230223_prediction_model.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/oghzzang/tutorial/blob/phd/230223_prediction_model.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9bwghj55ebAt",
        "outputId": "2cd8ebac-1f83-4242-ff8d-f28c16b6d30a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%matplotlib inline"
      ],
      "metadata": {
        "id": "qDd42-ePeip0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import sklearn\n",
        "from sklearn import metrics\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.model_selection import GridSearchCV, KFold, train_test_split\n",
        "from sklearn.model_selection import LeaveOneOut, cross_val_score, StratifiedKFold\n",
        "from sklearn.metrics import auc, plot_roc_curve, confusion_matrix, accuracy_score, roc_curve\n",
        "import pickle\n",
        "import joblib\n",
        "from joblib import dump, load\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.model_selection import cross_val_score\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import mean_squared_error\n",
        "import math\n",
        "from sklearn import preprocessing\n",
        "import pandas\n",
        "import seaborn\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "\n",
        "from sklearn.preprocessing import OrdinalEncoder\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.model_selection import ShuffleSplit\n",
        "from sklearn import preprocessing\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn import linear_model\n",
        "import numpy as np\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import roc_auc_score\n",
        "from sklearn.metrics import plot_roc_curve\n",
        "import seaborn\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.model_selection import train_test_split, GridSearchCV\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import accuracy_score\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "from sklearn.datasets import load_wine\n",
        "\n",
        "from sklearn.preprocessing import MinMaxScaler, StandardScaler  \n",
        "from sklearn.model_selection import train_test_split, KFold, cross_val_score, GridSearchCV, RandomizedSearchCV\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.pipeline import Pipeline, make_pipeline\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "import pandas as pd \n",
        "print(sklearn.__version__)\n",
        "\n",
        "import numpy as np\n",
        "from sklearn.metrics import roc_auc_score\n",
        "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier, BaggingClassifier, VotingClassifier, ExtraTreesClassifier, GradientBoostingClassifier\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.pipeline import make_pipeline\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import f1_score, confusion_matrix\n",
        "import random\n",
        "random.seed(300)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mf8t7IbAemIl",
        "outputId": "9adb56ec-d114-48b9-c035-9f8d63dcaae3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "1.0.2\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 1. 데이터셋 불러오기"
      ],
      "metadata": {
        "id": "sysqqULnKk_z"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "data = pd.read_csv('/content/drive/My Drive/Colab Notebooks/co_mitochondria/12.prediction model (n=180).csv')\n",
        "\n",
        "data.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 382
        },
        "id": "mcsOtcrxepp6",
        "outputId": "50ff1282-1306-4969-8e6f-fafe311a8586"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   obs  patient_no subject_no sample_barcode group  sample_YY  sex      batch  \\\n",
              "0  179    59493635      P0412          P0412    pd       2020    2  second pd   \n",
              "1  142    50290101      P0288    KBB00004393    pd       2019    2    batch 2   \n",
              "2   24    51572192      P0201    KBB00002159    pd       2016    1    batch 1   \n",
              "3  173    44625997      P0349          P0349    pd       2020    1  second pd   \n",
              "4  134    53389459      P0368          P0368    pd       2020    1  second pd   \n",
              "\n",
              "   dm  htn  ...  mtcn_2  mtcn_3  time_points  min_year max_year min_bxyear  \\\n",
              "0   1    1  ...       1       1            1      2020     2020          0   \n",
              "1   1    0  ...       1       1            2      2019     2020          0   \n",
              "2   0    1  ...       1       1            3      2016     2019          0   \n",
              "3   0    0  ...       1       1            1      2020     2020          0   \n",
              "4   0    0  ...       1       1            4      2017     2021         -3   \n",
              "\n",
              "   updrs_ii_mean  updrs_iii_mean  updrs_ii_iii_mean  updrs_ii_iii_group  \n",
              "0      24.000000       49.000000          73.000000                   2  \n",
              "1       6.500000       35.000000          41.500000                   2  \n",
              "2       4.500000       33.000000          37.500000                   2  \n",
              "3       5.000000       42.000000          47.000000                   2  \n",
              "4       4.666667       17.666667          22.333333                   1  \n",
              "\n",
              "[5 rows x 43 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c8ee6a07-4f71-4ba1-a323-eb74048f8ece\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>obs</th>\n",
              "      <th>patient_no</th>\n",
              "      <th>subject_no</th>\n",
              "      <th>sample_barcode</th>\n",
              "      <th>group</th>\n",
              "      <th>sample_YY</th>\n",
              "      <th>sex</th>\n",
              "      <th>batch</th>\n",
              "      <th>dm</th>\n",
              "      <th>htn</th>\n",
              "      <th>...</th>\n",
              "      <th>mtcn_2</th>\n",
              "      <th>mtcn_3</th>\n",
              "      <th>time_points</th>\n",
              "      <th>min_year</th>\n",
              "      <th>max_year</th>\n",
              "      <th>min_bxyear</th>\n",
              "      <th>updrs_ii_mean</th>\n",
              "      <th>updrs_iii_mean</th>\n",
              "      <th>updrs_ii_iii_mean</th>\n",
              "      <th>updrs_ii_iii_group</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>179</td>\n",
              "      <td>59493635</td>\n",
              "      <td>P0412</td>\n",
              "      <td>P0412</td>\n",
              "      <td>pd</td>\n",
              "      <td>2020</td>\n",
              "      <td>2</td>\n",
              "      <td>second pd</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2020</td>\n",
              "      <td>2020</td>\n",
              "      <td>0</td>\n",
              "      <td>24.000000</td>\n",
              "      <td>49.000000</td>\n",
              "      <td>73.000000</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>142</td>\n",
              "      <td>50290101</td>\n",
              "      <td>P0288</td>\n",
              "      <td>KBB00004393</td>\n",
              "      <td>pd</td>\n",
              "      <td>2019</td>\n",
              "      <td>2</td>\n",
              "      <td>batch 2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2019</td>\n",
              "      <td>2020</td>\n",
              "      <td>0</td>\n",
              "      <td>6.500000</td>\n",
              "      <td>35.000000</td>\n",
              "      <td>41.500000</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>24</td>\n",
              "      <td>51572192</td>\n",
              "      <td>P0201</td>\n",
              "      <td>KBB00002159</td>\n",
              "      <td>pd</td>\n",
              "      <td>2016</td>\n",
              "      <td>1</td>\n",
              "      <td>batch 1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2016</td>\n",
              "      <td>2019</td>\n",
              "      <td>0</td>\n",
              "      <td>4.500000</td>\n",
              "      <td>33.000000</td>\n",
              "      <td>37.500000</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>173</td>\n",
              "      <td>44625997</td>\n",
              "      <td>P0349</td>\n",
              "      <td>P0349</td>\n",
              "      <td>pd</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>second pd</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2020</td>\n",
              "      <td>2020</td>\n",
              "      <td>0</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>42.000000</td>\n",
              "      <td>47.000000</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>134</td>\n",
              "      <td>53389459</td>\n",
              "      <td>P0368</td>\n",
              "      <td>P0368</td>\n",
              "      <td>pd</td>\n",
              "      <td>2020</td>\n",
              "      <td>1</td>\n",
              "      <td>second pd</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>2017</td>\n",
              "      <td>2021</td>\n",
              "      <td>-3</td>\n",
              "      <td>4.666667</td>\n",
              "      <td>17.666667</td>\n",
              "      <td>22.333333</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>5 rows × 43 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c8ee6a07-4f71-4ba1-a323-eb74048f8ece')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-c8ee6a07-4f71-4ba1-a323-eb74048f8ece button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-c8ee6a07-4f71-4ba1-a323-eb74048f8ece');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "seaborn.pairplot(data[[\"log2.mtcn\", \"updrs_ii_mean\", \"updrs_iii_mean\", \"updrs_ii_iii_mean\"]], \n",
        "                 diag_kind = \"kde\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 743
        },
        "id": "OOiYh26LesZY",
        "outputId": "f9a51067-e4a8-402b-98a2-a0c0a8a5c269"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.PairGrid at 0x7f9fa9f34430>"
            ]
          },
          "metadata": {},
          "execution_count": 31
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x720 with 20 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data = data[['sex', 'sample_age', \n",
        "             'onset_age', 'dx_age',\n",
        "             'dm', 'htn',\n",
        "             'mtcn', 'updrs_ii_iii_group']]\n",
        "\n",
        "\n",
        "#encoder_ordinal = preprocessing.OrdinalEncoder()\n",
        "data['sex'] = data['sex'].astype('category')\n",
        "data['updrs_ii_iii_group'] = data['updrs_ii_iii_group'].astype('category')\n",
        "data['dm'] = data['dm'].astype('category')\n",
        "data['htn'] = data['htn'].astype('category')\n",
        "\n",
        "data['mtcn.log2'] = np.log2(data['mtcn'])\n",
        "data['sample_age.log2'] = np.log2(data['sample_age'])\n",
        "data['onset_age.log2'] = np.log2(data['onset_age'])\n",
        "data['dx_age.log2'] = np.log2(data['dx_age'])\n",
        "\n",
        "data = data[['sex', \n",
        "             'sample_age.log2', \n",
        "             'onset_age.log2', \n",
        "             'dx_age.log2',\n",
        "             'dm', 'htn',\n",
        "             'mtcn.log2', 'updrs_ii_iii_group']]\n",
        "\n",
        "#data['mtcn_3'] = data['mtcn_3'].astype('category')\n",
        "\n",
        "\n",
        "cols = data.columns\n",
        "num_cols = data._get_numeric_data().columns\n",
        "categorical_cols = list(set(cols) - set(num_cols))\n",
        "\n",
        "lbl_encode = LabelEncoder()\n",
        "for i in categorical_cols:\n",
        "    num = data.columns.get_loc(i)\n",
        "    data.iloc[:,num] = lbl_encode.fit_transform(data.iloc[:,num])\n",
        "\n",
        "#data['updrs_ii_iii_group'] = encoder_ordinal.fit_transform(data[['updrs_ii_iii_group']])\n",
        "\n",
        "\n",
        "corr = data.corr()\n",
        "sns.heatmap(corr)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "6b9AmPVTexcB",
        "outputId": "03dd252f-f94c-4f0b-973d-b6b998ae293c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f9fa8251790>"
            ]
          },
          "metadata": {},
          "execution_count": 32
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- 변수간에 높은 correlation 을 보이는 feature는 제외함. "
      ],
      "metadata": {
        "id": "tK8rM7XRK6pu"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "columns = np.full((corr.shape[0],), True, dtype=bool)\n",
        "for i in range(corr.shape[0]):\n",
        "    for j in range(i+1, corr.shape[0]):\n",
        "        if corr.iloc[i,j] >= 0.9:\n",
        "            if columns[j]:\n",
        "                columns[j] = False\n",
        "\n",
        "selected_columns = data.columns[columns]\n",
        "data = data[selected_columns]\n",
        "data"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 419
        },
        "id": "0zLCvxFUezlB",
        "outputId": "a541c2e9-49eb-4221-9039-8a02b4963135"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     sex  sample_age.log2  dm  htn  mtcn.log2  updrs_ii_iii_group\n",
              "0      1         6.285402   1    1   5.822701                   1\n",
              "1      1         6.000000   1    0   5.924114                   1\n",
              "2      0         6.066089   0    1   5.961564                   1\n",
              "3      0         6.000000   0    0   6.130713                   1\n",
              "4      0         5.523562   0    0   6.206385                   0\n",
              "..   ...              ...  ..  ...        ...                 ...\n",
              "175    0         6.108524   1    0   8.492410                   0\n",
              "176    0         6.129283   0    1   8.539976                   1\n",
              "177    0         5.584963   0    0   8.644584                   0\n",
              "178    0         6.044394   0    0   8.688077                   1\n",
              "179    1         5.906891   0    1   8.704231                   0\n",
              "\n",
              "[180 rows x 6 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-38b28bb6-146e-46f4-895b-7c8049d1f954\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>sex</th>\n",
              "      <th>sample_age.log2</th>\n",
              "      <th>dm</th>\n",
              "      <th>htn</th>\n",
              "      <th>mtcn.log2</th>\n",
              "      <th>updrs_ii_iii_group</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>6.285402</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5.822701</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>6.000000</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>5.924114</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>6.066089</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>5.961564</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>6.000000</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.130713</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>5.523562</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6.206385</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>175</th>\n",
              "      <td>0</td>\n",
              "      <td>6.108524</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>8.492410</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>176</th>\n",
              "      <td>0</td>\n",
              "      <td>6.129283</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>8.539976</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>177</th>\n",
              "      <td>0</td>\n",
              "      <td>5.584963</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.644584</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>178</th>\n",
              "      <td>0</td>\n",
              "      <td>6.044394</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.688077</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>179</th>\n",
              "      <td>1</td>\n",
              "      <td>5.906891</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>8.704231</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>180 rows × 6 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-38b28bb6-146e-46f4-895b-7c8049d1f954')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-38b28bb6-146e-46f4-895b-7c8049d1f954 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-38b28bb6-146e-46f4-895b-7c8049d1f954');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- 결과변수 : updrs_ii_iii_group\n",
        "- 설명변수 : sex, sample_age (log2), dm, htn, mtcn (log2)"
      ],
      "metadata": {
        "id": "4zwn9w1iLGTB"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 2. Split data for tuning hyperparameters"
      ],
      "metadata": {
        "id": "lrSOeTx2kWG4"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 1) 전체 180명의 데이터를 training set 80%, test set 20%로 분리함\n",
        "\n",
        "=> Q. 데이터 수가 적은 경우에 이정도 비율로 분리해서 사용하는게 적합할지??\n",
        "\n",
        "![01.dataset 분리.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "WbReXgNrWVEB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "X = data.iloc[:, :-1]\n",
        "y = data.iloc[:, -1]"
      ],
      "metadata": {
        "id": "XJ8MCRBoe1j9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, \n",
        "                                                    test_size = 0.2,\n",
        "                                                    shuffle=True, stratify=y,\n",
        "                                                    random_state = 100)"
      ],
      "metadata": {
        "id": "qq6tdHnCe3rF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 2) RandomizedSearchCV를 사용하여 hyperparameter tuning"
      ],
      "metadata": {
        "id": "f2CQyTbzEboG"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from pprint import pprint\n",
        "\n",
        "rfc = RandomForestClassifier(random_state = 100)\n",
        "\n",
        "# Look at parameters used by our current forest\n",
        "print('Parameters currently in use:\\n')\n",
        "pprint(rfc.get_params())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uL9u5k3je5hs",
        "outputId": "ef934762-cbd6-4a94-e2db-3fa4f1fc9cbe"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Parameters currently in use:\n",
            "\n",
            "{'bootstrap': True,\n",
            " 'ccp_alpha': 0.0,\n",
            " 'class_weight': None,\n",
            " 'criterion': 'gini',\n",
            " 'max_depth': None,\n",
            " 'max_features': 'auto',\n",
            " 'max_leaf_nodes': None,\n",
            " 'max_samples': None,\n",
            " 'min_impurity_decrease': 0.0,\n",
            " 'min_samples_leaf': 1,\n",
            " 'min_samples_split': 2,\n",
            " 'min_weight_fraction_leaf': 0.0,\n",
            " 'n_estimators': 100,\n",
            " 'n_jobs': None,\n",
            " 'oob_score': False,\n",
            " 'random_state': 100,\n",
            " 'verbose': 0,\n",
            " 'warm_start': False}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- n_estimators = number of trees in the foreset\n",
        "- max_features = max number of features considered for splitting a node\n",
        "- max_depth = max number of levels in each decision tree\n",
        "- min_samples_split = min number of data points placed in a node before the node is split\n",
        "- min_samples_leaf = min number of data points allowed in a leaf node\n",
        "- bootstrap = method for sampling data points (with or without replacement)"
      ],
      "metadata": {
        "id": "kxua-BQtmL1Q"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 3) 최적의 parameter 찾기"
      ],
      "metadata": {
        "id": "h3kI3kg7lrcE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "y_train.value_counts()\n",
        "y_test.value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7lKIBfjyoK2L",
        "outputId": "ddafc728-8eba-47ea-e5cb-b64ea444d1e3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    18\n",
              "1    18\n",
              "Name: updrs_ii_iii_group, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import RandomizedSearchCV\n",
        "\n",
        "# Number of trees in random forest\n",
        "n_estimators = [int(x) for x in np.linspace(start = 10, stop = 300, num = 30)]\n",
        "\n",
        "# Number of features to consider at every split\n",
        "max_features = ['auto', 'sqrt', 'log2', None]\n",
        "\n",
        "# Maximum number of levels in tree\n",
        "max_depth = [int(x) for x in np.linspace(3, 50, num = 25)]\n",
        "max_depth.append(None)\n",
        "\n",
        "# Minimum number of samples required to split a node\n",
        "min_samples_split = range(1,40)\n",
        "\n",
        "# Minimum number of samples required at each leaf node\n",
        "min_samples_leaf = range(1,40)\n",
        "\n",
        "# Method of selecting samples for training each tree\n",
        "bootstrap = [True, False]\n",
        "\n",
        "# Create the random grid\n",
        "random_grid = {'n_estimators': n_estimators,\n",
        "               'max_features': max_features,\n",
        "               'max_depth': max_depth,\n",
        "               'min_samples_split': min_samples_split,\n",
        "               'min_samples_leaf': min_samples_leaf,\n",
        "               'bootstrap': bootstrap}\n",
        "pprint(random_grid)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cRF1aoLIe-ir",
        "outputId": "b43756df-b059-4130-890a-3c19587414dc"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'bootstrap': [True, False],\n",
            " 'max_depth': [3,\n",
            "               4,\n",
            "               6,\n",
            "               8,\n",
            "               10,\n",
            "               12,\n",
            "               14,\n",
            "               16,\n",
            "               18,\n",
            "               20,\n",
            "               22,\n",
            "               24,\n",
            "               26,\n",
            "               28,\n",
            "               30,\n",
            "               32,\n",
            "               34,\n",
            "               36,\n",
            "               38,\n",
            "               40,\n",
            "               42,\n",
            "               44,\n",
            "               46,\n",
            "               48,\n",
            "               50,\n",
            "               None],\n",
            " 'max_features': ['auto', 'sqrt', 'log2', None],\n",
            " 'min_samples_leaf': range(1, 40),\n",
            " 'min_samples_split': range(1, 40),\n",
            " 'n_estimators': [10,\n",
            "                  20,\n",
            "                  30,\n",
            "                  40,\n",
            "                  50,\n",
            "                  60,\n",
            "                  70,\n",
            "                  80,\n",
            "                  90,\n",
            "                  100,\n",
            "                  110,\n",
            "                  120,\n",
            "                  130,\n",
            "                  140,\n",
            "                  150,\n",
            "                  160,\n",
            "                  170,\n",
            "                  180,\n",
            "                  190,\n",
            "                  200,\n",
            "                  210,\n",
            "                  220,\n",
            "                  230,\n",
            "                  240,\n",
            "                  250,\n",
            "                  260,\n",
            "                  270,\n",
            "                  280,\n",
            "                  290,\n",
            "                  300]}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Use the random grid to search for best hyperparameters\n",
        "# First create the base model to tune\n",
        "rf = RandomForestClassifier()\n",
        "\n",
        "# Random search of parameters, using 5 fold cross validation, \n",
        "# search across 100 different combinations, and use all available cores\n",
        "rf_random = RandomizedSearchCV(estimator = rf, \n",
        "                               param_distributions = random_grid, \n",
        "                               n_iter = 100, \n",
        "                               cv = 5, \n",
        "                               verbose=2, \n",
        "                               random_state=100, n_jobs = -1)"
      ],
      "metadata": {
        "id": "NqGDMYHbfBmr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Fit the random search model 모델 훈련\n",
        "rf_random.fit(X_train, y_train)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GgnwL-UkfDS9",
        "outputId": "3536fdf8-3d9a-4c9d-a00a-41702bb2d174"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Fitting 5 folds for each of 100 candidates, totalling 500 fits\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "RandomizedSearchCV(cv=5, estimator=RandomForestClassifier(), n_iter=100,\n",
              "                   n_jobs=-1,\n",
              "                   param_distributions={'bootstrap': [True, False],\n",
              "                                        'max_depth': [3, 4, 6, 8, 10, 12, 14,\n",
              "                                                      16, 18, 20, 22, 24, 26,\n",
              "                                                      28, 30, 32, 34, 36, 38,\n",
              "                                                      40, 42, 44, 46, 48, 50,\n",
              "                                                      None],\n",
              "                                        'max_features': ['auto', 'sqrt', 'log2',\n",
              "                                                         None],\n",
              "                                        'min_samples_leaf': range(1, 40),\n",
              "                                        'min_samples_split': range(1, 40),\n",
              "                                        'n_estimators': [10, 20, 30, 40, 50, 60,\n",
              "                                                         70, 80, 90, 100, 110,\n",
              "                                                         120, 130, 140, 150,\n",
              "                                                         160, 170, 180, 190,\n",
              "                                                         200, 210, 220, 230,\n",
              "                                                         240, 250, 260, 270,\n",
              "                                                         280, 290, 300]},\n",
              "                   random_state=100, verbose=2)"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 4) randomizedSearchCV로 찾은 최적의 hyperparameter를 확인하기"
      ],
      "metadata": {
        "id": "nqmtvTTa1GvB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print('최적 하이퍼파라미터: ', rf_random.best_params_)\n",
        "print('검증 정확도:', rf_random.best_score_)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "84Nz8P2GfFNc",
        "outputId": "0f7d6767-8fa0-40b7-aed7-47c3491a918d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "최적 하이퍼파라미터:  {'n_estimators': 240, 'min_samples_split': 36, 'min_samples_leaf': 18, 'max_features': 'sqrt', 'max_depth': 20, 'bootstrap': False}\n",
            "검증 정확도: 0.6118226600985222\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 5) 최적의 hyperparameter로 모델 만들기"
      ],
      "metadata": {
        "id": "uAfNkYUXWoQW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "rf_best = RandomForestClassifier(n_estimators=rf_random.best_params_['n_estimators'],\n",
        "                                  max_features=rf_random.best_params_['max_features'],\n",
        "                                  max_depth=rf_random.best_params_['max_depth'],\n",
        "                                  min_samples_split=rf_random.best_params_['min_samples_split'],\n",
        "                                  min_samples_leaf=rf_random.best_params_['min_samples_leaf'],\n",
        "                                  bootstrap=rf_random.best_params_['bootstrap'],\n",
        "                                  verbose=2, \n",
        "                                  random_state=100, n_jobs = -1)"
      ],
      "metadata": {
        "id": "Ql2Yq1YX1OOr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 6) 기본 모델과 비교하기\n",
        "\n",
        "- base model (rf)\n",
        "- tuned moel (rf_best)"
      ],
      "metadata": {
        "id": "KnpOjU1b1p0U"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "rf = RandomForestClassifier(random_state = 100)\n",
        "rf.fit(X_train, y_train)\n",
        "rf_score = rf.score(X_test, y_test)\n",
        "rf_pred = rf.predict(X_test)\n",
        "rf_accuracy = accuracy_score(y_test, rf_pred)\n",
        "rf_f1 = f1_score(y_test, rf_pred)\n",
        "\n",
        "print(f\"기본 모델 성능: {rf_score}\")\n",
        "print(\"Accuracy: {:.2f}%\".format(rf_accuracy*100))\n",
        "print(\"F1 score: {:.2f}\".format(rf_f1))\n",
        "\n",
        "##--\n",
        "rf_best.fit(X_train, y_train)\n",
        "rf_best_score = rf_best.score(X_test, y_test)\n",
        "rf_best_pred = rf_best.predict(X_test)\n",
        "print(f\"튜닝한 모델 성능: {rf_best_score}\")\n",
        "rf_best_accuracy = accuracy_score(y_test, rf_best_pred)\n",
        "rf_best_f1 = f1_score(y_test, rf_best_pred)\n",
        "\n",
        "print(\"Accuracy: {:.2f}%\".format(rf_best_accuracy*100))\n",
        "print(\"F1 score: {:.2f}\".format(rf_best_f1))\n",
        "\n",
        "#print('검증 정확도:', base_model.score(X_test, y_test))\n",
        "#y_test_pred = base_model.predict(X_test)\n",
        "#print('F1:', f1_score(y_test, y_test_pred))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AIw63k4K1pY2",
        "outputId": "87efe1f5-2f7b-44f5-8028-fb80eae3c9e6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "기본 모델 성능: 0.5555555555555556\n",
            "Accuracy: 55.56%\n",
            "F1 score: 0.56\n",
            "building tree 1 of 240\n",
            "building tree 2 of 240\n",
            "building tree 3 of 240\n",
            "building tree 4 of 240\n",
            "building tree 5 of 240\n",
            "building tree 6 of 240building tree 7 of 240\n",
            "\n",
            "building tree 8 of 240\n",
            "building tree 9 of 240\n",
            "building tree 10 of 240\n",
            "building tree 11 of 240\n",
            "building tree 12 of 240\n",
            "building tree 13 of 240\n",
            "building tree 14 of 240\n",
            "building tree 15 of 240\n",
            "building tree 16 of 240\n",
            "building tree 17 of 240\n",
            "building tree 18 of 240\n",
            "building tree 19 of 240\n",
            "building tree 20 of 240\n",
            "building tree 21 of 240building tree 22 of 240\n",
            "\n",
            "building tree 23 of 240\n",
            "building tree 24 of 240\n",
            "building tree 25 of 240\n",
            "building tree 26 of 240\n",
            "building tree 27 of 240\n",
            "building tree 28 of 240\n",
            "building tree 29 of 240\n",
            "building tree 30 of 240\n",
            "building tree 31 of 240\n",
            "building tree 32 of 240\n",
            "building tree 33 of 240\n",
            "building tree 34 of 240\n",
            "building tree 35 of 240\n",
            "building tree 36 of 240\n",
            "building tree 37 of 240\n",
            "building tree 38 of 240\n",
            "building tree 39 of 240\n",
            "building tree 40 of 240\n",
            "building tree 41 of 240\n",
            "building tree 42 of 240\n",
            "building tree 43 of 240\n",
            "building tree 44 of 240\n",
            "building tree 45 of 240\n",
            "building tree 46 of 240\n",
            "building tree 47 of 240\n",
            "building tree 48 of 240\n",
            "building tree 49 of 240\n",
            "building tree 50 of 240\n",
            "building tree 51 of 240\n",
            "building tree 52 of 240\n",
            "building tree 53 of 240\n",
            "building tree 54 of 240\n",
            "building tree 55 of 240\n",
            "building tree 56 of 240\n",
            "building tree 57 of 240\n",
            "building tree 58 of 240\n",
            "building tree 59 of 240\n",
            "building tree 60 of 240\n",
            "building tree 61 of 240\n",
            "building tree 62 of 240\n",
            "building tree 63 of 240\n",
            "building tree 64 of 240building tree 65 of 240\n",
            "\n",
            "building tree 66 of 240\n",
            "building tree 67 of 240\n",
            "building tree 68 of 240building tree 69 of 240\n",
            "\n",
            "building tree 70 of 240\n",
            "building tree 71 of 240\n",
            "building tree 72 of 240\n",
            "building tree 73 of 240\n",
            "building tree 74 of 240\n",
            "building tree 75 of 240\n",
            "building tree 76 of 240\n",
            "building tree 77 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  37 tasks      | elapsed:    0.1s\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 78 of 240\n",
            "building tree 79 of 240\n",
            "building tree 80 of 240\n",
            "building tree 81 of 240\n",
            "building tree 82 of 240\n",
            "building tree 83 of 240\n",
            "building tree 84 of 240\n",
            "building tree 85 of 240\n",
            "building tree 86 of 240\n",
            "building tree 87 of 240\n",
            "building tree 88 of 240\n",
            "building tree 89 of 240\n",
            "building tree 90 of 240\n",
            "building tree 91 of 240\n",
            "building tree 92 of 240building tree 93 of 240\n",
            "\n",
            "building tree 94 of 240\n",
            "building tree 95 of 240\n",
            "building tree 96 of 240\n",
            "building tree 97 of 240\n",
            "building tree 98 of 240\n",
            "building tree 99 of 240\n",
            "building tree 100 of 240\n",
            "building tree 101 of 240\n",
            "building tree 102 of 240building tree 103 of 240\n",
            "\n",
            "building tree 104 of 240\n",
            "building tree 105 of 240\n",
            "building tree 106 of 240\n",
            "building tree 107 of 240\n",
            "building tree 108 of 240\n",
            "building tree 109 of 240\n",
            "building tree 110 of 240\n",
            "building tree 111 of 240\n",
            "building tree 112 of 240\n",
            "building tree 113 of 240building tree 114 of 240\n",
            "building tree 115 of 240\n",
            "\n",
            "building tree 116 of 240\n",
            "building tree 117 of 240\n",
            "building tree 118 of 240\n",
            "building tree 119 of 240\n",
            "building tree 120 of 240\n",
            "building tree 121 of 240\n",
            "building tree 122 of 240\n",
            "building tree 123 of 240\n",
            "building tree 124 of 240\n",
            "building tree 125 of 240\n",
            "building tree 126 of 240\n",
            "building tree 127 of 240\n",
            "building tree 128 of 240\n",
            "building tree 129 of 240\n",
            "building tree 130 of 240\n",
            "building tree 131 of 240\n",
            "building tree 132 of 240\n",
            "building tree 133 of 240\n",
            "building tree 134 of 240\n",
            "building tree 135 of 240\n",
            "building tree 136 of 240\n",
            "building tree 137 of 240\n",
            "building tree 138 of 240\n",
            "building tree 139 of 240\n",
            "building tree 140 of 240building tree 141 of 240\n",
            "\n",
            "building tree 142 of 240\n",
            "building tree 143 of 240\n",
            "building tree 144 of 240\n",
            "building tree 145 of 240\n",
            "building tree 146 of 240\n",
            "building tree 147 of 240\n",
            "building tree 148 of 240\n",
            "building tree 149 of 240\n",
            "building tree 150 of 240\n",
            "building tree 151 of 240\n",
            "building tree 152 of 240\n",
            "building tree 153 of 240\n",
            "building tree 154 of 240\n",
            "building tree 155 of 240\n",
            "building tree 156 of 240\n",
            "building tree 157 of 240\n",
            "building tree 158 of 240\n",
            "building tree 159 of 240\n",
            "building tree 160 of 240\n",
            "building tree 161 of 240\n",
            "building tree 162 of 240\n",
            "building tree 163 of 240\n",
            "building tree 164 of 240\n",
            "building tree 165 of 240\n",
            "building tree 166 of 240\n",
            "building tree 167 of 240\n",
            "building tree 168 of 240\n",
            "building tree 169 of 240\n",
            "building tree 170 of 240\n",
            "building tree 171 of 240\n",
            "building tree 172 of 240\n",
            "building tree 173 of 240\n",
            "building tree 174 of 240building tree 175 of 240\n",
            "building tree 176 of 240\n",
            "\n",
            "building tree 177 of 240\n",
            "building tree 178 of 240\n",
            "building tree 179 of 240building tree 180 of 240\n",
            "\n",
            "building tree 181 of 240building tree 182 of 240\n",
            "\n",
            "building tree 183 of 240\n",
            "building tree 184 of 240\n",
            "building tree 185 of 240\n",
            "building tree 186 of 240\n",
            "building tree 187 of 240\n",
            "building tree 188 of 240\n",
            "building tree 189 of 240building tree 190 of 240\n",
            "\n",
            "building tree 191 of 240building tree 192 of 240\n",
            "\n",
            "building tree 193 of 240\n",
            "building tree 194 of 240\n",
            "building tree 195 of 240\n",
            "building tree 196 of 240building tree 197 of 240\n",
            "building tree 198 of 240\n",
            "building tree 199 of 240\n",
            "building tree 200 of 240\n",
            "building tree 201 of 240"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done 158 tasks      | elapsed:    0.3s\n",
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.4s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "building tree 202 of 240building tree 203 of 240\n",
            "building tree 204 of 240\n",
            "building tree 205 of 240\n",
            "building tree 206 of 240\n",
            "building tree 207 of 240\n",
            "building tree 208 of 240\n",
            "building tree 209 of 240\n",
            "\n",
            "building tree 210 of 240building tree 211 of 240\n",
            "\n",
            "building tree 212 of 240building tree 213 of 240\n",
            "\n",
            "building tree 214 of 240building tree 215 of 240\n",
            "\n",
            "building tree 216 of 240\n",
            "building tree 217 of 240\n",
            "building tree 218 of 240\n",
            "building tree 219 of 240building tree 220 of 240\n",
            "\n",
            "building tree 221 of 240\n",
            "building tree 222 of 240\n",
            "building tree 223 of 240\n",
            "building tree 224 of 240\n",
            "building tree 225 of 240\n",
            "building tree 226 of 240\n",
            "building tree 227 of 240\n",
            "building tree 228 of 240\n",
            "building tree 229 of 240\n",
            "building tree 230 of 240\n",
            "building tree 231 of 240\n",
            "building tree 232 of 240\n",
            "building tree 233 of 240\n",
            "building tree 234 of 240\n",
            "building tree 235 of 240\n",
            "building tree 236 of 240building tree 237 of 240\n",
            "\n",
            "building tree 238 of 240\n",
            "building tree 239 of 240\n",
            "building tree 240 of 240\n",
            "튜닝한 모델 성능: 0.6388888888888888\n",
            "Accuracy: 63.89%\n",
            "F1 score: 0.65\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# 선정한 하이퍼 파라미터를 이용해서 RandomForest 수행하기"
      ],
      "metadata": {
        "id": "sODUuTuQsGNE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score, roc_auc_score, confusion_matrix, roc_curve, precision_score, recall_score\n",
        "from sklearn.model_selection import KFold, cross_val_predict\n",
        "\n",
        "# Perform 5-fold cross validation using KFold\n",
        "cv = KFold(n_splits=5, shuffle=True, random_state=100)\n",
        "accuracy_scores = []\n",
        "auc_scores = []\n",
        "sensitivity_scores = []\n",
        "specificity_scores = []\n",
        "ppv_scores = []\n",
        "important_features = []\n",
        "\n",
        "for train_index, test_index in cv.split(X):\n",
        "    # Split data into training and testing sets for this fold\n",
        "    X_train, X_test = X.iloc[train_index], X.iloc[test_index]\n",
        "    y_train, y_test = y.iloc[train_index], y.iloc[test_index]\n",
        "    \n",
        "    # Fit model on training data\n",
        "    rf_best.fit(X_train, y_train)\n",
        "    \n",
        "    # Make predictions on testing data\n",
        "    y_pred = rf_best.predict(X_test)\n",
        "    \n",
        "    # Calculate metrics for this fold\n",
        "    accuracy_scores.append(accuracy_score(y_test, y_pred))\n",
        "    auc_scores.append(roc_auc_score(y_test, y_pred))\n",
        "    tn, fp, fn, tp = confusion_matrix(y_test, y_pred).ravel()\n",
        "    sensitivity_scores.append(tp / (tp + fn))\n",
        "    specificity_scores.append(tn / (tn + fp))\n",
        "    ppv_scores.append(tp / (tp + fp))\n",
        "    important_features.append(rf_best.feature_importances_)\n",
        "\n",
        "# Print the mean evaluation metrics across all folds\n",
        "print('Cross validated accuracy:', np.mean(accuracy_scores))\n",
        "print('Cross validated accuracy (std):', np.std(accuracy_scores))\n",
        "print('Cross validated AUC:', np.mean(auc_scores))\n",
        "print('Cross validated AUC (std):', np.std(auc_scores))\n",
        "print('Cross validated sensitivity:', np.mean(sensitivity_scores))\n",
        "print('Cross validated sensitivity (std):', np.std(sensitivity_scores))\n",
        "print('Cross validated specificity:', np.mean(specificity_scores))\n",
        "print('Cross validated specificity (std):', np.std(specificity_scores))\n",
        "print('Cross validated PPV:', np.mean(ppv_scores))\n",
        "print('Cross validated PPV (std):', np.std(ppv_scores))\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Yg5RTRuh8fu7",
        "outputId": "1f80d7e9-1eab-4d13-8286-a49ede2e13e4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done  37 tasks      | elapsed:    0.1s\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 1 of 240\n",
            "building tree 2 of 240\n",
            "building tree 3 of 240\n",
            "building tree 4 of 240\n",
            "building tree 5 of 240\n",
            "building tree 6 of 240\n",
            "building tree 7 of 240building tree 8 of 240\n",
            "\n",
            "building tree 9 of 240\n",
            "building tree 10 of 240\n",
            "building tree 11 of 240\n",
            "building tree 12 of 240building tree 13 of 240\n",
            "building tree 14 of 240\n",
            "building tree 15 of 240\n",
            "\n",
            "building tree 16 of 240building tree 17 of 240\n",
            "building tree 18 of 240\n",
            "building tree 19 of 240\n",
            "building tree 20 of 240\n",
            "building tree 21 of 240\n",
            "building tree 22 of 240\n",
            "building tree 23 of 240\n",
            "building tree 24 of 240\n",
            "\n",
            "building tree 25 of 240building tree 26 of 240\n",
            "building tree 27 of 240\n",
            "building tree 28 of 240\n",
            "building tree 29 of 240\n",
            "building tree 30 of 240\n",
            "\n",
            "building tree 31 of 240\n",
            "building tree 32 of 240\n",
            "building tree 33 of 240building tree 34 of 240\n",
            "building tree 35 of 240\n",
            "building tree 36 of 240\n",
            "\n",
            "building tree 37 of 240\n",
            "building tree 38 of 240\n",
            "building tree 39 of 240building tree 40 of 240\n",
            "\n",
            "building tree 41 of 240\n",
            "building tree 42 of 240\n",
            "building tree 43 of 240building tree 44 of 240\n",
            "building tree 45 of 240\n",
            "building tree 46 of 240\n",
            "\n",
            "building tree 47 of 240\n",
            "building tree 48 of 240\n",
            "building tree 49 of 240building tree 50 of 240\n",
            "\n",
            "building tree 51 of 240\n",
            "building tree 52 of 240\n",
            "building tree 53 of 240\n",
            "building tree 54 of 240\n",
            "building tree 55 of 240building tree 56 of 240\n",
            "\n",
            "building tree 57 of 240\n",
            "building tree 58 of 240\n",
            "building tree 59 of 240building tree 60 of 240\n",
            "building tree 61 of 240\n",
            "\n",
            "building tree 62 of 240\n",
            "building tree 63 of 240\n",
            "building tree 64 of 240\n",
            "building tree 65 of 240\n",
            "building tree 66 of 240\n",
            "building tree 67 of 240\n",
            "building tree 68 of 240building tree 69 of 240\n",
            "\n",
            "building tree 70 of 240\n",
            "building tree 71 of 240\n",
            "building tree 72 of 240\n",
            "building tree 73 of 240\n",
            "building tree 74 of 240\n",
            "building tree 75 of 240\n",
            "building tree 76 of 240\n",
            "building tree 77 of 240\n",
            "building tree 78 of 240\n",
            "building tree 79 of 240building tree 80 of 240\n",
            "\n",
            "building tree 81 of 240\n",
            "building tree 82 of 240\n",
            "building tree 83 of 240\n",
            "building tree 84 of 240\n",
            "building tree 85 of 240\n",
            "building tree 86 of 240\n",
            "building tree 87 of 240\n",
            "building tree 88 of 240\n",
            "building tree 89 of 240\n",
            "building tree 90 of 240\n",
            "building tree 91 of 240building tree 92 of 240\n",
            "\n",
            "building tree 93 of 240\n",
            "building tree 94 of 240\n",
            "building tree 95 of 240building tree 96 of 240\n",
            "\n",
            "building tree 97 of 240\n",
            "building tree 98 of 240\n",
            "building tree 99 of 240\n",
            "building tree 100 of 240\n",
            "building tree 101 of 240\n",
            "building tree 102 of 240\n",
            "building tree 103 of 240building tree 104 of 240\n",
            "\n",
            "building tree 105 of 240\n",
            "building tree 106 of 240\n",
            "building tree 107 of 240\n",
            "building tree 108 of 240\n",
            "building tree 109 of 240building tree 110 of 240\n",
            "\n",
            "building tree 111 of 240\n",
            "building tree 112 of 240\n",
            "building tree 113 of 240\n",
            "building tree 114 of 240\n",
            "building tree 115 of 240\n",
            "building tree 116 of 240\n",
            "building tree 117 of 240\n",
            "building tree 118 of 240\n",
            "building tree 119 of 240\n",
            "building tree 120 of 240\n",
            "building tree 121 of 240\n",
            "building tree 122 of 240\n",
            "building tree 123 of 240\n",
            "building tree 124 of 240\n",
            "building tree 125 of 240\n",
            "building tree 126 of 240\n",
            "building tree 127 of 240\n",
            "building tree 128 of 240\n",
            "building tree 129 of 240\n",
            "building tree 130 of 240\n",
            "building tree 131 of 240\n",
            "building tree 132 of 240\n",
            "building tree 133 of 240\n",
            "building tree 134 of 240building tree 135 of 240\n",
            "\n",
            "building tree 136 of 240\n",
            "building tree 137 of 240\n",
            "building tree 138 of 240\n",
            "building tree 139 of 240\n",
            "building tree 140 of 240\n",
            "building tree 141 of 240building tree 142 of 240\n",
            "\n",
            "building tree 143 of 240\n",
            "building tree 144 of 240\n",
            "building tree 145 of 240\n",
            "building tree 146 of 240\n",
            "building tree 147 of 240\n",
            "building tree 148 of 240\n",
            "building tree 149 of 240\n",
            "building tree 150 of 240\n",
            "building tree 151 of 240\n",
            "building tree 152 of 240\n",
            "building tree 153 of 240\n",
            "building tree 154 of 240\n",
            "building tree 155 of 240\n",
            "building tree 156 of 240\n",
            "building tree 157 of 240\n",
            "building tree 158 of 240\n",
            "building tree 159 of 240\n",
            "building tree 160 of 240\n",
            "building tree 161 of 240\n",
            "building tree 162 of 240\n",
            "building tree 163 of 240\n",
            "building tree 164 of 240building tree 165 of 240\n",
            "\n",
            "building tree 166 of 240\n",
            "building tree 167 of 240building tree 168 of 240\n",
            "building tree 169 of 240\n",
            "building tree 170 of 240\n",
            "building tree 171 of 240\n",
            "building tree 172 of 240\n",
            "building tree 173 of 240\n",
            "building tree 174 of 240\n",
            "building tree 175 of 240\n",
            "\n",
            "building tree 176 of 240\n",
            "building tree 177 of 240\n",
            "building tree 178 of 240building tree 179 of 240\n",
            "\n",
            "building tree 180 of 240\n",
            "building tree 181 of 240building tree 182 of 240\n",
            "building tree 183 of 240\n",
            "\n",
            "building tree 184 of 240building tree 185 of 240\n",
            "\n",
            "building tree 186 of 240\n",
            "building tree 187 of 240\n",
            "building tree 188 of 240\n",
            "building tree 189 of 240\n",
            "building tree 190 of 240\n",
            "building tree 191 of 240\n",
            "building tree 192 of 240\n",
            "building tree 193 of 240\n",
            "building tree 194 of 240\n",
            "building tree 195 of 240\n",
            "building tree 196 of 240\n",
            "building tree 197 of 240building tree 198 of 240\n",
            "\n",
            "building tree 199 of 240\n",
            "building tree 200 of 240building tree 201 of 240\n",
            "building tree 202 of 240\n",
            "building tree 203 of 240\n",
            "\n",
            "building tree 204 of 240building tree 205 of 240\n",
            "\n",
            "building tree 206 of 240building tree 207 of 240\n",
            "\n",
            "building tree 208 of 240\n",
            "building tree 209 of 240building tree 210 of 240\n",
            "\n",
            "building tree 211 of 240\n",
            "building tree 212 of 240\n",
            "building tree 213 of 240\n",
            "building tree 214 of 240\n",
            "building tree 215 of 240\n",
            "building tree 216 of 240\n",
            "building tree 217 of 240\n",
            "building tree 218 of 240building tree 219 of 240\n",
            "\n",
            "building tree 220 of 240\n",
            "building tree 221 of 240\n",
            "building tree 222 of 240building tree 223 of 240\n",
            "\n",
            "building tree 224 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done 158 tasks      | elapsed:    0.3s\n",
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.5s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 225 of 240\n",
            "building tree 226 of 240\n",
            "building tree 227 of 240\n",
            "building tree 228 of 240building tree 229 of 240\n",
            "\n",
            "building tree 230 of 240\n",
            "building tree 231 of 240\n",
            "building tree 232 of 240\n",
            "building tree 233 of 240\n",
            "building tree 234 of 240building tree 235 of 240\n",
            "\n",
            "building tree 236 of 240\n",
            "building tree 237 of 240\n",
            "building tree 238 of 240\n",
            "building tree 239 of 240\n",
            "building tree 240 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n",
            "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 2 concurrent workers.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 1 of 240\n",
            "building tree 2 of 240\n",
            "building tree 3 of 240building tree 4 of 240\n",
            "\n",
            "building tree 5 of 240\n",
            "building tree 6 of 240\n",
            "building tree 7 of 240building tree 8 of 240\n",
            "\n",
            "building tree 9 of 240\n",
            "building tree 10 of 240\n",
            "building tree 11 of 240\n",
            "building tree 12 of 240\n",
            "building tree 13 of 240\n",
            "building tree 14 of 240\n",
            "building tree 15 of 240\n",
            "building tree 16 of 240\n",
            "building tree 17 of 240\n",
            "building tree 18 of 240\n",
            "building tree 19 of 240building tree 20 of 240\n",
            "\n",
            "building tree 21 of 240\n",
            "building tree 22 of 240building tree 23 of 240\n",
            "\n",
            "building tree 24 of 240\n",
            "building tree 25 of 240building tree 26 of 240\n",
            "\n",
            "building tree 27 of 240\n",
            "building tree 28 of 240\n",
            "building tree 29 of 240\n",
            "building tree 30 of 240\n",
            "building tree 31 of 240\n",
            "building tree 32 of 240\n",
            "building tree 33 of 240\n",
            "building tree 34 of 240\n",
            "building tree 35 of 240\n",
            "building tree 36 of 240\n",
            "building tree 37 of 240\n",
            "building tree 38 of 240\n",
            "building tree 39 of 240\n",
            "building tree 40 of 240\n",
            "building tree 41 of 240\n",
            "building tree 42 of 240building tree 43 of 240\n",
            "\n",
            "building tree 44 of 240\n",
            "building tree 45 of 240\n",
            "building tree 46 of 240\n",
            "building tree 47 of 240\n",
            "building tree 48 of 240\n",
            "building tree 49 of 240building tree 50 of 240\n",
            "\n",
            "building tree 51 of 240\n",
            "building tree 52 of 240\n",
            "building tree 53 of 240\n",
            "building tree 54 of 240\n",
            "building tree 55 of 240\n",
            "building tree 56 of 240\n",
            "building tree 57 of 240\n",
            "building tree 58 of 240\n",
            "building tree 59 of 240\n",
            "building tree 60 of 240\n",
            "building tree 61 of 240\n",
            "building tree 62 of 240\n",
            "building tree 63 of 240\n",
            "building tree 64 of 240\n",
            "building tree 65 of 240building tree 66 of 240\n",
            "\n",
            "building tree 67 of 240\n",
            "building tree 68 of 240\n",
            "building tree 69 of 240\n",
            "building tree 70 of 240\n",
            "building tree 71 of 240\n",
            "building tree 72 of 240\n",
            "building tree 73 of 240\n",
            "building tree 74 of 240\n",
            "building tree 75 of 240\n",
            "building tree 76 of 240\n",
            "building tree 77 of 240\n",
            "building tree 78 of 240\n",
            "building tree 79 of 240\n",
            "building tree 80 of 240\n",
            "building tree 81 of 240\n",
            "building tree 82 of 240\n",
            "building tree 83 of 240\n",
            "building tree 84 of 240\n",
            "building tree 85 of 240\n",
            "building tree 86 of 240\n",
            "building tree 87 of 240building tree 88 of 240\n",
            "\n",
            "building tree 89 of 240\n",
            "building tree 90 of 240\n",
            "building tree 91 of 240building tree 92 of 240\n",
            "\n",
            "building tree 93 of 240\n",
            "building tree 94 of 240\n",
            "building tree 95 of 240\n",
            "building tree 96 of 240\n",
            "building tree 97 of 240\n",
            "building tree 98 of 240\n",
            "building tree 99 of 240\n",
            "building tree 100 of 240\n",
            "building tree 101 of 240\n",
            "building tree 102 of 240\n",
            "building tree 103 of 240\n",
            "building tree 104 of 240\n",
            "building tree 105 of 240\n",
            "building tree 106 of 240\n",
            "building tree 107 of 240\n",
            "building tree 108 of 240\n",
            "building tree 109 of 240\n",
            "building tree 110 of 240\n",
            "building tree 111 of 240\n",
            "building tree 112 of 240\n",
            "building tree 113 of 240\n",
            "building tree 114 of 240\n",
            "building tree 115 of 240\n",
            "building tree 116 of 240\n",
            "building tree 117 of 240\n",
            "building tree 118 of 240\n",
            "building tree 119 of 240\n",
            "building tree 120 of 240\n",
            "building tree 121 of 240\n",
            "building tree 122 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done  37 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=-1)]: Done 158 tasks      | elapsed:    0.3s\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 123 of 240\n",
            "building tree 124 of 240\n",
            "building tree 125 of 240\n",
            "building tree 126 of 240\n",
            "building tree 127 of 240\n",
            "building tree 128 of 240\n",
            "building tree 129 of 240\n",
            "building tree 130 of 240\n",
            "building tree 131 of 240\n",
            "building tree 132 of 240\n",
            "building tree 133 of 240\n",
            "building tree 134 of 240\n",
            "building tree 135 of 240building tree 136 of 240\n",
            "\n",
            "building tree 137 of 240\n",
            "building tree 138 of 240\n",
            "building tree 139 of 240\n",
            "building tree 140 of 240\n",
            "building tree 141 of 240\n",
            "building tree 142 of 240\n",
            "building tree 143 of 240\n",
            "building tree 144 of 240\n",
            "building tree 145 of 240\n",
            "building tree 146 of 240\n",
            "building tree 147 of 240\n",
            "building tree 148 of 240\n",
            "building tree 149 of 240\n",
            "building tree 150 of 240\n",
            "building tree 151 of 240\n",
            "building tree 152 of 240\n",
            "building tree 153 of 240building tree 154 of 240\n",
            "\n",
            "building tree 155 of 240\n",
            "building tree 156 of 240\n",
            "building tree 157 of 240\n",
            "building tree 158 of 240\n",
            "building tree 159 of 240\n",
            "building tree 160 of 240\n",
            "building tree 161 of 240\n",
            "building tree 162 of 240\n",
            "building tree 163 of 240\n",
            "building tree 164 of 240\n",
            "building tree 165 of 240\n",
            "building tree 166 of 240\n",
            "building tree 167 of 240\n",
            "building tree 168 of 240\n",
            "building tree 169 of 240\n",
            "building tree 170 of 240\n",
            "building tree 171 of 240\n",
            "building tree 172 of 240\n",
            "building tree 173 of 240\n",
            "building tree 174 of 240building tree 175 of 240\n",
            "\n",
            "building tree 176 of 240\n",
            "building tree 177 of 240\n",
            "building tree 178 of 240\n",
            "building tree 179 of 240\n",
            "building tree 180 of 240\n",
            "building tree 181 of 240\n",
            "building tree 182 of 240\n",
            "building tree 183 of 240\n",
            "building tree 184 of 240\n",
            "building tree 185 of 240\n",
            "building tree 186 of 240\n",
            "building tree 187 of 240\n",
            "building tree 188 of 240\n",
            "building tree 189 of 240\n",
            "building tree 190 of 240\n",
            "building tree 191 of 240\n",
            "building tree 192 of 240\n",
            "building tree 193 of 240\n",
            "building tree 194 of 240\n",
            "building tree 195 of 240\n",
            "building tree 196 of 240\n",
            "building tree 197 of 240\n",
            "building tree 198 of 240\n",
            "building tree 199 of 240\n",
            "building tree 200 of 240\n",
            "building tree 201 of 240\n",
            "building tree 202 of 240\n",
            "building tree 203 of 240\n",
            "building tree 204 of 240\n",
            "building tree 205 of 240\n",
            "building tree 206 of 240\n",
            "building tree 207 of 240\n",
            "building tree 208 of 240\n",
            "building tree 209 of 240\n",
            "building tree 210 of 240\n",
            "building tree 211 of 240\n",
            "building tree 212 of 240\n",
            "building tree 213 of 240\n",
            "building tree 214 of 240\n",
            "building tree 215 of 240\n",
            "building tree 216 of 240\n",
            "building tree 217 of 240\n",
            "building tree 218 of 240\n",
            "building tree 219 of 240\n",
            "building tree 220 of 240\n",
            "building tree 221 of 240\n",
            "building tree 222 of 240\n",
            "building tree 223 of 240\n",
            "building tree 224 of 240\n",
            "building tree 225 of 240\n",
            "building tree 226 of 240\n",
            "building tree 227 of 240\n",
            "building tree 228 of 240\n",
            "building tree 229 of 240\n",
            "building tree 230 of 240\n",
            "building tree 231 of 240\n",
            "building tree 232 of 240\n",
            "building tree 233 of 240\n",
            "building tree 234 of 240\n",
            "building tree 235 of 240\n",
            "building tree 236 of 240\n",
            "building tree 237 of 240\n",
            "building tree 238 of 240\n",
            "building tree 239 of 240\n",
            "building tree 240 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.4s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n",
            "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 2 concurrent workers.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 1 of 240\n",
            "building tree 2 of 240\n",
            "building tree 3 of 240\n",
            "building tree 4 of 240building tree 5 of 240\n",
            "\n",
            "building tree 6 of 240\n",
            "building tree 7 of 240building tree 8 of 240\n",
            "\n",
            "building tree 9 of 240\n",
            "building tree 10 of 240\n",
            "building tree 11 of 240\n",
            "building tree 12 of 240\n",
            "building tree 13 of 240\n",
            "building tree 14 of 240\n",
            "building tree 15 of 240\n",
            "building tree 16 of 240\n",
            "building tree 17 of 240\n",
            "building tree 18 of 240\n",
            "building tree 19 of 240\n",
            "building tree 20 of 240\n",
            "building tree 21 of 240\n",
            "building tree 22 of 240\n",
            "building tree 23 of 240\n",
            "building tree 24 of 240building tree 25 of 240\n",
            "building tree 26 of 240\n",
            "\n",
            "building tree 27 of 240\n",
            "building tree 28 of 240\n",
            "building tree 29 of 240\n",
            "building tree 30 of 240\n",
            "building tree 31 of 240\n",
            "building tree 32 of 240\n",
            "building tree 33 of 240\n",
            "building tree 34 of 240\n",
            "building tree 35 of 240\n",
            "building tree 36 of 240\n",
            "building tree 37 of 240\n",
            "building tree 38 of 240\n",
            "building tree 39 of 240\n",
            "building tree 40 of 240\n",
            "building tree 41 of 240\n",
            "building tree 42 of 240\n",
            "building tree 43 of 240\n",
            "building tree 44 of 240\n",
            "building tree 45 of 240\n",
            "building tree 46 of 240\n",
            "building tree 47 of 240\n",
            "building tree 48 of 240\n",
            "building tree 49 of 240\n",
            "building tree 50 of 240\n",
            "building tree 51 of 240building tree 52 of 240\n",
            "\n",
            "building tree 53 of 240\n",
            "building tree 54 of 240\n",
            "building tree 55 of 240\n",
            "building tree 56 of 240\n",
            "building tree 57 of 240\n",
            "building tree 58 of 240\n",
            "building tree 59 of 240\n",
            "building tree 60 of 240\n",
            "building tree 61 of 240\n",
            "building tree 62 of 240building tree 63 of 240\n",
            "building tree 64 of 240\n",
            "building tree 65 of 240\n",
            "building tree 66 of 240\n",
            "building tree 67 of 240\n",
            "building tree 68 of 240\n",
            "\n",
            "building tree 69 of 240\n",
            "building tree 70 of 240\n",
            "building tree 71 of 240\n",
            "building tree 72 of 240\n",
            "building tree 73 of 240\n",
            "building tree 74 of 240\n",
            "building tree 75 of 240\n",
            "building tree 76 of 240\n",
            "building tree 77 of 240building tree 78 of 240\n",
            "building tree 79 of 240\n",
            "\n",
            "building tree 80 of 240\n",
            "building tree 81 of 240\n",
            "building tree 82 of 240\n",
            "building tree 83 of 240building tree 84 of 240\n",
            "\n",
            "building tree 85 of 240\n",
            "building tree 86 of 240\n",
            "building tree 87 of 240\n",
            "building tree 88 of 240\n",
            "building tree 89 of 240\n",
            "building tree 90 of 240\n",
            "building tree 91 of 240\n",
            "building tree 92 of 240building tree 93 of 240\n",
            "\n",
            "building tree 94 of 240building tree 95 of 240\n",
            "building tree 96 of 240\n",
            "\n",
            "building tree 97 of 240\n",
            "building tree 98 of 240\n",
            "building tree 99 of 240\n",
            "building tree 100 of 240\n",
            "building tree 101 of 240building tree 102 of 240\n",
            "building tree 103 of 240\n",
            "\n",
            "building tree 104 of 240building tree 105 of 240\n",
            "building tree 106 of 240\n",
            "building tree 107 of 240\n",
            "\n",
            "building tree 108 of 240\n",
            "building tree 109 of 240building tree 110 of 240\n",
            "\n",
            "building tree 111 of 240\n",
            "building tree 112 of 240\n",
            "building tree 113 of 240\n",
            "building tree 114 of 240\n",
            "building tree 115 of 240\n",
            "building tree 116 of 240\n",
            "building tree 117 of 240\n",
            "building tree 118 of 240\n",
            "building tree 119 of 240\n",
            "building tree 120 of 240\n",
            "building tree 121 of 240\n",
            "building tree 122 of 240\n",
            "building tree 123 of 240\n",
            "building tree 124 of 240\n",
            "building tree 125 of 240\n",
            "building tree 126 of 240\n",
            "building tree 127 of 240\n",
            "building tree 128 of 240\n",
            "building tree 129 of 240\n",
            "building tree 130 of 240building tree 131 of 240\n",
            "\n",
            "building tree 132 of 240\n",
            "building tree 133 of 240\n",
            "building tree 134 of 240\n",
            "building tree 135 of 240\n",
            "building tree 136 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done  37 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=-1)]: Done 158 tasks      | elapsed:    0.2s\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 137 of 240\n",
            "building tree 138 of 240building tree 139 of 240\n",
            "\n",
            "building tree 140 of 240\n",
            "building tree 141 of 240\n",
            "building tree 142 of 240\n",
            "building tree 143 of 240\n",
            "building tree 144 of 240\n",
            "building tree 145 of 240\n",
            "building tree 146 of 240\n",
            "building tree 147 of 240\n",
            "building tree 148 of 240\n",
            "building tree 149 of 240\n",
            "building tree 150 of 240\n",
            "building tree 151 of 240\n",
            "building tree 152 of 240building tree 153 of 240\n",
            "\n",
            "building tree 154 of 240\n",
            "building tree 155 of 240\n",
            "building tree 156 of 240\n",
            "building tree 157 of 240\n",
            "building tree 158 of 240\n",
            "building tree 159 of 240\n",
            "building tree 160 of 240\n",
            "building tree 161 of 240\n",
            "building tree 162 of 240\n",
            "building tree 163 of 240\n",
            "building tree 164 of 240\n",
            "building tree 165 of 240\n",
            "building tree 166 of 240building tree 167 of 240\n",
            "\n",
            "building tree 168 of 240\n",
            "building tree 169 of 240\n",
            "building tree 170 of 240\n",
            "building tree 171 of 240\n",
            "building tree 172 of 240\n",
            "building tree 173 of 240\n",
            "building tree 174 of 240building tree 175 of 240\n",
            "\n",
            "building tree 176 of 240\n",
            "building tree 177 of 240\n",
            "building tree 178 of 240\n",
            "building tree 179 of 240\n",
            "building tree 180 of 240\n",
            "building tree 181 of 240building tree 182 of 240\n",
            "\n",
            "building tree 183 of 240building tree 184 of 240\n",
            "\n",
            "building tree 185 of 240\n",
            "building tree 186 of 240\n",
            "building tree 187 of 240\n",
            "building tree 188 of 240\n",
            "building tree 189 of 240\n",
            "building tree 190 of 240\n",
            "building tree 191 of 240\n",
            "building tree 192 of 240\n",
            "building tree 193 of 240building tree 194 of 240\n",
            "building tree 195 of 240\n",
            "building tree 196 of 240\n",
            "\n",
            "building tree 197 of 240\n",
            "building tree 198 of 240\n",
            "building tree 199 of 240\n",
            "building tree 200 of 240\n",
            "building tree 201 of 240\n",
            "building tree 202 of 240\n",
            "building tree 203 of 240\n",
            "building tree 204 of 240\n",
            "building tree 205 of 240\n",
            "building tree 206 of 240\n",
            "building tree 207 of 240\n",
            "building tree 208 of 240\n",
            "building tree 209 of 240\n",
            "building tree 210 of 240\n",
            "building tree 211 of 240\n",
            "building tree 212 of 240building tree 213 of 240\n",
            "\n",
            "building tree 214 of 240\n",
            "building tree 215 of 240\n",
            "building tree 216 of 240\n",
            "building tree 217 of 240\n",
            "building tree 218 of 240building tree 219 of 240\n",
            "\n",
            "building tree 220 of 240\n",
            "building tree 221 of 240\n",
            "building tree 222 of 240\n",
            "building tree 223 of 240\n",
            "building tree 224 of 240\n",
            "building tree 225 of 240building tree 226 of 240\n",
            "\n",
            "building tree 227 of 240\n",
            "building tree 228 of 240\n",
            "building tree 229 of 240\n",
            "building tree 230 of 240\n",
            "building tree 231 of 240\n",
            "building tree 232 of 240\n",
            "building tree 233 of 240\n",
            "building tree 234 of 240\n",
            "building tree 235 of 240\n",
            "building tree 236 of 240\n",
            "building tree 237 of 240\n",
            "building tree 238 of 240\n",
            "building tree 239 of 240\n",
            "building tree 240 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.4s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n",
            "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 2 concurrent workers.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 1 of 240\n",
            "building tree 2 of 240\n",
            "building tree 3 of 240\n",
            "building tree 4 of 240\n",
            "building tree 5 of 240\n",
            "building tree 6 of 240building tree 7 of 240\n",
            "\n",
            "building tree 8 of 240\n",
            "building tree 9 of 240\n",
            "building tree 10 of 240\n",
            "building tree 11 of 240\n",
            "building tree 12 of 240\n",
            "building tree 13 of 240building tree 14 of 240\n",
            "\n",
            "building tree 15 of 240\n",
            "building tree 16 of 240\n",
            "building tree 17 of 240\n",
            "building tree 18 of 240\n",
            "building tree 19 of 240\n",
            "building tree 20 of 240\n",
            "building tree 21 of 240\n",
            "building tree 22 of 240\n",
            "building tree 23 of 240\n",
            "building tree 24 of 240\n",
            "building tree 25 of 240\n",
            "building tree 26 of 240\n",
            "building tree 27 of 240\n",
            "building tree 28 of 240\n",
            "building tree 29 of 240\n",
            "building tree 30 of 240\n",
            "building tree 31 of 240\n",
            "building tree 32 of 240\n",
            "building tree 33 of 240\n",
            "building tree 34 of 240\n",
            "building tree 35 of 240\n",
            "building tree 36 of 240\n",
            "building tree 37 of 240building tree 38 of 240\n",
            "\n",
            "building tree 39 of 240\n",
            "building tree 40 of 240\n",
            "building tree 41 of 240\n",
            "building tree 42 of 240\n",
            "building tree 43 of 240\n",
            "building tree 44 of 240\n",
            "building tree 45 of 240\n",
            "building tree 46 of 240\n",
            "building tree 47 of 240\n",
            "building tree 48 of 240\n",
            "building tree 49 of 240\n",
            "building tree 50 of 240\n",
            "building tree 51 of 240\n",
            "building tree 52 of 240\n",
            "building tree 53 of 240\n",
            "building tree 54 of 240\n",
            "building tree 55 of 240\n",
            "building tree 56 of 240\n",
            "building tree 57 of 240\n",
            "building tree 58 of 240\n",
            "building tree 59 of 240\n",
            "building tree 60 of 240\n",
            "building tree 61 of 240\n",
            "building tree 62 of 240\n",
            "building tree 63 of 240\n",
            "building tree 64 of 240\n",
            "building tree 65 of 240\n",
            "building tree 66 of 240building tree 67 of 240\n",
            "\n",
            "building tree 68 of 240\n",
            "building tree 69 of 240\n",
            "building tree 70 of 240\n",
            "building tree 71 of 240\n",
            "building tree 72 of 240\n",
            "building tree 73 of 240\n",
            "building tree 74 of 240\n",
            "building tree 75 of 240\n",
            "building tree 76 of 240\n",
            "building tree 77 of 240\n",
            "building tree 78 of 240\n",
            "building tree 79 of 240\n",
            "building tree 80 of 240\n",
            "building tree 81 of 240\n",
            "building tree 82 of 240\n",
            "building tree 83 of 240\n",
            "building tree 84 of 240\n",
            "building tree 85 of 240\n",
            "building tree 86 of 240\n",
            "building tree 87 of 240\n",
            "building tree 88 of 240\n",
            "building tree 89 of 240building tree 90 of 240\n",
            "\n",
            "building tree 91 of 240\n",
            "building tree 92 of 240\n",
            "building tree 93 of 240\n",
            "building tree 94 of 240\n",
            "building tree 95 of 240\n",
            "building tree 96 of 240\n",
            "building tree 97 of 240\n",
            "building tree 98 of 240\n",
            "building tree 99 of 240\n",
            "building tree 100 of 240\n",
            "building tree 101 of 240\n",
            "building tree 102 of 240\n",
            "building tree 103 of 240\n",
            "building tree 104 of 240\n",
            "building tree 105 of 240\n",
            "building tree 106 of 240\n",
            "building tree 107 of 240\n",
            "building tree 108 of 240\n",
            "building tree 109 of 240building tree 110 of 240\n",
            "\n",
            "building tree 111 of 240\n",
            "building tree 112 of 240\n",
            "building tree 113 of 240\n",
            "building tree 114 of 240\n",
            "building tree 115 of 240\n",
            "building tree 116 of 240building tree 117 of 240\n",
            "\n",
            "building tree 118 of 240\n",
            "building tree 119 of 240\n",
            "building tree 120 of 240\n",
            "building tree 121 of 240\n",
            "building tree 122 of 240\n",
            "building tree 123 of 240\n",
            "building tree 124 of 240\n",
            "building tree 125 of 240\n",
            "building tree 126 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done  37 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=-1)]: Done 158 tasks      | elapsed:    0.3s\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 127 of 240building tree 128 of 240\n",
            "building tree 129 of 240\n",
            "\n",
            "building tree 130 of 240\n",
            "building tree 131 of 240building tree 132 of 240\n",
            "\n",
            "building tree 133 of 240\n",
            "building tree 134 of 240\n",
            "building tree 135 of 240\n",
            "building tree 136 of 240\n",
            "building tree 137 of 240\n",
            "building tree 138 of 240\n",
            "building tree 139 of 240\n",
            "building tree 140 of 240\n",
            "building tree 141 of 240\n",
            "building tree 142 of 240\n",
            "building tree 143 of 240\n",
            "building tree 144 of 240\n",
            "building tree 145 of 240building tree 146 of 240\n",
            "\n",
            "building tree 147 of 240\n",
            "building tree 148 of 240\n",
            "building tree 149 of 240\n",
            "building tree 150 of 240\n",
            "building tree 151 of 240\n",
            "building tree 152 of 240\n",
            "building tree 153 of 240\n",
            "building tree 154 of 240\n",
            "building tree 155 of 240\n",
            "building tree 156 of 240\n",
            "building tree 157 of 240\n",
            "building tree 158 of 240\n",
            "building tree 159 of 240\n",
            "building tree 160 of 240building tree 161 of 240\n",
            "\n",
            "building tree 162 of 240\n",
            "building tree 163 of 240\n",
            "building tree 164 of 240\n",
            "building tree 165 of 240\n",
            "building tree 166 of 240building tree 167 of 240\n",
            "\n",
            "building tree 168 of 240\n",
            "building tree 169 of 240\n",
            "building tree 170 of 240\n",
            "building tree 171 of 240\n",
            "building tree 172 of 240\n",
            "building tree 173 of 240\n",
            "building tree 174 of 240\n",
            "building tree 175 of 240\n",
            "building tree 176 of 240\n",
            "building tree 177 of 240\n",
            "building tree 178 of 240\n",
            "building tree 179 of 240\n",
            "building tree 180 of 240\n",
            "building tree 181 of 240\n",
            "building tree 182 of 240building tree 183 of 240\n",
            "building tree 184 of 240\n",
            "building tree 185 of 240\n",
            "building tree 186 of 240\n",
            "building tree 187 of 240\n",
            "building tree 188 of 240\n",
            "building tree 189 of 240\n",
            "building tree 190 of 240\n",
            "building tree 191 of 240\n",
            "building tree 192 of 240\n",
            "building tree 193 of 240\n",
            "building tree 194 of 240\n",
            "building tree 195 of 240\n",
            "building tree 196 of 240\n",
            "\n",
            "building tree 197 of 240\n",
            "building tree 198 of 240\n",
            "building tree 199 of 240\n",
            "building tree 200 of 240\n",
            "building tree 201 of 240building tree 202 of 240\n",
            "building tree 203 of 240\n",
            "\n",
            "building tree 204 of 240\n",
            "building tree 205 of 240\n",
            "building tree 206 of 240\n",
            "building tree 207 of 240\n",
            "building tree 208 of 240\n",
            "building tree 209 of 240\n",
            "building tree 210 of 240\n",
            "building tree 211 of 240\n",
            "building tree 212 of 240building tree 213 of 240\n",
            "building tree 214 of 240\n",
            "\n",
            "building tree 215 of 240\n",
            "building tree 216 of 240\n",
            "building tree 217 of 240\n",
            "building tree 218 of 240\n",
            "building tree 219 of 240\n",
            "building tree 220 of 240\n",
            "building tree 221 of 240\n",
            "building tree 222 of 240\n",
            "building tree 223 of 240\n",
            "building tree 224 of 240\n",
            "building tree 225 of 240\n",
            "building tree 226 of 240\n",
            "building tree 227 of 240\n",
            "building tree 228 of 240\n",
            "building tree 229 of 240\n",
            "building tree 230 of 240\n",
            "building tree 231 of 240\n",
            "building tree 232 of 240\n",
            "building tree 233 of 240\n",
            "building tree 234 of 240\n",
            "building tree 235 of 240\n",
            "building tree 236 of 240\n",
            "building tree 237 of 240\n",
            "building tree 238 of 240\n",
            "building tree 239 of 240\n",
            "building tree 240 of 240\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.4s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n",
            "[Parallel(n_jobs=-1)]: Using backend ThreadingBackend with 2 concurrent workers.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 1 of 240building tree 2 of 240\n",
            "\n",
            "building tree 3 of 240\n",
            "building tree 4 of 240\n",
            "building tree 5 of 240building tree 6 of 240\n",
            "\n",
            "building tree 7 of 240\n",
            "building tree 8 of 240\n",
            "building tree 9 of 240\n",
            "building tree 10 of 240\n",
            "building tree 11 of 240\n",
            "building tree 12 of 240\n",
            "building tree 13 of 240building tree 14 of 240\n",
            "building tree 15 of 240\n",
            "\n",
            "building tree 16 of 240building tree 17 of 240\n",
            "\n",
            "building tree 18 of 240\n",
            "building tree 19 of 240\n",
            "building tree 20 of 240\n",
            "building tree 21 of 240\n",
            "building tree 22 of 240\n",
            "building tree 23 of 240\n",
            "building tree 24 of 240\n",
            "building tree 25 of 240\n",
            "building tree 26 of 240\n",
            "building tree 27 of 240\n",
            "building tree 28 of 240\n",
            "building tree 29 of 240\n",
            "building tree 30 of 240building tree 31 of 240\n",
            "\n",
            "building tree 32 of 240\n",
            "building tree 33 of 240\n",
            "building tree 34 of 240\n",
            "building tree 35 of 240\n",
            "building tree 36 of 240\n",
            "building tree 37 of 240\n",
            "building tree 38 of 240\n",
            "building tree 39 of 240\n",
            "building tree 40 of 240building tree 41 of 240\n",
            "\n",
            "building tree 42 of 240\n",
            "building tree 43 of 240\n",
            "building tree 44 of 240\n",
            "building tree 45 of 240\n",
            "building tree 46 of 240\n",
            "building tree 47 of 240\n",
            "building tree 48 of 240\n",
            "building tree 49 of 240\n",
            "building tree 50 of 240\n",
            "building tree 51 of 240\n",
            "building tree 52 of 240\n",
            "building tree 53 of 240\n",
            "building tree 54 of 240\n",
            "building tree 55 of 240\n",
            "building tree 56 of 240\n",
            "building tree 57 of 240\n",
            "building tree 58 of 240\n",
            "building tree 59 of 240\n",
            "building tree 60 of 240\n",
            "building tree 61 of 240\n",
            "building tree 62 of 240\n",
            "building tree 63 of 240building tree 64 of 240\n",
            "\n",
            "building tree 65 of 240building tree 66 of 240\n",
            "building tree 67 of 240\n",
            "\n",
            "building tree 68 of 240\n",
            "building tree 69 of 240\n",
            "building tree 70 of 240\n",
            "building tree 71 of 240\n",
            "building tree 72 of 240\n",
            "building tree 73 of 240\n",
            "building tree 74 of 240building tree 75 of 240\n",
            "\n",
            "building tree 76 of 240\n",
            "building tree 77 of 240\n",
            "building tree 78 of 240\n",
            "building tree 79 of 240\n",
            "building tree 80 of 240building tree 81 of 240\n",
            "\n",
            "building tree 82 of 240building tree 83 of 240\n",
            "\n",
            "building tree 84 of 240\n",
            "building tree 85 of 240\n",
            "building tree 86 of 240\n",
            "building tree 87 of 240\n",
            "building tree 88 of 240\n",
            "building tree 89 of 240\n",
            "building tree 90 of 240\n",
            "building tree 91 of 240\n",
            "building tree 92 of 240\n",
            "building tree 93 of 240\n",
            "building tree 94 of 240building tree 95 of 240\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done  37 tasks      | elapsed:    0.1s\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "building tree 96 of 240\n",
            "building tree 97 of 240\n",
            "building tree 98 of 240\n",
            "building tree 99 of 240\n",
            "building tree 100 of 240\n",
            "building tree 101 of 240\n",
            "building tree 102 of 240\n",
            "building tree 103 of 240\n",
            "building tree 104 of 240\n",
            "building tree 105 of 240\n",
            "building tree 106 of 240\n",
            "building tree 107 of 240\n",
            "building tree 108 of 240\n",
            "building tree 109 of 240\n",
            "building tree 110 of 240\n",
            "building tree 111 of 240\n",
            "building tree 112 of 240\n",
            "building tree 113 of 240\n",
            "building tree 114 of 240\n",
            "building tree 115 of 240\n",
            "building tree 116 of 240\n",
            "building tree 117 of 240\n",
            "building tree 118 of 240\n",
            "building tree 119 of 240\n",
            "building tree 120 of 240\n",
            "building tree 121 of 240\n",
            "building tree 122 of 240building tree 123 of 240\n",
            "\n",
            "building tree 124 of 240\n",
            "building tree 125 of 240\n",
            "building tree 126 of 240\n",
            "building tree 127 of 240\n",
            "building tree 128 of 240\n",
            "building tree 129 of 240\n",
            "building tree 130 of 240\n",
            "building tree 131 of 240building tree 132 of 240\n",
            "\n",
            "building tree 133 of 240\n",
            "building tree 134 of 240\n",
            "building tree 135 of 240\n",
            "building tree 136 of 240\n",
            "building tree 137 of 240\n",
            "building tree 138 of 240building tree 139 of 240\n",
            "\n",
            "building tree 140 of 240\n",
            "building tree 141 of 240\n",
            "building tree 142 of 240\n",
            "building tree 143 of 240\n",
            "building tree 144 of 240\n",
            "building tree 145 of 240\n",
            "building tree 146 of 240\n",
            "building tree 147 of 240\n",
            "building tree 148 of 240\n",
            "building tree 149 of 240\n",
            "building tree 150 of 240\n",
            "building tree 151 of 240\n",
            "building tree 152 of 240\n",
            "building tree 153 of 240\n",
            "building tree 154 of 240\n",
            "building tree 155 of 240\n",
            "building tree 156 of 240\n",
            "building tree 157 of 240\n",
            "building tree 158 of 240\n",
            "building tree 159 of 240building tree 160 of 240\n",
            "building tree 161 of 240\n",
            "building tree 162 of 240\n",
            "building tree 163 of 240\n",
            "building tree 164 of 240\n",
            "building tree 165 of 240\n",
            "building tree 166 of 240\n",
            "building tree 167 of 240\n",
            "building tree 168 of 240\n",
            "building tree 169 of 240\n",
            "\n",
            "building tree 170 of 240\n",
            "building tree 171 of 240building tree 172 of 240\n",
            "building tree 173 of 240\n",
            "building tree 174 of 240\n",
            "building tree 175 of 240\n",
            "\n",
            "building tree 176 of 240\n",
            "building tree 177 of 240building tree 178 of 240\n",
            "building tree 179 of 240\n",
            "building tree 180 of 240\n",
            "building tree 181 of 240\n",
            "\n",
            "building tree 182 of 240\n",
            "building tree 183 of 240\n",
            "building tree 184 of 240building tree 185 of 240\n",
            "building tree 186 of 240\n",
            "\n",
            "building tree 187 of 240\n",
            "building tree 188 of 240\n",
            "building tree 189 of 240\n",
            "building tree 190 of 240\n",
            "building tree 191 of 240\n",
            "building tree 192 of 240\n",
            "building tree 193 of 240\n",
            "building tree 194 of 240\n",
            "building tree 195 of 240\n",
            "building tree 196 of 240\n",
            "building tree 197 of 240\n",
            "building tree 198 of 240\n",
            "building tree 199 of 240\n",
            "building tree 200 of 240\n",
            "building tree 201 of 240\n",
            "building tree 202 of 240\n",
            "building tree 203 of 240\n",
            "building tree 204 of 240\n",
            "building tree 205 of 240building tree 206 of 240\n",
            "\n",
            "building tree 207 of 240\n",
            "building tree 208 of 240\n",
            "building tree 209 of 240\n",
            "building tree 210 of 240\n",
            "building tree 211 of 240\n",
            "building tree 212 of 240\n",
            "building tree 213 of 240\n",
            "building tree 214 of 240\n",
            "building tree 215 of 240\n",
            "building tree 216 of 240\n",
            "building tree 217 of 240\n",
            "building tree 218 of 240\n",
            "building tree 219 of 240\n",
            "building tree 220 of 240"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Done 158 tasks      | elapsed:    0.3s\n",
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.4s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.0s finished\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "building tree 221 of 240building tree 222 of 240\n",
            "building tree 223 of 240\n",
            "building tree 224 of 240\n",
            "building tree 225 of 240\n",
            "\n",
            "building tree 226 of 240\n",
            "building tree 227 of 240\n",
            "building tree 228 of 240building tree 229 of 240\n",
            "\n",
            "building tree 230 of 240\n",
            "building tree 231 of 240\n",
            "building tree 232 of 240\n",
            "building tree 233 of 240\n",
            "building tree 234 of 240\n",
            "building tree 235 of 240\n",
            "building tree 236 of 240\n",
            "building tree 237 of 240\n",
            "building tree 238 of 240\n",
            "building tree 239 of 240building tree 240 of 240\n",
            "\n",
            "Cross validated accuracy: 0.6333333333333334\n",
            "Cross validated accuracy (std): 0.11027462911821782\n",
            "Cross validated AUC: 0.6294141492322607\n",
            "Cross validated AUC (std): 0.10889570450471715\n",
            "Cross validated sensitivity: 0.698477421271539\n",
            "Cross validated sensitivity (std): 0.09023617182408873\n",
            "Cross validated specificity: 0.5603508771929825\n",
            "Cross validated specificity (std): 0.15505675645509975\n",
            "Cross validated PPV: 0.6255134464853135\n",
            "Cross validated PPV (std): 0.12666875110302042\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot ROC curve\n",
        "\n",
        "y_pred_prob = cross_val_predict(rf_best, X, y, cv=cv, method='predict_proba')\n",
        "fpr, tpr, thresholds = roc_curve(y, y_pred_prob[:, 1])\n",
        "plt.plot(fpr, tpr)\n",
        "plt.xlabel('False Positive Rate')\n",
        "plt.ylabel('True Positive Rate')\n",
        "plt.title('ROC Curve')\n",
        "plt.show()\n",
        "#plt.savefig(\"/content/drive/My Drive/Colab Notebooks/co_mitochondria/14.prediction.roc.pdf\")\n",
        "\n",
        "\n",
        "# Print feature importances\n",
        "mean_importances = np.mean(important_features, axis=0)\n",
        "feature_names = X.columns\n",
        "sorted_idx = mean_importances.argsort()\n",
        "for i in sorted_idx:\n",
        "    print(feature_names[i], mean_importances[i])\n",
        "\n",
        "plt.bar(X.columns, mean_importances)\n",
        "plt.title('Feature Importance')\n",
        "plt.xlabel('Feature')\n",
        "plt.ylabel('Importance')\n",
        "plt.show()\n",
        "#plt.savefig(\"/content/drive/My Drive/Colab Notebooks/co_mitochondria/14.prediction.feature.pdf\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "nG1q_NIWV1W1",
        "outputId": "a2919051-4e62-4186-8834-06e518a7f45f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done 206 tasks      | elapsed:    1.2s\n",
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    1.3s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.2s finished\n",
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done 218 tasks      | elapsed:    0.4s\n",
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.4s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n",
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done 218 tasks      | elapsed:    0.5s\n",
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.6s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n",
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.5s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.2s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.2s finished\n",
            "[Parallel(n_jobs=-1)]: Using backend LokyBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=-1)]: Done 218 tasks      | elapsed:    0.8s\n",
            "[Parallel(n_jobs=-1)]: Done 240 out of 240 | elapsed:    0.9s finished\n",
            "[Parallel(n_jobs=2)]: Using backend ThreadingBackend with 2 concurrent workers.\n",
            "[Parallel(n_jobs=2)]: Done  37 tasks      | elapsed:    0.0s\n",
            "[Parallel(n_jobs=2)]: Done 158 tasks      | elapsed:    0.1s\n",
            "[Parallel(n_jobs=2)]: Done 240 out of 240 | elapsed:    0.1s finished\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "dm 0.002917148666334672\n",
            "htn 0.029474542466854346\n",
            "sex 0.03914525636335033\n",
            "sample_age.log2 0.4516031479914494\n",
            "mtcn.log2 0.4768599045120112\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "saved_model = pickle.dumps(rf_best)\n",
        "saved_model\n",
        "\n",
        "joblib.dump(rf_best, '/content/drive/My Drive/Colab Notebooks/co_mitochondria/14.prediction_rf_best.pkl')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bWphwVaGRXmu",
        "outputId": "e16379b7-d4ca-4331-9422-bfbd586ed0eb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['/content/drive/My Drive/Colab Notebooks/co_mitochondria/14.prediction_rf_best.pkl']"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "- mtcn 이 가장 중요한 feature \n",
        "- AUC = 0.63 +/- 0.11"
      ],
      "metadata": {
        "id": "c_YCg9PPRARR"
      }
    }
  ]
}