03T03:40:42.944074Z",
     "iopub.status.idle": "2021-11-03T04:38:09.200525Z",
     "shell.execute_reply": "2021-11-03T04:38:09.199877Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 50 candidates, totalling 250 fits\n",
      "[CV] END ..............................C=10.0, kernel=linear; total time=   7.9s\n",
      "[CV] END ..............................C=10.0, kernel=linear; total time=   7.8s\n",
      "[CV] END ..............................C=10.0, kernel=linear; total time=   7.8s\n",
      "[CV] END ..............................C=10.0, kernel=linear; total time=   7.8s\n",
      "[CV] END ..............................C=10.0, kernel=linear; total time=   7.8s\n",
      "[CV] END ..............................C=30.0, kernel=linear; total time=   7.8s\n",
      "[CV] END ..............................C=30.0, kernel=linear; total time=   7.7s\n",
      "[CV] END ..............................C=30.0, kernel=linear; total time=   7.9s\n",
      "[CV] END ..............................C=30.0, kernel=linear; total time=   7.8s\n",
      "[CV] END ..............................C=30.0, kernel=linear; total time=   7.8s\n",
      "[CV] END .............................C=100.0, kernel=linear; total time=   7.9s\n",
      "[CV] END .............................C=100.0, kernel=linear; total time=   7.7s\n",
      "[CV] END .............................C=100.0, kernel=linear; total time=   7.6s\n",
      "[CV] END .............................C=100.0, kernel=linear; total time=   7.8s\n",
      "[CV] END .............................C=100.0, kernel=linear; total time=   7.8s\n",
      "[CV] END .............................C=300.0, kernel=linear; total time=   8.0s\n",
      "[CV] END .............................C=300.0, kernel=linear; total time=   7.7s\n",
      "[CV] END .............................C=300.0, kernel=linear; total time=   7.7s\n",
      "[CV] END .............................C=300.0, kernel=linear; total time=   7.8s\n",
      "[CV] END .............................C=300.0, kernel=linear; total time=   7.9s\n",
      "[CV] END ............................C=1000.0, kernel=linear; total time=   8.0s\n",
      "[CV] END ............................C=1000.0, kernel=linear; total time=   8.0s\n",
      "[CV] END ............................C=1000.0, kernel=linear; total time=   8.1s\n",
      "[CV] END ............................C=1000.0, kernel=linear; total time=   8.0s\n",
      "[CV] END ............................C=1000.0, kernel=linear; total time=   8.1s\n",
      "[CV] END ............................C=3000.0, kernel=linear; total time=   9.1s\n",
      "[CV] END ............................C=3000.0, kernel=linear; total time=   8.7s\n",
      "[CV] END ............................C=3000.0, kernel=linear; total time=   8.7s\n",
      "[CV] END ............................C=3000.0, kernel=linear; total time=   8.7s\n",
      "[CV] END ............................C=3000.0, kernel=linear; total time=   8.7s\n",
      "[CV] END ...........................C=10000.0, kernel=linear; total time=  11.1s\n",
      "[CV] END ...........................C=10000.0, kernel=linear; total time=  11.1s\n",
      "[CV] END ...........................C=10000.0, kernel=linear; total time=  10.8s\n",
      "[CV] END ...........................C=10000.0, kernel=linear; total time=  10.5s\n",
      "[CV] END ...........................C=10000.0, kernel=linear; total time=  10.8s\n",
      "[CV] END ...........................C=30000.0, kernel=linear; total time=  16.9s\n",
      "[CV] END ...........................C=30000.0, kernel=linear; total time=  16.1s\n",
      "[CV] END ...........................C=30000.0, kernel=linear; total time=  15.0s\n",
      "[CV] END ...........................C=30000.0, kernel=linear; total time=  16.5s\n",
      "[CV] END ...........................C=30000.0, kernel=linear; total time=  17.2s\n",
      "[CV] END ......................C=1.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=1.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=1.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=1.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=1.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=1.0, gamma=0.03, kernel=rbf; total time=  13.1s\n",
      "[CV] END ......................C=1.0, gamma=0.03, kernel=rbf; total time=  13.1s\n",
      "[CV] END ......................C=1.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=1.0, gamma=0.03, kernel=rbf; total time=  13.1s\n",
      "[CV] END ......................C=1.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=1.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=1.0, gamma=0.1, kernel=rbf; total time=  14.6s\n",
      "[CV] END .......................C=1.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=1.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=1.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=1.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=1.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=1.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=1.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=1.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=1.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=1.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=1.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=1.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=1.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=1.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END .......................C=1.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END .......................C=1.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END .......................C=1.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END .......................C=1.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END ......................C=3.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=3.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=3.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=3.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=3.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END ......................C=3.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=3.0, gamma=0.03, kernel=rbf; total time=  13.1s\n",
      "[CV] END ......................C=3.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=3.0, gamma=0.03, kernel=rbf; total time=  13.1s\n",
      "[CV] END ......................C=3.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=3.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=3.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=3.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=3.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=3.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END .......................C=3.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=3.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=3.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=3.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=3.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END .......................C=3.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=3.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=3.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=3.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=3.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END .......................C=3.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END .......................C=3.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END .......................C=3.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END .......................C=3.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END .......................C=3.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END .....................C=10.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=10.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=10.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=10.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=10.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=10.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END .....................C=10.0, gamma=0.03, kernel=rbf; total time=  13.1s\n",
      "[CV] END .....................C=10.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END .....................C=10.0, gamma=0.03, kernel=rbf; total time=  13.1s\n",
      "[CV] END .....................C=10.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=10.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END ......................C=10.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END ......................C=10.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END ......................C=10.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END ......................C=10.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END ......................C=10.0, gamma=0.3, kernel=rbf; total time=  13.4s\n",
      "[CV] END ......................C=10.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END ......................C=10.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END ......................C=10.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END ......................C=10.0, gamma=0.3, kernel=rbf; total time=  13.5s\n",
      "[CV] END ......................C=10.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=10.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=10.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=10.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=10.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=10.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END ......................C=10.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END ......................C=10.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END ......................C=10.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END ......................C=10.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END .....................C=30.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=30.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=30.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=30.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=30.0, gamma=0.01, kernel=rbf; total time=  12.8s\n",
      "[CV] END .....................C=30.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END .....................C=30.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END .....................C=30.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END .....................C=30.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END .....................C=30.0, gamma=0.03, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=30.0, gamma=0.1, kernel=rbf; total time=  14.4s\n",
      "[CV] END ......................C=30.0, gamma=0.1, kernel=rbf; total time=  14.5s\n",
      "[CV] END ......................C=30.0, gamma=0.1, kernel=rbf; total time=  14.4s\n",
      "[CV] END ......................C=30.0, gamma=0.1, kernel=rbf; total time=  14.4s\n",
      "[CV] END ......................C=30.0, gamma=0.1, kernel=rbf; total time=  14.4s\n",
      "[CV] END ......................C=30.0, gamma=0.3, kernel=rbf; total time=  13.4s\n",
      "[CV] END ......................C=30.0, gamma=0.3, kernel=rbf; total time=  14.8s\n",
      "[CV] END ......................C=30.0, gamma=0.3, kernel=rbf; total time=  18.3s\n",
      "[CV] END ......................C=30.0, gamma=0.3, kernel=rbf; total time=  17.2s\n",
      "[CV] END ......................C=30.0, gamma=0.3, kernel=rbf; total time=  17.9s\n",
      "[CV] END ......................C=30.0, gamma=1.0, kernel=rbf; total time=  17.3s\n",
      "[CV] END ......................C=30.0, gamma=1.0, kernel=rbf; total time=  15.4s\n",
      "[CV] END ......................C=30.0, gamma=1.0, kernel=rbf; total time=  15.3s\n",
      "[CV] END ......................C=30.0, gamma=1.0, kernel=rbf; total time=  16.2s\n",
      "[CV] END ......................C=30.0, gamma=1.0, kernel=rbf; total time=  13.0s\n",
      "[CV] END ......................C=30.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END ......................C=30.0, gamma=3.0, kernel=rbf; total time=  17.6s\n",
      "[CV] END ......................C=30.0, gamma=3.0, kernel=rbf; total time=  17.7s\n",
      "[CV] END ......................C=30.0, gamma=3.0, kernel=rbf; total time=  19.0s\n",
      "[CV] END ......................C=30.0, gamma=3.0, kernel=rbf; total time=  18.3s\n",
      "[CV] END ....................C=100.0, gamma=0.01, kernel=rbf; total time=  17.0s\n",
      "[CV] END ....................C=100.0, gamma=0.01, kernel=rbf; total time=  16.6s\n",
      "[CV] END ....................C=100.0, gamma=0.01, kernel=rbf; total time=  17.6s\n",
      "[CV] END ....................C=100.0, gamma=0.01, kernel=rbf; total time=  17.7s\n",
      "[CV] END ....................C=100.0, gamma=0.01, kernel=rbf; total time=  17.9s\n",
      "[CV] END ....................C=100.0, gamma=0.03, kernel=rbf; total time=  19.1s\n",
      "[CV] END ....................C=100.0, gamma=0.03, kernel=rbf; total time=  18.5s\n",
      "[CV] END ....................C=100.0, gamma=0.03, kernel=rbf; total time=  18.1s\n",
      "[CV] END ....................C=100.0, gamma=0.03, kernel=rbf; total time=  17.9s\n",
      "[CV] END ....................C=100.0, gamma=0.03, kernel=rbf; total time=  19.1s\n",
      "[CV] END .....................C=100.0, gamma=0.1, kernel=rbf; total time=  19.7s\n",
      "[CV] END .....................C=100.0, gamma=0.1, kernel=rbf; total time=  20.2s\n",
      "[CV] END .....................C=100.0, gamma=0.1, kernel=rbf; total time=  18.9s\n",
      "[CV] END .....................C=100.0, gamma=0.1, kernel=rbf; total time=  18.0s\n",
      "[CV] END .....................C=100.0, gamma=0.1, kernel=rbf; total time=  19.5s\n",
      "[CV] END .....................C=100.0, gamma=0.3, kernel=rbf; total time=  17.3s\n",
      "[CV] END .....................C=100.0, gamma=0.3, kernel=rbf; total time=  17.9s\n",
      "[CV] END .....................C=100.0, gamma=0.3, kernel=rbf; total time=  18.0s\n",
      "[CV] END .....................C=100.0, gamma=0.3, kernel=rbf; total time=  17.4s\n",
      "[CV] END .....................C=100.0, gamma=0.3, kernel=rbf; total time=  18.2s\n",
      "[CV] END .....................C=100.0, gamma=1.0, kernel=rbf; total time=  17.1s\n",
      "[CV] END .....................C=100.0, gamma=1.0, kernel=rbf; total time=  17.9s\n",
      "[CV] END .....................C=100.0, gamma=1.0, kernel=rbf; total time=  17.2s\n",
      "[CV] END .....................C=100.0, gamma=1.0, kernel=rbf; total time=  17.3s\n",
      "[CV] END .....................C=100.0, gamma=1.0, kernel=rbf; total time=  16.8s\n",
      "[CV] END .....................C=100.0, gamma=3.0, kernel=rbf; total time=  18.6s\n",
      "[CV] END .....................C=100.0, gamma=3.0, kernel=rbf; total time=  18.3s\n",
      "[CV] END .....................C=100.0, gamma=3.0, kernel=rbf; total time=  18.6s\n",
      "[CV] END .....................C=100.0, gamma=3.0, kernel=rbf; total time=  18.5s\n",
      "[CV] END .....................C=100.0, gamma=3.0, kernel=rbf; total time=  18.5s\n",
      "[CV] END ....................C=300.0, gamma=0.01, kernel=rbf; total time=  16.4s\n",
      "[CV] END ....................C=300.0, gamma=0.01, kernel=rbf; total time=  16.8s\n",
      "[CV] END ....................C=300.0, gamma=0.01, kernel=rbf; total time=  17.1s\n",
      "[CV] END ....................C=300.0, gamma=0.01, kernel=rbf; total time=  16.8s\n",
      "[CV] END ....................C=300.0, gamma=0.01, kernel=rbf; total time=  16.6s\n",
      "[CV] END ....................C=300.0, gamma=0.03, kernel=rbf; total time=  16.6s\n",
      "[CV] END ....................C=300.0, gamma=0.03, kernel=rbf; total time=  17.0s\n",
      "[CV] END ....................C=300.0, gamma=0.03, kernel=rbf; total time=  12.6s\n",
      "[CV] END ....................C=300.0, gamma=0.03, kernel=rbf; total time=  12.6s\n",
      "[CV] END ....................C=300.0, gamma=0.03, kernel=rbf; total time=  13.1s\n",
      "[CV] END .....................C=300.0, gamma=0.1, kernel=rbf; total time=  18.2s\n",
      "[CV] END .....................C=300.0, gamma=0.1, kernel=rbf; total time=  17.7s\n",
      "[CV] END .....................C=300.0, gamma=0.1, kernel=rbf; total time=  18.3s\n",
      "[CV] END .....................C=300.0, gamma=0.1, kernel=rbf; total time=  18.1s\n",
      "[CV] END .....................C=300.0, gamma=0.1, kernel=rbf; total time=  18.1s\n",
      "[CV] END .....................C=300.0, gamma=0.3, kernel=rbf; total time=  17.6s\n",
      "[CV] END .....................C=300.0, gamma=0.3, kernel=rbf; total time=  15.3s\n",
      "[CV] END .....................C=300.0, gamma=0.3, kernel=rbf; total time=  13.2s\n",
      "[CV] END .....................C=300.0, gamma=0.3, kernel=rbf; total time=  13.1s\n",
      "[CV] END .....................C=300.0, gamma=0.3, kernel=rbf; total time=  13.2s\n",
      "[CV] END .....................C=300.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END .....................C=300.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END .....................C=300.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END .....................C=300.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END .....................C=300.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END .....................C=300.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END .....................C=300.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END .....................C=300.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END .....................C=300.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END .....................C=300.0, gamma=3.0, kernel=rbf; total time=  13.9s\n",
      "[CV] END ...................C=1000.0, gamma=0.01, kernel=rbf; total time=  12.2s\n",
      "[CV] END ...................C=1000.0, gamma=0.01, kernel=rbf; total time=  12.2s\n",
      "[CV] END ...................C=1000.0, gamma=0.01, kernel=rbf; total time=  12.2s\n",
      "[CV] END ...................C=1000.0, gamma=0.01, kernel=rbf; total time=  12.2s\n",
      "[CV] END ...................C=1000.0, gamma=0.01, kernel=rbf; total time=  12.2s\n",
      "[CV] END ...................C=1000.0, gamma=0.03, kernel=rbf; total time=  12.3s\n",
      "[CV] END ...................C=1000.0, gamma=0.03, kernel=rbf; total time=  12.3s\n",
      "[CV] END ...................C=1000.0, gamma=0.03, kernel=rbf; total time=  12.3s\n",
      "[CV] END ...................C=1000.0, gamma=0.03, kernel=rbf; total time=  12.4s\n",
      "[CV] END ...................C=1000.0, gamma=0.03, kernel=rbf; total time=  12.3s\n",
      "[CV] END ....................C=1000.0, gamma=0.1, kernel=rbf; total time=  13.7s\n",
      "[CV] END ....................C=1000.0, gamma=0.1, kernel=rbf; total time=  13.7s\n",
      "[CV] END ....................C=1000.0, gamma=0.1, kernel=rbf; total time=  13.7s\n",
      "[CV] END ....................C=1000.0, gamma=0.1, kernel=rbf; total time=  13.7s\n",
      "[CV] END ....................C=1000.0, gamma=0.1, kernel=rbf; total time=  13.7s\n",
      "[CV] END ....................C=1000.0, gamma=0.3, kernel=rbf; total time=  13.0s\n",
      "[CV] END ....................C=1000.0, gamma=0.3, kernel=rbf; total time=  13.0s\n",
      "[CV] END ....................C=1000.0, gamma=0.3, kernel=rbf; total time=  13.0s\n",
      "[CV] END ....................C=1000.0, gamma=0.3, kernel=rbf; total time=  13.0s\n",
      "[CV] END ....................C=1000.0, gamma=0.3, kernel=rbf; total time=  13.0s\n",
      "[CV] END ....................C=1000.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END ....................C=1000.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END ....................C=1000.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END ....................C=1000.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END ....................C=1000.0, gamma=1.0, kernel=rbf; total time=  12.9s\n",
      "[CV] END ....................C=1000.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END ....................C=1000.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END ....................C=1000.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END ....................C=1000.0, gamma=3.0, kernel=rbf; total time=  14.0s\n",
      "[CV] END ....................C=1000.0, gamma=3.0, kernel=rbf; total time=  14.0s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, estimator=SVR(),\n",
       "             param_grid=[{'C': [10.0, 30.0, 100.0, 300.0, 1000.0, 3000.0,\n",
       "                                10000.0, 30000.0],\n",
       "                          'kernel': ['linear']},\n",
       "                         {'C': [1.0, 3.0, 10.0, 30.0, 100.0, 300.0, 1000.0],\n",
       "                          'gamma': [0.01, 0.03, 0.1, 0.3, 1.0, 3.0],\n",
       "                          'kernel': ['rbf']}],\n",
       "             scoring='neg_mean_squared_error', verbose=2)"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param_grid = [\n",
    "        {'kernel': ['linear'], 'C': [10., 30., 100., 300., 1000., 3000., 10000., 30000.0]},\n",
    "        {'kernel': ['rbf'], 'C': [1.0, 3.0, 10., 30., 100., 300., 1000.0],\n",
    "         'gamma': [0.01, 0.03, 0.1, 0.3, 1.0, 3.0]},\n",
    "    ]\n",
    "\n",
    "svm_reg = SVR()\n",
    "grid_search = GridSearchCV(svm_reg, param_grid, cv=5, scoring='neg_mean_squared_error', verbose=2)\n",
    "grid_search.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "최상 모델의 (5-폴드 교차 검증으로 평가한) 점수는 다음과 같습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T04:38:09.205126Z",
     "iopub.status.busy": "2021-11-03T04:38:09.204525Z",
     "iopub.status.idle": "2021-11-03T04:38:09.207046Z",
     "shell.execute_reply": "2021-11-03T04:38:09.207492Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "70286.61835383571"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "negative_mse = grid_search.best_score_\n",
    "rmse = np.sqrt(-negative_mse)\n",
    "rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`RandomForestRegressor`보다 훨씬 좋지 않네요. 최상의 하이퍼파라미터를 확인해 보겠습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T04:38:09.211320Z",
     "iopub.status.busy": "2021-11-03T04:38:09.209766Z",
     "iopub.status.idle": "2021-11-03T04:38:09.213885Z",
     "shell.execute_reply": "2021-11-03T04:38:09.214295Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 30000.0, 'kernel': 'linear'}"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_search.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "선형 커널이 RBF 커널보다 성능이 나은 것 같습니다. `C`는 테스트한 것 중에 최대값이 선택되었습니다. 따라서 (작은 값들은 지우고) 더 큰 값의 `C`로 그리드서치를 다시 실행해 보아야 합니다. 아마도 더 큰 값의 `C`에서 성능이 높아질 것입니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "질문: `GridSearchCV`를 `RandomizedSearchCV`로 바꿔보세요.\n",
    "\n",
    "**경고**: 사용하는 하드웨어에 따라 다음 셀을 실행하는데 45분 또는 그 이상 걸릴 수 있습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T04:38:09.229776Z",
     "iopub.status.busy": "2021-11-03T04:38:09.229105Z",
     "iopub.status.idle": "2021-11-03T05:54:17.028868Z",
     "shell.execute_reply": "2021-11-03T05:54:17.029317Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 50 candidates, totalling 250 fits\n",
      "[CV] END C=629.782329591372, gamma=3.010121430917521, kernel=linear; total time=   8.0s\n",
      "[CV] END C=629.782329591372, gamma=3.010121430917521, kernel=linear; total time=   7.8s\n",
      "[CV] END C=629.782329591372, gamma=3.010121430917521, kernel=linear; total time=   7.9s\n",
      "[CV] END C=629.782329591372, gamma=3.010121430917521, kernel=linear; total time=   7.9s\n",
      "[CV] END C=629.782329591372, gamma=3.010121430917521, kernel=linear; total time=   7.9s\n",
      "[CV] END C=26290.206464300216, gamma=0.9084469696321253, kernel=rbf; total time=  15.5s\n",
      "[CV] END C=26290.206464300216, gamma=0.9084469696321253, kernel=rbf; total time=  15.6s\n",
      "[CV] END C=26290.206464300216, gamma=0.9084469696321253, kernel=rbf; total time=  16.0s\n",
      "[CV] END C=26290.206464300216, gamma=0.9084469696321253, kernel=rbf; total time=  15.9s\n",
      "[CV] END C=26290.206464300216, gamma=0.9084469696321253, kernel=rbf; total time=  15.6s\n",
      "[CV] END C=84.14107900575871, gamma=0.059838768608680676, kernel=rbf; total time=  13.8s\n",
      "[CV] END C=84.14107900575871, gamma=0.059838768608680676, kernel=rbf; total time=  13.8s\n",
      "[CV] END C=84.14107900575871, gamma=0.059838768608680676, kernel=rbf; total time=  13.7s\n",
      "[CV] END C=84.14107900575871, gamma=0.059838768608680676, kernel=rbf; total time=  13.8s\n",
      "[CV] END C=84.14107900575871, gamma=0.059838768608680676, kernel=rbf; total time=  13.7s\n",
      "[CV] END C=432.37884813148855, gamma=0.15416196746656105, kernel=linear; total time=   7.9s\n",
      "[CV] END C=432.37884813148855, gamma=0.15416196746656105, kernel=linear; total time=   7.7s\n",
      "[CV] END C=432.37884813148855, gamma=0.15416196746656105, kernel=linear; total time=   7.8s\n",
      "[CV] END C=432.37884813148855, gamma=0.15416196746656105, kernel=linear; total time=   7.9s\n",
      "[CV] END C=432.37884813148855, gamma=0.15416196746656105, kernel=linear; total time=   8.0s\n",
      "[CV] END C=24.17508294611391, gamma=3.503557475158312, kernel=rbf; total time=  14.5s\n",
      "[CV] END C=24.17508294611391, gamma=3.503557475158312, kernel=rbf; total time=  14.4s\n",
      "[CV] END C=24.17508294611391, gamma=3.503557475158312, kernel=rbf; total time=  14.5s\n",
      "[CV] END C=24.17508294611391, gamma=3.503557475158312, kernel=rbf; total time=  14.5s\n",
      "[CV] END C=24.17508294611391, gamma=3.503557475158312, kernel=rbf; total time=  14.4s\n",
      "[CV] END C=113564.03940586245, gamma=0.0007790692366582295, kernel=rbf; total time=  12.8s\n",
      "[CV] END C=113564.03940586245, gamma=0.0007790692366582295, kernel=rbf; total time=  12.9s\n",
      "[CV] END C=113564.03940586245, gamma=0.0007790692366582295, kernel=rbf; total time=  12.9s\n",
      "[CV] END C=113564.03940586245, gamma=0.0007790692366582295, kernel=rbf; total time=  12.9s\n",
      "[CV] END C=113564.03940586245, gamma=0.0007790692366582295, kernel=rbf; total time=  12.9s\n",
      "[CV] END C=108.30488238805073, gamma=0.3627537294604771, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=108.30488238805073, gamma=0.3627537294604771, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=108.30488238805073, gamma=0.3627537294604771, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=108.30488238805073, gamma=0.3627537294604771, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=108.30488238805073, gamma=0.3627537294604771, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=21.344953672647435, gamma=0.023332523598323388, kernel=linear; total time=   7.8s\n",
      "[CV] END C=21.344953672647435, gamma=0.023332523598323388, kernel=linear; total time=   7.7s\n",
      "[CV] END C=21.344953672647435, gamma=0.023332523598323388, kernel=linear; total time=   7.9s\n",
      "[CV] END C=21.344953672647435, gamma=0.023332523598323388, kernel=linear; total time=   7.7s\n",
      "[CV] END C=21.344953672647435, gamma=0.023332523598323388, kernel=linear; total time=   7.8s\n",
      "[CV] END C=5603.270317432516, gamma=0.15023452872733867, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=5603.270317432516, gamma=0.15023452872733867, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=5603.270317432516, gamma=0.15023452872733867, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=5603.270317432516, gamma=0.15023452872733867, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=5603.270317432516, gamma=0.15023452872733867, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=157055.10989448498, gamma=0.26497040005002437, kernel=rbf; total time=  29.9s\n",
      "[CV] END C=157055.10989448498, gamma=0.26497040005002437, kernel=rbf; total time=  31.0s\n",
      "[CV] END C=157055.10989448498, gamma=0.26497040005002437, kernel=rbf; total time=  32.4s\n",
      "[CV] END C=157055.10989448498, gamma=0.26497040005002437, kernel=rbf; total time=  29.0s\n",
      "[CV] END C=157055.10989448498, gamma=0.26497040005002437, kernel=rbf; total time=  32.0s\n",
      "[CV] END C=27652.464358739708, gamma=0.2227358621286903, kernel=linear; total time=  18.0s\n",
      "[CV] END C=27652.464358739708, gamma=0.2227358621286903, kernel=linear; total time=  15.8s\n",
      "[CV] END C=27652.464358739708, gamma=0.2227358621286903, kernel=linear; total time=  15.3s\n",
      "[CV] END C=27652.464358739708, gamma=0.2227358621286903, kernel=linear; total time=  16.7s\n",
      "[CV] END C=27652.464358739708, gamma=0.2227358621286903, kernel=linear; total time=  16.5s\n",
      "[CV] END C=171377.39570378003, gamma=0.628789100540856, kernel=linear; total time=  45.8s\n",
      "[CV] END C=171377.39570378003, gamma=0.628789100540856, kernel=linear; total time=  53.5s\n",
      "[CV] END C=171377.39570378003, gamma=0.628789100540856, kernel=linear; total time=  44.3s\n",
      "[CV] END C=171377.39570378003, gamma=0.628789100540856, kernel=linear; total time=  60.0s\n",
      "[CV] END C=171377.39570378003, gamma=0.628789100540856, kernel=linear; total time=  46.8s\n",
      "[CV] END C=5385.293820172355, gamma=0.18696125197741642, kernel=linear; total time=   9.8s\n",
      "[CV] END C=5385.293820172355, gamma=0.18696125197741642, kernel=linear; total time=   9.3s\n",
      "[CV] END C=5385.293820172355, gamma=0.18696125197741642, kernel=linear; total time=   9.4s\n",
      "[CV] END C=5385.293820172355, gamma=0.18696125197741642, kernel=linear; total time=   9.3s\n",
      "[CV] END C=5385.293820172355, gamma=0.18696125197741642, kernel=linear; total time=   9.3s\n",
      "[CV] END C=22.59903216621323, gamma=2.850796878935603, kernel=rbf; total time=  13.8s\n",
      "[CV] END C=22.59903216621323, gamma=2.850796878935603, kernel=rbf; total time=  13.9s\n",
      "[CV] END C=22.59903216621323, gamma=2.850796878935603, kernel=rbf; total time=  13.9s\n",
      "[CV] END C=22.59903216621323, gamma=2.850796878935603, kernel=rbf; total time=  13.9s\n",
      "[CV] END C=22.59903216621323, gamma=2.850796878935603, kernel=rbf; total time=  13.9s\n",
      "[CV] END C=34246.75194632794, gamma=0.3632878599687583, kernel=linear; total time=  19.6s\n",
      "[CV] END C=34246.75194632794, gamma=0.3632878599687583, kernel=linear; total time=  19.6s\n",
      "[CV] END C=34246.75194632794, gamma=0.3632878599687583, kernel=linear; total time=  16.0s\n",
      "[CV] END C=34246.75194632794, gamma=0.3632878599687583, kernel=linear; total time=  17.3s\n",
      "[CV] END C=34246.75194632794, gamma=0.3632878599687583, kernel=linear; total time=  19.5s\n",
      "[CV] END C=167.7278956080511, gamma=0.2757870542258224, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=167.7278956080511, gamma=0.2757870542258224, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=167.7278956080511, gamma=0.2757870542258224, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=167.7278956080511, gamma=0.2757870542258224, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=167.7278956080511, gamma=0.2757870542258224, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=61.54360542501371, gamma=0.6835472281341501, kernel=linear; total time=   7.8s\n",
      "[CV] END C=61.54360542501371, gamma=0.6835472281341501, kernel=linear; total time=   7.6s\n",
      "[CV] END C=61.54360542501371, gamma=0.6835472281341501, kernel=linear; total time=   7.6s\n",
      "[CV] END C=61.54360542501371, gamma=0.6835472281341501, kernel=linear; total time=   7.7s\n",
      "[CV] END C=61.54360542501371, gamma=0.6835472281341501, kernel=linear; total time=   7.7s\n",
      "[CV] END C=98.73897389920914, gamma=0.4960365360493639, kernel=rbf; total time=  13.1s\n",
      "[CV] END C=98.73897389920914, gamma=0.4960365360493639, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=98.73897389920914, gamma=0.4960365360493639, kernel=rbf; total time=  13.1s\n",
      "[CV] END C=98.73897389920914, gamma=0.4960365360493639, kernel=rbf; total time=  13.1s\n",
      "[CV] END C=98.73897389920914, gamma=0.4960365360493639, kernel=rbf; total time=  13.1s\n",
      "[CV] END C=8935.505635947808, gamma=0.37354658165762367, kernel=rbf; total time=  13.1s\n",
      "[CV] END C=8935.505635947808, gamma=0.37354658165762367, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=8935.505635947808, gamma=0.37354658165762367, kernel=rbf; total time=  13.1s\n",
      "[CV] END C=8935.505635947808, gamma=0.37354658165762367, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=8935.505635947808, gamma=0.37354658165762367, kernel=rbf; total time=  13.1s\n",
      "[CV] END C=135.76775824842434, gamma=0.838636245624803, kernel=linear; total time=   7.8s\n",
      "[CV] END C=135.76775824842434, gamma=0.838636245624803, kernel=linear; total time=   7.6s\n",
      "[CV] END C=135.76775824842434, gamma=0.838636245624803, kernel=linear; total time=   7.6s\n",
      "[CV] END C=135.76775824842434, gamma=0.838636245624803, kernel=linear; total time=   7.8s\n",
      "[CV] END C=135.76775824842434, gamma=0.838636245624803, kernel=linear; total time=   7.8s\n",
      "[CV] END C=151136.20282548846, gamma=1.4922453771381408, kernel=rbf; total time= 2.5min\n",
      "[CV] END C=151136.20282548846, gamma=1.4922453771381408, kernel=rbf; total time= 2.4min\n",
      "[CV] END C=151136.20282548846, gamma=1.4922453771381408, kernel=rbf; total time= 2.6min\n",
      "[CV] END C=151136.20282548846, gamma=1.4922453771381408, kernel=rbf; total time= 2.3min\n",
      "[CV] END C=151136.20282548846, gamma=1.4922453771381408, kernel=rbf; total time= 2.3min\n",
      "[CV] END C=761.4316758498783, gamma=2.6126336514161914, kernel=linear; total time=   8.0s\n",
      "[CV] END C=761.4316758498783, gamma=2.6126336514161914, kernel=linear; total time=   7.9s\n",
      "[CV] END C=761.4316758498783, gamma=2.6126336514161914, kernel=linear; total time=   7.8s\n",
      "[CV] END C=761.4316758498783, gamma=2.6126336514161914, kernel=linear; total time=   7.9s\n",
      "[CV] END C=761.4316758498783, gamma=2.6126336514161914, kernel=linear; total time=   8.1s\n",
      "[CV] END C=97392.81883041795, gamma=0.09265545895311562, kernel=linear; total time=  35.4s\n",
      "[CV] END C=97392.81883041795, gamma=0.09265545895311562, kernel=linear; total time=  36.0s\n",
      "[CV] END C=97392.81883041795, gamma=0.09265545895311562, kernel=linear; total time=  30.2s\n",
      "[CV] END C=97392.81883041795, gamma=0.09265545895311562, kernel=linear; total time=  47.9s\n",
      "[CV] END C=97392.81883041795, gamma=0.09265545895311562, kernel=linear; total time=  37.4s\n",
      "[CV] END C=2423.0759984939164, gamma=3.248614270240346, kernel=linear; total time=   8.5s\n",
      "[CV] END C=2423.0759984939164, gamma=3.248614270240346, kernel=linear; total time=   8.5s\n",
      "[CV] END C=2423.0759984939164, gamma=3.248614270240346, kernel=linear; total time=   8.3s\n",
      "[CV] END C=2423.0759984939164, gamma=3.248614270240346, kernel=linear; total time=   8.7s\n",
      "[CV] END C=2423.0759984939164, gamma=3.248614270240346, kernel=linear; total time=   8.6s\n",
      "[CV] END C=717.3632997255095, gamma=0.3165604432088257, kernel=linear; total time=   8.0s\n",
      "[CV] END C=717.3632997255095, gamma=0.3165604432088257, kernel=linear; total time=   7.9s\n",
      "[CV] END C=717.3632997255095, gamma=0.3165604432088257, kernel=linear; total time=   7.9s\n",
      "[CV] END C=717.3632997255095, gamma=0.3165604432088257, kernel=linear; total time=   7.9s\n",
      "[CV] END C=717.3632997255095, gamma=0.3165604432088257, kernel=linear; total time=   7.9s\n",
      "[CV] END C=4446.667521184072, gamma=3.3597284456608496, kernel=rbf; total time=  14.4s\n",
      "[CV] END C=4446.667521184072, gamma=3.3597284456608496, kernel=rbf; total time=  14.5s\n",
      "[CV] END C=4446.667521184072, gamma=3.3597284456608496, kernel=rbf; total time=  14.5s\n",
      "[CV] END C=4446.667521184072, gamma=3.3597284456608496, kernel=rbf; total time=  14.5s\n",
      "[CV] END C=4446.667521184072, gamma=3.3597284456608496, kernel=rbf; total time=  14.5s\n",
      "[CV] END C=2963.564121207815, gamma=0.15189814782062885, kernel=linear; total time=   8.5s\n",
      "[CV] END C=2963.564121207815, gamma=0.15189814782062885, kernel=linear; total time=   9.0s\n",
      "[CV] END C=2963.564121207815, gamma=0.15189814782062885, kernel=linear; total time=   8.8s\n",
      "[CV] END C=2963.564121207815, gamma=0.15189814782062885, kernel=linear; total time=   8.6s\n",
      "[CV] END C=2963.564121207815, gamma=0.15189814782062885, kernel=linear; total time=   8.5s\n",
      "[CV] END C=91.64267381686706, gamma=0.01575994483585621, kernel=linear; total time=   7.8s\n",
      "[CV] END C=91.64267381686706, gamma=0.01575994483585621, kernel=linear; total time=   7.6s\n",
      "[CV] END C=91.64267381686706, gamma=0.01575994483585621, kernel=linear; total time=   7.6s\n",
      "[CV] END C=91.64267381686706, gamma=0.01575994483585621, kernel=linear; total time=   7.7s\n",
      "[CV] END C=91.64267381686706, gamma=0.01575994483585621, kernel=linear; total time=   7.7s\n",
      "[CV] END C=24547.601975705915, gamma=0.22153944050588595, kernel=rbf; total time=  13.7s\n",
      "[CV] END C=24547.601975705915, gamma=0.22153944050588595, kernel=rbf; total time=  13.7s\n",
      "[CV] END C=24547.601975705915, gamma=0.22153944050588595, kernel=rbf; total time=  13.7s\n",
      "[CV] END C=24547.601975705915, gamma=0.22153944050588595, kernel=rbf; total time=  13.7s\n",
      "[CV] END C=24547.601975705915, gamma=0.22153944050588595, kernel=rbf; total time=  13.8s\n",
      "[CV] END C=22.76927941060928, gamma=0.22169760231351215, kernel=rbf; total time=  13.6s\n",
      "[CV] END C=22.76927941060928, gamma=0.22169760231351215, kernel=rbf; total time=  13.7s\n",
      "[CV] END C=22.76927941060928, gamma=0.22169760231351215, kernel=rbf; total time=  13.6s\n",
      "[CV] END C=22.76927941060928, gamma=0.22169760231351215, kernel=rbf; total time=  13.7s\n",
      "[CV] END C=22.76927941060928, gamma=0.22169760231351215, kernel=rbf; total time=  13.6s\n",
      "[CV] END C=16483.850529752886, gamma=1.4752145260435134, kernel=linear; total time=  13.2s\n",
      "[CV] END C=16483.850529752886, gamma=1.4752145260435134, kernel=linear; total time=  12.1s\n",
      "[CV] END C=16483.850529752886, gamma=1.4752145260435134, kernel=linear; total time=  12.0s\n",
      "[CV] END C=16483.850529752886, gamma=1.4752145260435134, kernel=linear; total time=  13.7s\n",
      "[CV] END C=16483.850529752886, gamma=1.4752145260435134, kernel=linear; total time=  12.6s\n",
      "[CV] END C=101445.66881340064, gamma=1.052904084582266, kernel=rbf; total time=  58.8s\n",
      "[CV] END C=101445.66881340064, gamma=1.052904084582266, kernel=rbf; total time= 1.1min\n",
      "[CV] END C=101445.66881340064, gamma=1.052904084582266, kernel=rbf; total time=  54.6s\n",
      "[CV] END C=101445.66881340064, gamma=1.052904084582266, kernel=rbf; total time= 1.0min\n",
      "[CV] END C=101445.66881340064, gamma=1.052904084582266, kernel=rbf; total time= 1.2min\n",
      "[CV] END C=56681.80859029545, gamma=0.9763011917123741, kernel=rbf; total time=  23.9s\n",
      "[CV] END C=56681.80859029545, gamma=0.9763011917123741, kernel=rbf; total time=  25.0s\n",
      "[CV] END C=56681.80859029545, gamma=0.9763011917123741, kernel=rbf; total time=  24.0s\n",
      "[CV] END C=56681.80859029545, gamma=0.9763011917123741, kernel=rbf; total time=  28.1s\n",
      "[CV] END C=56681.80859029545, gamma=0.9763011917123741, kernel=rbf; total time=  25.4s\n",
      "[CV] END C=48.15822390928914, gamma=0.4633351167983427, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=48.15822390928914, gamma=0.4633351167983427, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=48.15822390928914, gamma=0.4633351167983427, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=48.15822390928914, gamma=0.4633351167983427, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=48.15822390928914, gamma=0.4633351167983427, kernel=rbf; total time=  13.2s\n",
      "[CV] END C=399.7268155705774, gamma=1.3078757839577408, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=399.7268155705774, gamma=1.3078757839577408, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=399.7268155705774, gamma=1.3078757839577408, kernel=rbf; total time=  12.9s\n",
      "[CV] END C=399.7268155705774, gamma=1.3078757839577408, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=399.7268155705774, gamma=1.3078757839577408, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=251.14073886281363, gamma=0.8238105204914145, kernel=linear; total time=   7.9s\n",
      "[CV] END C=251.14073886281363, gamma=0.8238105204914145, kernel=linear; total time=   7.7s\n",
      "[CV] END C=251.14073886281363, gamma=0.8238105204914145, kernel=linear; total time=   7.7s\n",
      "[CV] END C=251.14073886281363, gamma=0.8238105204914145, kernel=linear; total time=   7.8s\n",
      "[CV] END C=251.14073886281363, gamma=0.8238105204914145, kernel=linear; total time=   7.8s\n",
      "[CV] END C=60.17373642891687, gamma=1.2491263443165994, kernel=linear; total time=   7.8s\n",
      "[CV] END C=60.17373642891687, gamma=1.2491263443165994, kernel=linear; total time=   7.6s\n",
      "[CV] END C=60.17373642891687, gamma=1.2491263443165994, kernel=linear; total time=   7.6s\n",
      "[CV] END C=60.17373642891687, gamma=1.2491263443165994, kernel=linear; total time=   7.7s\n",
      "[CV] END C=60.17373642891687, gamma=1.2491263443165994, kernel=linear; total time=   7.8s\n",
      "[CV] END C=15415.161544891856, gamma=0.2691677514619319, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=15415.161544891856, gamma=0.2691677514619319, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=15415.161544891856, gamma=0.2691677514619319, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=15415.161544891856, gamma=0.2691677514619319, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=15415.161544891856, gamma=0.2691677514619319, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=1888.9148509967113, gamma=0.739678838777267, kernel=linear; total time=   8.3s\n",
      "[CV] END C=1888.9148509967113, gamma=0.739678838777267, kernel=linear; total time=   8.3s\n",
      "[CV] END C=1888.9148509967113, gamma=0.739678838777267, kernel=linear; total time=   8.2s\n",
      "[CV] END C=1888.9148509967113, gamma=0.739678838777267, kernel=linear; total time=   8.2s\n",
      "[CV] END C=1888.9148509967113, gamma=0.739678838777267, kernel=linear; total time=   8.2s\n",
      "[CV] END C=55.53838911232773, gamma=0.578634378499143, kernel=linear; total time=   7.8s\n",
      "[CV] END C=55.53838911232773, gamma=0.578634378499143, kernel=linear; total time=   7.7s\n",
      "[CV] END C=55.53838911232773, gamma=0.578634378499143, kernel=linear; total time=   7.6s\n",
      "[CV] END C=55.53838911232773, gamma=0.578634378499143, kernel=linear; total time=   7.8s\n",
      "[CV] END C=55.53838911232773, gamma=0.578634378499143, kernel=linear; total time=   7.7s\n",
      "[CV] END C=26.714480823948186, gamma=1.0117295509275495, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=26.714480823948186, gamma=1.0117295509275495, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=26.714480823948186, gamma=1.0117295509275495, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=26.714480823948186, gamma=1.0117295509275495, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=26.714480823948186, gamma=1.0117295509275495, kernel=rbf; total time=  13.0s\n",
      "[CV] END C=3582.0552780489566, gamma=1.1891370222133257, kernel=linear; total time=   8.9s\n",
      "[CV] END C=3582.0552780489566, gamma=1.1891370222133257, kernel=linear; total time=   9.0s\n",
      "[CV] END C=3582.0552780489566, gamma=1.1891370222133257, kernel=linear; total time=   8.6s\n",
      "[CV] END C=3582.0552780489566, gamma=1.1891370222133257, kernel=linear; total time=   9.8s\n",
      "[CV] END C=3582.0552780489566, gamma=1.1891370222133257, kernel=linear; total time=   9.0s\n",
      "[CV] END C=198.7004781812736, gamma=0.5282819748826726, kernel=linear; total time=   7.8s\n",
      "[CV] END C=198.7004781812736, gamma=0.5282819748826726, kernel=linear; total time=   7.6s\n",
      "[CV] END C=198.7004781812736, gamma=0.5282819748826726, kernel=linear; total time=   7.7s\n",
      "[CV] END C=198.7004781812736, gamma=0.5282819748826726, kernel=linear; total time=   7.8s\n",
      "[CV] END C=198.7004781812736, gamma=0.5282819748826726, kernel=linear; total time=   7.8s\n",
      "[CV] END C=129.8000604143307, gamma=2.8621383676481322, kernel=linear; total time=   7.8s\n",
      "[CV] END C=129.8000604143307, gamma=2.8621383676481322, kernel=linear; total time=   7.6s\n",
      "[CV] END C=129.8000604143307, gamma=2.8621383676481322, kernel=linear; total time=   7.6s\n",
      "[CV] END C=129.8000604143307, gamma=2.8621383676481322, kernel=linear; total time=   7.7s\n",
      "[CV] END C=129.8000604143307, gamma=2.8621383676481322, kernel=linear; total time=   7.8s\n",
      "[CV] END C=288.4269299593897, gamma=0.17580835850006285, kernel=rbf; total time=  13.4s\n",
      "[CV] END C=288.4269299593897, gamma=0.17580835850006285, kernel=rbf; total time=  13.5s\n",
      "[CV] END C=288.4269299593897, gamma=0.17580835850006285, kernel=rbf; total time=  13.5s\n",
      "[CV] END C=288.4269299593897, gamma=0.17580835850006285, kernel=rbf; total time=  13.5s\n",
      "[CV] END C=288.4269299593897, gamma=0.17580835850006285, kernel=rbf; total time=  13.5s\n",
      "[CV] END C=6287.039489427172, gamma=0.3504567255332862, kernel=linear; total time=   9.7s\n",
      "[CV] END C=6287.039489427172, gamma=0.3504567255332862, kernel=linear; total time=   9.8s\n",
      "[CV] END C=6287.039489427172, gamma=0.3504567255332862, kernel=linear; total time=   9.2s\n",
      "[CV] END C=6287.039489427172, gamma=0.3504567255332862, kernel=linear; total time=   9.6s\n",
      "[CV] END C=6287.039489427172, gamma=0.3504567255332862, kernel=linear; total time=   9.8s\n",
      "[CV] END C=61217.04421344494, gamma=1.6279689407405564, kernel=rbf; total time=  45.0s\n",
      "[CV] END C=61217.04421344494, gamma=1.6279689407405564, kernel=rbf; total time=  52.2s\n",
      "[CV] END C=61217.04421344494, gamma=1.6279689407405564, kernel=rbf; total time=  37.6s\n",
      "[CV] END C=61217.04421344494, gamma=1.6279689407405564, kernel=rbf; total time=  39.0s\n",
      "[CV] END C=61217.04421344494, gamma=1.6279689407405564, kernel=rbf; total time=  46.8s\n",
      "[CV] END C=926.9787684096649, gamma=2.147979593060577, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=926.9787684096649, gamma=2.147979593060577, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=926.9787684096649, gamma=2.147979593060577, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=926.9787684096649, gamma=2.147979593060577, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=926.9787684096649, gamma=2.147979593060577, kernel=rbf; total time=  13.3s\n",
      "[CV] END C=33946.157064934, gamma=2.2642426492862313, kernel=linear; total time=  15.6s\n",
      "[CV] END C=33946.157064934, gamma=2.2642426492862313, kernel=linear; total time=  18.9s\n",
      "[CV] END C=33946.157064934, gamma=2.2642426492862313, kernel=linear; total time=  16.6s\n",
      "[CV] END C=33946.157064934, gamma=2.2642426492862313, kernel=linear; total time=  19.1s\n",
      "[CV] END C=33946.157064934, gamma=2.2642426492862313, kernel=linear; total time=  18.5s\n",
      "[CV] END C=84789.82947739525, gamma=0.3176359085304841, kernel=linear; total time=  28.3s\n",
      "[CV] END C=84789.82947739525, gamma=0.3176359085304841, kernel=linear; total time=  33.6s\n",
      "[CV] END C=84789.82947739525, gamma=0.3176359085304841, kernel=linear; total time=  29.7s\n",
      "[CV] END C=84789.82947739525, gamma=0.3176359085304841, kernel=linear; total time=  32.0s\n",
      "[CV] END C=84789.82947739525, gamma=0.3176359085304841, kernel=linear; total time=  33.6s\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, estimator=SVR(), n_iter=50,\n",
       "                   param_distributions={'C': <scipy.stats._distn_infrastructure.rv_frozen object at 0x7f7071284860>,\n",
       "                                        'gamma': <scipy.stats._distn_infrastructure.rv_frozen object at 0x7f70712840b8>,\n",
       "                                        'kernel': ['linear', 'rbf']},\n",
       "                   random_state=42, scoring='neg_mean_squared_error',\n",
       "                   verbose=2)"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from scipy.stats import expon, reciprocal\n",
    "\n",
    "# expon(), reciprocal()와 그외 다른 확률 분포 함수에 대해서는\n",
    "# https://docs.scipy.org/doc/scipy/reference/stats.html를 참고하세요.\n",
    "\n",
    "# 노트: kernel 매개변수가 \"linear\"일 때는 gamma가 무시됩니다.\n",
    "param_distribs = {\n",
    "        'kernel': ['linear', 'rbf'],\n",
    "        'C': reciprocal(20, 200000),\n",
    "        'gamma': expon(scale=1.0),\n",
    "    }\n",
    "\n",
    "svm_reg = SVR()\n",
    "rnd_search = RandomizedSearchCV(svm_reg, param_distributions=param_distribs,\n",
    "                                n_iter=50, cv=5, scoring='neg_mean_squared_error',\n",
    "                                verbose=2, random_state=42)\n",
    "rnd_search.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "최상 모델의 (5-폴드 교차 검증으로 평가한) 점수는 다음과 같습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.034290Z",
     "iopub.status.busy": "2021-11-03T05:54:17.033700Z",
     "iopub.status.idle": "2021-11-03T05:54:17.036149Z",
     "shell.execute_reply": "2021-11-03T05:54:17.036584Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "54751.69009488048"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "negative_mse = rnd_search.best_score_\n",
    "rmse = np.sqrt(-negative_mse)\n",
    "rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "이제 `RandomForestRegressor`의 성능에 훨씬 가까워졌습니다(하지만 아직 차이가 납니다). 최상의 하이퍼파라미터를 확인해 보겠습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.039836Z",
     "iopub.status.busy": "2021-11-03T05:54:17.039239Z",
     "iopub.status.idle": "2021-11-03T05:54:17.043409Z",
     "shell.execute_reply": "2021-11-03T05:54:17.043827Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 157055.10989448498, 'gamma': 0.26497040005002437, 'kernel': 'rbf'}"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rnd_search.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "이번에는 RBF 커널에 대해 최적의 하이퍼파라미터 조합을 찾았습니다. 보통 랜덤서치가 같은 시간안에 그리드서치보다 더 좋은 하이퍼파라미터를 찾습니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "여기서 사용된 `scale=1.0`인 지수 분포를 살펴보겠습니다. 일부 샘플은 1.0보다 아주 크거나 작습니다. 하지만 로그 분포를 보면 대부분의 값이 exp(-2)와 exp(+2), 즉 0.1과 7.4 사이에 집중되어 있음을 알 수 있습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.049802Z",
     "iopub.status.busy": "2021-11-03T05:54:17.049168Z",
     "iopub.status.idle": "2021-11-03T05:54:17.375619Z",
     "shell.execute_reply": "2021-11-03T05:54:17.376051Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "expon_distrib = expon(scale=1.)\n",
    "samples = expon_distrib.rvs(10000, random_state=42)\n",
    "plt.figure(figsize=(10, 4))\n",
    "plt.subplot(121)\n",
    "plt.title(\"Exponential distribution (scale=1.0)\")\n",
    "plt.hist(samples, bins=50)\n",
    "plt.subplot(122)\n",
    "plt.title(\"Log of this distribution\")\n",
    "plt.hist(np.log(samples), bins=50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`C`에 사용된 분포는 매우 다릅니다. 주어진 범위안에서 균등 분포로 샘플링됩니다. 그래서 오른쪽 로그 분포가 거의 일정하게 나타납니다. 이런 분포는 원하는 스케일이 정확이 무엇인지 모를 때 사용하면 좋습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.381839Z",
     "iopub.status.busy": "2021-11-03T05:54:17.378566Z",
     "iopub.status.idle": "2021-11-03T05:54:17.787344Z",
     "shell.execute_reply": "2021-11-03T05:54:17.787790Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "reciprocal_distrib = reciprocal(20, 200000)\n",
    "samples = reciprocal_distrib.rvs(10000, random_state=42)\n",
    "plt.figure(figsize=(10, 4))\n",
    "plt.subplot(121)\n",
    "plt.title(\"Reciprocal distribution (scale=1.0)\")\n",
    "plt.hist(samples, bins=50)\n",
    "plt.subplot(122)\n",
    "plt.title(\"Log of this distribution\")\n",
    "plt.hist(np.log(samples), bins=50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`reciprocal()` 함수는 하이퍼파라미터의 스케일에 대해 전혀 감을 잡을 수 없을 때 사용합니다(오른쪽 그래프에서 볼 수 있듯이 주어진 범위안에서 모든 값이 균등합니다). 반면 지수 분포는 하이퍼파라미터의 스케일을 (어느정도) 알고 있을 때 사용하는 것이 좋습니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "질문: 가장 중요한 특성을 선택하는 변환기를 준비 파이프라인에 추가해보세요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.793916Z",
     "iopub.status.busy": "2021-11-03T05:54:17.793283Z",
     "iopub.status.idle": "2021-11-03T05:54:17.796052Z",
     "shell.execute_reply": "2021-11-03T05:54:17.795497Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "\n",
    "def indices_of_top_k(arr, k):\n",
    "    return np.sort(np.argpartition(np.array(arr), -k)[-k:])\n",
    "\n",
    "class TopFeatureSelector(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, feature_importances, k):\n",
    "        self.feature_importances = feature_importances\n",
    "        self.k = k\n",
    "    def fit(self, X, y=None):\n",
    "        self.feature_indices_ = indices_of_top_k(self.feature_importances, self.k)\n",
    "        return self\n",
    "    def transform(self, X):\n",
    "        return X[:, self.feature_indices_]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "노트: 이 특성 선택 클래스는 이미 어떤 식으로든 특성 중요도를 계산했다고 가정합니다(가령 `RandomForestRegressor`을 사용하여). `TopFeatureSelector`의 `fit()` 메서드에서 직접 계산할 수도 있지만 (캐싱을 사용하지 않을 경우) 그리드서치나 랜덤서치의 모든 하이퍼파라미터 조합에 대해 계산이 일어나기 때문에 매우 느려집니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "선택할 특성의 개수를 지정합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.800473Z",
     "iopub.status.busy": "2021-11-03T05:54:17.799291Z",
     "iopub.status.idle": "2021-11-03T05:54:17.801110Z",
     "shell.execute_reply": "2021-11-03T05:54:17.801556Z"
    }
   },
   "outputs": [],
   "source": [
    "k = 5"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "최상의 k개 특성의 인덱스를 확인해 보겠습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.806770Z",
     "iopub.status.busy": "2021-11-03T05:54:17.805823Z",
     "iopub.status.idle": "2021-11-03T05:54:17.809246Z",
     "shell.execute_reply": "2021-11-03T05:54:17.808715Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0,  1,  7,  9, 12])"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_k_feature_indices = indices_of_top_k(feature_importances, k)\n",
    "top_k_feature_indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.813295Z",
     "iopub.status.busy": "2021-11-03T05:54:17.812633Z",
     "iopub.status.idle": "2021-11-03T05:54:17.815241Z",
     "shell.execute_reply": "2021-11-03T05:54:17.815669Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['longitude', 'latitude', 'median_income', 'pop_per_hhold',\n",
       "       'INLAND'], dtype='<U18')"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(attributes)[top_k_feature_indices]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "최상의 k개 특성이 맞는지 다시 확인합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.820236Z",
     "iopub.status.busy": "2021-11-03T05:54:17.819636Z",
     "iopub.status.idle": "2021-11-03T05:54:17.824882Z",
     "shell.execute_reply": "2021-11-03T05:54:17.824292Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0.3790092248170967, 'median_income'),\n",
       " (0.16570630316895876, 'INLAND'),\n",
       " (0.10703132208204354, 'pop_per_hhold'),\n",
       " (0.06965425227942929, 'longitude'),\n",
       " (0.0604213840080722, 'latitude')]"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(zip(feature_importances, attributes), reverse=True)[:k]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "좋습니다. 이제 이전에 정의한 준비 파이프라인과 특성 선택기를 추가한 새로운 파이프라인을 만듭니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.829108Z",
     "iopub.status.busy": "2021-11-03T05:54:17.828299Z",
     "iopub.status.idle": "2021-11-03T05:54:17.829904Z",
     "shell.execute_reply": "2021-11-03T05:54:17.830353Z"
    }
   },
   "outputs": [],
   "source": [
    "preparation_and_feature_selection_pipeline = Pipeline([\n",
    "    ('preparation', full_pipeline),\n",
    "    ('feature_selection', TopFeatureSelector(feature_importances, k))\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.833409Z",
     "iopub.status.busy": "2021-11-03T05:54:17.832686Z",
     "iopub.status.idle": "2021-11-03T05:54:17.870332Z",
     "shell.execute_reply": "2021-11-03T05:54:17.869564Z"
    }
   },
   "outputs": [],
   "source": [
    "housing_prepared_top_k_features = preparation_and_feature_selection_pipeline.fit_transform(housing)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "처음 3개 샘플의 특성을 확인해 보겠습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.875203Z",
     "iopub.status.busy": "2021-11-03T05:54:17.874575Z",
     "iopub.status.idle": "2021-11-03T05:54:17.879215Z",
     "shell.execute_reply": "2021-11-03T05:54:17.879722Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.94135046,  1.34743822, -0.8936472 ,  0.00622264,  1.        ],\n",
       "       [ 1.17178212, -1.19243966,  1.292168  , -0.04081077,  0.        ],\n",
       "       [ 0.26758118, -0.1259716 , -0.52543365, -0.07537122,  1.        ]])"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_prepared_top_k_features[0:3]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "최상의 k개 특성이 맞는지 다시 확인합니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.882774Z",
     "iopub.status.busy": "2021-11-03T05:54:17.882071Z",
     "iopub.status.idle": "2021-11-03T05:54:17.885915Z",
     "shell.execute_reply": "2021-11-03T05:54:17.886333Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-0.94135046,  1.34743822, -0.8936472 ,  0.00622264,  1.        ],\n",
       "       [ 1.17178212, -1.19243966,  1.292168  , -0.04081077,  0.        ],\n",
       "       [ 0.26758118, -0.1259716 , -0.52543365, -0.07537122,  1.        ]])"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_prepared[0:3, top_k_feature_indices]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "성공입니다!  :)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "질문: 전체 데이터 준비 과정과 최종 예측을 하나의 파이프라인으로 만들어보세요."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.891010Z",
     "iopub.status.busy": "2021-11-03T05:54:17.889385Z",
     "iopub.status.idle": "2021-11-03T05:54:17.893061Z",
     "shell.execute_reply": "2021-11-03T05:54:17.892487Z"
    }
   },
   "outputs": [],
   "source": [
    "prepare_select_and_predict_pipeline = Pipeline([\n",
    "    ('preparation', full_pipeline),\n",
    "    ('feature_selection', TopFeatureSelector(feature_importances, k)),\n",
    "    ('svm_reg', SVR(**rnd_search.best_params_))\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:17.900199Z",
     "iopub.status.busy": "2021-11-03T05:54:17.899506Z",
     "iopub.status.idle": "2021-11-03T05:54:36.283554Z",
     "shell.execute_reply": "2021-11-03T05:54:36.284020Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(steps=[('preparation',\n",
       "                 ColumnTransformer(transformers=[('num',\n",
       "                                                  Pipeline(steps=[('imputer',\n",
       "                                                                   SimpleImputer(strategy='median')),\n",
       "                                                                  ('attribs_adder',\n",
       "                                                                   CombinedAttributesAdder()),\n",
       "                                                                  ('std_scaler',\n",
       "                                                                   StandardScaler())]),\n",
       "                                                  ['longitude', 'latitude',\n",
       "                                                   'housing_median_age',\n",
       "                                                   'total_rooms',\n",
       "                                                   'total_bedrooms',\n",
       "                                                   'population', 'households',\n",
       "                                                   'median_income']),\n",
       "                                                 ('cat', OneHotEncoder(...\n",
       "                 TopFeatureSelector(feature_importances=array([6.96542523e-02, 6.04213840e-02, 4.21882202e-02, 1.52450557e-02,\n",
       "       1.55545295e-02, 1.58491147e-02, 1.49346552e-02, 3.79009225e-01,\n",
       "       5.47789150e-02, 1.07031322e-01, 4.82031213e-02, 6.79266007e-03,\n",
       "       1.65706303e-01, 7.83480660e-05, 1.52473276e-03, 3.02816106e-03]),\n",
       "                                    k=5)),\n",
       "                ('svm_reg',\n",
       "                 SVR(C=157055.10989448498, gamma=0.26497040005002437))])"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prepare_select_and_predict_pipeline.fit(housing, housing_labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "몇 개의 샘플에 전체 파이프라인을 적용해 보겠습니다:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:36.291856Z",
     "iopub.status.busy": "2021-11-03T05:54:36.286763Z",
     "iopub.status.idle": "2021-11-03T05:54:36.300247Z",
     "shell.execute_reply": "2021-11-03T05:54:36.299729Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predictions:\t [ 83384.49158095 299407.90439234  92272.03345144 150173.16199041]\n",
      "Labels:\t\t [72100.0, 279600.0, 82700.0, 112500.0]\n"
     ]
    }
   ],
   "source": [
    "some_data = housing.iloc[:4]\n",
    "some_labels = housing_labels.iloc[:4]\n",
    "\n",
    "print(\"Predictions:\\t\", prepare_select_and_predict_pipeline.predict(some_data))\n",
    "print(\"Labels:\\t\\t\", list(some_labels))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "전체 파이프라인이 잘 작동하는 것 같습니다. 물론 예측 성능이 아주 좋지는 않습니다. `SVR`보다 `RandomForestRegressor`가 더 나은 것 같습니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "질문: `GridSearchCV`를 사용해 준비 단계의 옵션을 자동으로 탐색해보세요.\n",
    "\n",
    "**경고**: 사용하는 하드웨어에 따라 다음 셀을 실행하는데 45분 또는 그 이상 걸릴 수 있습니다.\n",
    "\n",
    "**노트**: 아래 코드에서 훈련 도중 경고를 피하기 위해 `OneHotEncoder`의 `handle_unknown` 하이퍼파라미터를 `'ignore'`로 지정했습니다. 그렇지 않으면 `OneHotEncoder`는 기본적으로 `handle_unkown='error'`를 사용하기 때문에 데이터를 변활할 때 훈련할 때 없던 범주가 있으면 에러를 냅니다. 기본값을 사용하면 훈련 세트에 모든 카테고리가 들어 있지 않은 폴드를 평가할 때 `GridSearchCV`가 에러를 일으킵니다. `'ISLAND'` 범주에는 샘플이 하나이기 때문에 일어날 가능성이 높습니다. 일부 폴드에서는 테스트 세트 안에 포함될 수 있습니다. 따라서 이런 폴드는 `GridSearchCV`에서 무시하여 피하는 것이 좋습니다."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T05:54:36.307334Z",
     "iopub.status.busy": "2021-11-03T05:54:36.306593Z",
     "iopub.status.idle": "2021-11-03T07:14:06.433521Z",
     "shell.execute_reply": "2021-11-03T07:14:06.432930Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 48 candidates, totalling 240 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=mean; total time=   9.6s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=mean; total time=  11.4s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=mean; total time=  11.4s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=mean; total time=  11.5s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=mean; total time=  11.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=median; total time=   9.6s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=median; total time=  11.4s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=median; total time=  11.4s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=median; total time=  11.5s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=median; total time=  11.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=most_frequent; total time=   9.5s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=most_frequent; total time=  11.4s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=most_frequent; total time=  11.4s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=most_frequent; total time=  11.5s\n",
      "[CV] END feature_selection__k=1, preparation__num__imputer__strategy=most_frequent; total time=  11.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=mean; total time=   9.6s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=mean; total time=  11.8s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=mean; total time=  11.7s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=mean; total time=  11.8s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=mean; total time=  11.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=median; total time=   9.6s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=median; total time=  11.8s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=median; total time=  12.0s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=median; total time=  11.8s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=median; total time=  11.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=most_frequent; total time=   9.6s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=most_frequent; total time=  11.8s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=most_frequent; total time=  11.7s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=most_frequent; total time=  11.8s\n",
      "[CV] END feature_selection__k=2, preparation__num__imputer__strategy=most_frequent; total time=  11.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=mean; total time=   9.7s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=mean; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=mean; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=mean; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=mean; total time=  12.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=median; total time=   9.7s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=median; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=median; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=median; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=median; total time=  12.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=most_frequent; total time=   9.7s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=most_frequent; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=most_frequent; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=most_frequent; total time=  11.9s\n",
      "[CV] END feature_selection__k=3, preparation__num__imputer__strategy=most_frequent; total time=  12.0s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=mean; total time=  11.3s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=mean; total time=  13.6s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=mean; total time=  13.6s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=mean; total time=  13.6s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=mean; total time=  13.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=median; total time=  11.3s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=median; total time=  13.6s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=median; total time=  13.6s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=median; total time=  13.6s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=median; total time=  13.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=most_frequent; total time=  11.3s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=most_frequent; total time=  13.7s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=most_frequent; total time=  13.6s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=most_frequent; total time=  13.6s\n",
      "[CV] END feature_selection__k=4, preparation__num__imputer__strategy=most_frequent; total time=  13.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=mean; total time=  11.6s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=mean; total time=  14.3s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=mean; total time=  14.1s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=mean; total time=  14.1s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=mean; total time=  14.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=median; total time=  11.6s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=median; total time=  14.3s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=median; total time=  14.1s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=median; total time=  14.1s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=median; total time=  14.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=most_frequent; total time=  11.6s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=most_frequent; total time=  14.3s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=most_frequent; total time=  14.1s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=most_frequent; total time=  14.1s\n",
      "[CV] END feature_selection__k=5, preparation__num__imputer__strategy=most_frequent; total time=  14.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=mean; total time=  12.0s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=mean; total time=  14.4s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=mean; total time=  14.9s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=mean; total time=  14.6s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=mean; total time=  14.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=median; total time=  12.0s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=median; total time=  14.4s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=median; total time=  14.9s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=median; total time=  14.6s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=median; total time=  14.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=most_frequent; total time=  12.0s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=most_frequent; total time=  14.4s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=most_frequent; total time=  14.9s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=most_frequent; total time=  14.6s\n",
      "[CV] END feature_selection__k=6, preparation__num__imputer__strategy=most_frequent; total time=  14.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=mean; total time=  12.5s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=mean; total time=  15.7s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=mean; total time=  15.6s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=mean; total time=  14.8s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=mean; total time=  15.8s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=median; total time=  12.7s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=median; total time=  15.7s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=median; total time=  15.4s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=median; total time=  15.5s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=median; total time=  15.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=most_frequent; total time=  12.6s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=most_frequent; total time=  15.3s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=most_frequent; total time=  15.0s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=most_frequent; total time=  15.2s\n",
      "[CV] END feature_selection__k=7, preparation__num__imputer__strategy=most_frequent; total time=  15.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=mean; total time=  16.0s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=mean; total time=  16.8s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=mean; total time=  17.0s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=mean; total time=  17.4s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=mean; total time=  17.1s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=median; total time=  15.1s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=median; total time=  16.6s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=median; total time=  17.6s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=median; total time=  17.6s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=median; total time=  18.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=most_frequent; total time=  14.7s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=most_frequent; total time=  17.6s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=most_frequent; total time=  16.9s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=most_frequent; total time=  18.9s\n",
      "[CV] END feature_selection__k=8, preparation__num__imputer__strategy=most_frequent; total time=  17.9s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=mean; total time=  18.9s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=mean; total time=  21.3s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=mean; total time=  21.7s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=mean; total time=  21.6s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=mean; total time=  21.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=median; total time=  18.3s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=median; total time=  22.7s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=median; total time=  22.2s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=median; total time=  21.3s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=median; total time=  22.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=most_frequent; total time=  16.6s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=most_frequent; total time=  23.5s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=most_frequent; total time=  21.6s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=most_frequent; total time=  20.0s\n",
      "[CV] END feature_selection__k=9, preparation__num__imputer__strategy=most_frequent; total time=  22.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=mean; total time=  24.6s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=mean; total time=  25.5s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=mean; total time=  21.9s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=mean; total time=  26.3s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=mean; total time=  23.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=median; total time=  20.4s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=median; total time=  25.9s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=median; total time=  25.1s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=median; total time=  26.3s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=median; total time=  24.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=most_frequent; total time=  23.5s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=most_frequent; total time=  26.0s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=most_frequent; total time=  26.9s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=most_frequent; total time=  27.4s\n",
      "[CV] END feature_selection__k=10, preparation__num__imputer__strategy=most_frequent; total time=  23.6s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=mean; total time=  22.9s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=mean; total time=  29.0s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=mean; total time=  27.5s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=mean; total time=  29.8s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=mean; total time=  26.3s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=median; total time=  22.5s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=median; total time=  26.6s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=median; total time=  24.6s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=median; total time=  30.0s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=median; total time=  26.9s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=most_frequent; total time=  27.0s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=most_frequent; total time=  27.4s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=most_frequent; total time=  29.2s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=most_frequent; total time=  29.0s\n",
      "[CV] END feature_selection__k=11, preparation__num__imputer__strategy=most_frequent; total time=  29.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=mean; total time=  24.1s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=mean; total time=  31.6s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=mean; total time=  31.6s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=mean; total time=  30.4s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=mean; total time=  28.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=median; total time=  27.8s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=median; total time=  30.5s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=median; total time=  26.8s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=median; total time=  25.5s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=median; total time=  30.7s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=most_frequent; total time=  29.3s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=most_frequent; total time=  29.7s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=most_frequent; total time=  33.2s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=most_frequent; total time=  27.9s\n",
      "[CV] END feature_selection__k=12, preparation__num__imputer__strategy=most_frequent; total time=  31.5s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=mean; total time=  29.6s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=mean; total time=  28.7s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=mean; total time=  31.8s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=mean; total time=  34.3s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=mean; total time=  28.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=median; total time=  30.1s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=median; total time=  33.4s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=median; total time=  28.3s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=median; total time=  33.7s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=median; total time=  33.4s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:775: UserWarning: Scoring failed. The score on this train-test partition for these parameters will be set to nan. Details: \n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 762, in _score\n",
      "    scores = scorer(estimator, X_test, y_test)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 221, in __call__\n",
      "    sample_weight=sample_weight,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 258, in _score\n",
      "    y_pred = method_caller(estimator, \"predict\", X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/metrics/_scorer.py\", line 68, in _cached_call\n",
      "    return getattr(estimator, method)(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/metaestimators.py\", line 113, in <lambda>\n",
      "    out = lambda *args, **kwargs: self.fn(obj, *args, **kwargs)  # noqa\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 469, in predict\n",
      "    Xt = transform.transform(Xt)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 753, in transform\n",
      "    column_as_strings=fit_dataframe_and_transform_dataframe,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/compose/_column_transformer.py\", line 615, in _fit_transform\n",
      "    for idx, (name, trans, column, weight) in enumerate(transformers, 1)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 1007, in __call__\n",
      "    while self.dispatch_one_batch(iterator):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 835, in dispatch_one_batch\n",
      "    self._dispatch(tasks)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 754, in _dispatch\n",
      "    job = self._backend.apply_async(batch, callback=cb)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 209, in apply_async\n",
      "    result = ImmediateResult(func)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/_parallel_backends.py\", line 590, in __init__\n",
      "    self.results = batch()\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in __call__\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/parallel.py\", line 256, in <listcomp>\n",
      "    for func, args, kwargs in self.items]\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/utils/fixes.py\", line 209, in __call__\n",
      "    return self.function(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 874, in _transform_one\n",
      "    res = transformer.transform(X)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 513, in transform\n",
      "    warn_on_unknown=warn_on_unknown,\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/preprocessing/_encoders.py\", line 142, in _transform\n",
      "    raise ValueError(msg)\n",
      "ValueError: Found unknown categories ['ISLAND'] in column 0 during transform\n",
      "\n",
      "  UserWarning,\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=most_frequent; total time=  30.3s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=most_frequent; total time=  27.5s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=most_frequent; total time=  32.8s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=most_frequent; total time=  29.5s\n",
      "[CV] END feature_selection__k=13, preparation__num__imputer__strategy=most_frequent; total time=  29.1s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=mean; total time=   0.0s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=mean; total time=  30.6s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=mean; total time=  34.1s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=mean; total time=  32.4s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=mean; total time=  33.9s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=median; total time=   0.0s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=median; total time=  31.6s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=median; total time=  29.2s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=median; total time=  33.7s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=median; total time=  30.0s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=most_frequent; total time=   0.0s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=most_frequent; total time=  34.2s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=most_frequent; total time=  34.6s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=most_frequent; total time=  28.7s\n",
      "[CV] END feature_selection__k=14, preparation__num__imputer__strategy=most_frequent; total time=  33.2s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=mean; total time=   0.0s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=mean; total time=  31.1s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=mean; total time=  32.4s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=mean; total time=  30.9s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=mean; total time=  31.5s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=median; total time=   0.0s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=median; total time=  32.2s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=median; total time=  33.7s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=median; total time=  33.1s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=median; total time=  33.1s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=most_frequent; total time=   0.0s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=most_frequent; total time=  34.9s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=most_frequent; total time=  34.2s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=most_frequent; total time=  33.7s\n",
      "[CV] END feature_selection__k=15, preparation__num__imputer__strategy=most_frequent; total time=  33.2s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=mean; total time=   0.0s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=mean; total time=  32.6s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=mean; total time=  29.9s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=mean; total time=  33.1s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=mean; total time=  32.7s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=median; total time=   0.0s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=median; total time=  32.9s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=median; total time=  28.1s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=median; total time=  32.4s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=median; total time=  28.2s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=most_frequent; total time=   0.0s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=most_frequent; total time=  31.9s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=most_frequent; total time=  33.2s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=most_frequent; total time=  34.0s\n",
      "[CV] END feature_selection__k=16, preparation__num__imputer__strategy=most_frequent; total time=  30.2s\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py:372: FitFailedWarning: \n",
      "9 fits failed out of a total of 240.\n",
      "The score on these train-test partitions for these parameters will be set to nan.\n",
      "If these failures are not expected, you can try to debug them by setting error_score='raise'.\n",
      "\n",
      "Below are more details about the failures:\n",
      "--------------------------------------------------------------------------------\n",
      "9 fits failed with the following error:\n",
      "Traceback (most recent call last):\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_validation.py\", line 681, in _fit_and_score\n",
      "    estimator.fit(X_train, y_train, **fit_params)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 390, in fit\n",
      "    Xt = self._fit(X, y, **fit_params_steps)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 355, in _fit\n",
      "    **fit_params_steps[name],\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/joblib/memory.py\", line 355, in __call__\n",
      "    return self.func(*args, **kwargs)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/pipeline.py\", line 891, in _fit_transform_one\n",
      "    res = transformer.fit_transform(X, y, **fit_params)\n",
      "  File \"/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/base.py\", line 847, in fit_transform\n",
      "    return self.fit(X, y, **fit_params).transform(X)\n",
      "  File \"/tmp/ipykernel_1868/2088711179.py\", line 14, in transform\n",
      "    return X[:, self.feature_indices_]\n",
      "IndexError: index 15 is out of bounds for axis 1 with size 15\n",
      "\n",
      "  warnings.warn(some_fits_failed_message, FitFailedWarning)\n",
      "/home/haesun/handson-ml2/.env/lib/python3.7/site-packages/sklearn/model_selection/_search.py:972: UserWarning: One or more of the test scores are non-finite: [nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan nan\n",
      " nan nan nan nan nan nan nan nan nan nan nan nan]\n",
      "  category=UserWarning,\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5,\n",
       "             estimator=Pipeline(steps=[('preparation',\n",
       "                                        ColumnTransformer(transformers=[('num',\n",
       "                                                                         Pipeline(steps=[('imputer',\n",
       "                                                                                          SimpleImputer(strategy='median')),\n",
       "                                                                                         ('attribs_adder',\n",
       "                                                                                          CombinedAttributesAdder()),\n",
       "                                                                                         ('std_scaler',\n",
       "                                                                                          StandardScaler())]),\n",
       "                                                                         ['longitude',\n",
       "                                                                          'latitude',\n",
       "                                                                          'housing_median_age',\n",
       "                                                                          'total_rooms',\n",
       "                                                                          'total_bedrooms',\n",
       "                                                                          'population',\n",
       "                                                                          'households',\n",
       "                                                                          'median_inc...\n",
       "       5.47789150e-02, 1.07031322e-01, 4.82031213e-02, 6.79266007e-03,\n",
       "       1.65706303e-01, 7.83480660e-05, 1.52473276e-03, 3.02816106e-03]),\n",
       "                                                           k=5)),\n",
       "                                       ('svm_reg',\n",
       "                                        SVR(C=157055.10989448498,\n",
       "                                            gamma=0.26497040005002437))]),\n",
       "             param_grid=[{'feature_selection__k': [1, 2, 3, 4, 5, 6, 7, 8, 9,\n",
       "                                                   10, 11, 12, 13, 14, 15, 16],\n",
       "                          'preparation__num__imputer__strategy': ['mean',\n",
       "                                                                  'median',\n",
       "                                                                  'most_frequent']}],\n",
       "             scoring='neg_mean_squared_error', verbose=2)"
      ]
     },
     "execution_count": 136,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "full_pipeline.named_transformers_[\"cat\"].handle_unknown = 'ignore'\n",
    "\n",
    "param_grid = [{\n",
    "    'preparation__num__imputer__strategy': ['mean', 'median', 'most_frequent'],\n",
    "    'feature_selection__k': list(range(1, len(feature_importances) + 1))\n",
    "}]\n",
    "\n",
    "grid_search_prep = GridSearchCV(prepare_select_and_predict_pipeline, param_grid, cv=5,\n",
    "                                scoring='neg_mean_squared_error', verbose=2)\n",
    "grid_search_prep.fit(housing, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-11-03T07:14:06.437916Z",
     "iopub.status.busy": "2021-11-03T07:14:06.437325Z",
     "iopub.status.idle": "2021-11-03T07:14:06.439725Z",
     "shell.execute_reply": "2021-11-03T07:14:06.440133Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'feature_selection__k': 1, 'preparation__num__imputer__strategy': 'mean'}"
      ]
     },
     "execution_count": 137,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_search_prep.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "최상의 `Imputer` 정책은 `most_frequent`이고 거의 모든 특성이 유용합니다(16개 중 15개). 마지막 특성(`ISLAND`)은 잡음이 추가될 뿐입니다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "축하합니다! 이제 머신러닝에 대해 꽤 많은 것을 알게 되었습니다. :)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "nav_menu": {
   "height": "279px",
   "width": "309px"
  },
  "toc": {
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": "block",
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}