QoS_Prediction_Challenge_by_ITU_AI_ML_in_5G_Challenge.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "e2IXWdcTJPf1"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from typing import Union, List, Literal, Dict\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.cm as cm"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ddt5s0HfNk-y"
      },
      "outputs": [],
      "source": [
        "from Helper_Util import *"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "sjE1IVVvnbYo"
      },
      "outputs": [],
      "source": [
        "Filepath = '/content/drive/MyDrive/QoS Prediction Challenge by ITU ML in 5G Challenge'\n",
        "pd.set_option('display.max_colwidth',200)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "dvQAuRDKRLuH"
      },
      "outputs": [],
      "source": [
        "def Load_Data(file_path: str) -> List[pd.DataFrame]:\n",
        "  train = pd.read_csv(file_path+'/Train.csv')\n",
        "  test = pd.read_csv(file_path+'/Test.csv')\n",
        "  sub = pd.read_csv(file_path+'/SampleSubmission.csv')\n",
        "  variables = pd.read_csv(file_path+'/VariableDefinitions.csv')\n",
        "\n",
        "  return train, test, sub, variables\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2FIz8CAonNHC"
      },
      "outputs": [],
      "source": [
        "Train, Test, Sub, Variables = Load_Data(Filepath)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "seUcMl_BqUyz"
      },
      "source": [
        "### Data Exploration"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "id": "TYbkTAUEnTY1",
        "outputId": "41992617-8fd8-46eb-f994-0d540b829000"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-c311872d-8c3d-4f15-ba46-e022972d5198\">\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>id</th>\n",
              "      <th>timestamp</th>\n",
              "      <th>device</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>...</th>\n",
              "      <th>dewPoint</th>\n",
              "      <th>humidity</th>\n",
              "      <th>pressure</th>\n",
              "      <th>windSpeed</th>\n",
              "      <th>cloudCover</th>\n",
              "      <th>uvIndex</th>\n",
              "      <th>visibility</th>\n",
              "      <th>Traffic Jam Factor</th>\n",
              "      <th>area</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Id_qt2qn56050</td>\n",
              "      <td>1624367008</td>\n",
              "      <td>pc1</td>\n",
              "      <td>-84.498750</td>\n",
              "      <td>-14.586875</td>\n",
              "      <td>-49.855625</td>\n",
              "      <td>16.457</td>\n",
              "      <td>35076.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.86</td>\n",
              "      <td>0.62</td>\n",
              "      <td>1013.6</td>\n",
              "      <td>2.39</td>\n",
              "      <td>0.95</td>\n",
              "      <td>4.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>0.38979</td>\n",
              "      <td>Residential</td>\n",
              "      <td>51500000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Id_f1zf07cwb6</td>\n",
              "      <td>1624372465</td>\n",
              "      <td>pc3</td>\n",
              "      <td>-86.818125</td>\n",
              "      <td>-11.982500</td>\n",
              "      <td>-54.838125</td>\n",
              "      <td>16.674</td>\n",
              "      <td>42704.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.52</td>\n",
              "      <td>0.59</td>\n",
              "      <td>1013.5</td>\n",
              "      <td>2.29</td>\n",
              "      <td>0.82</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>3.78322</td>\n",
              "      <td>Residential</td>\n",
              "      <td>50300000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Id_uoxdz7di5b</td>\n",
              "      <td>1624371871</td>\n",
              "      <td>pc3</td>\n",
              "      <td>-101.916250</td>\n",
              "      <td>-13.257500</td>\n",
              "      <td>-71.241250</td>\n",
              "      <td>9.392</td>\n",
              "      <td>36017.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.71</td>\n",
              "      <td>0.61</td>\n",
              "      <td>1013.8</td>\n",
              "      <td>2.13</td>\n",
              "      <td>0.82</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.08425</td>\n",
              "      <td>Residential</td>\n",
              "      <td>16500000.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3 rows × 41 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c311872d-8c3d-4f15-ba46-e022972d5198')\"\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-c311872d-8c3d-4f15-ba46-e022972d5198 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-c311872d-8c3d-4f15-ba46-e022972d5198');\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",
              "  "
            ],
            "text/plain": [
              "              id   timestamp device  PCell_RSRP_max  PCell_RSRQ_max  \\\n",
              "0  Id_qt2qn56050  1624367008    pc1      -84.498750      -14.586875   \n",
              "1  Id_f1zf07cwb6  1624372465    pc3      -86.818125      -11.982500   \n",
              "2  Id_uoxdz7di5b  1624371871    pc3     -101.916250      -13.257500   \n",
              "\n",
              "   PCell_RSSI_max  PCell_SNR_max  PCell_Downlink_Num_RBs  \\\n",
              "0      -49.855625         16.457                 35076.0   \n",
              "1      -54.838125         16.674                 42704.0   \n",
              "2      -71.241250          9.392                 36017.0   \n",
              "\n",
              "   PCell_Downlink_Average_MCS  PCell_Downlink_bandwidth_MHz  ...  dewPoint  \\\n",
              "0                        23.0                          20.0  ...     13.86   \n",
              "1                        24.0                          20.0  ...     14.52   \n",
              "2                        22.0                          20.0  ...     14.71   \n",
              "\n",
              "   humidity  pressure  windSpeed  cloudCover  uvIndex  visibility  \\\n",
              "0      0.62    1013.6       2.39        0.95      4.0      16.093   \n",
              "1      0.59    1013.5       2.29        0.82      3.0      16.093   \n",
              "2      0.61    1013.8       2.13        0.82      3.0      16.093   \n",
              "\n",
              "   Traffic Jam Factor         area      target  \n",
              "0             0.38979  Residential  51500000.0  \n",
              "1             3.78322  Residential  50300000.0  \n",
              "2             2.08425  Residential  16500000.0  \n",
              "\n",
              "[3 rows x 41 columns]"
            ]
          },
          "execution_count": 7,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "Train.head(3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "id": "BwNVwCAlpX5F",
        "outputId": "4507a43f-bc90-4aa4-95b3-d28a7935defc"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-46fc50c4-124a-40ed-b990-5fcdf49e91da\">\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>id</th>\n",
              "      <th>timestamp</th>\n",
              "      <th>device</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>...</th>\n",
              "      <th>apparentTemperature</th>\n",
              "      <th>dewPoint</th>\n",
              "      <th>humidity</th>\n",
              "      <th>pressure</th>\n",
              "      <th>windSpeed</th>\n",
              "      <th>cloudCover</th>\n",
              "      <th>uvIndex</th>\n",
              "      <th>visibility</th>\n",
              "      <th>Traffic Jam Factor</th>\n",
              "      <th>area</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Id_ln0e0hfrgx</td>\n",
              "      <td>1624374179</td>\n",
              "      <td>pc2</td>\n",
              "      <td>-97.121875</td>\n",
              "      <td>-12.923750</td>\n",
              "      <td>-87.554375</td>\n",
              "      <td>13.293</td>\n",
              "      <td>39276.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>20</td>\n",
              "      <td>...</td>\n",
              "      <td>23.05</td>\n",
              "      <td>14.55</td>\n",
              "      <td>0.59</td>\n",
              "      <td>1013.6</td>\n",
              "      <td>2.27</td>\n",
              "      <td>0.83</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>3.07702</td>\n",
              "      <td>Avenue</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Id_svf7nz9fxv</td>\n",
              "      <td>1624366264</td>\n",
              "      <td>pc4</td>\n",
              "      <td>-65.202500</td>\n",
              "      <td>-9.944375</td>\n",
              "      <td>-35.501875</td>\n",
              "      <td>21.823</td>\n",
              "      <td>39566.0</td>\n",
              "      <td>25.0</td>\n",
              "      <td>20</td>\n",
              "      <td>...</td>\n",
              "      <td>21.56</td>\n",
              "      <td>13.92</td>\n",
              "      <td>0.62</td>\n",
              "      <td>1013.6</td>\n",
              "      <td>2.42</td>\n",
              "      <td>0.95</td>\n",
              "      <td>4.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>0.66000</td>\n",
              "      <td>Park</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Id_ww2mh07gwj</td>\n",
              "      <td>1624375795</td>\n",
              "      <td>pc2</td>\n",
              "      <td>-79.298125</td>\n",
              "      <td>-17.468750</td>\n",
              "      <td>-74.823750</td>\n",
              "      <td>24.114</td>\n",
              "      <td>41900.0</td>\n",
              "      <td>27.0</td>\n",
              "      <td>20</td>\n",
              "      <td>...</td>\n",
              "      <td>22.93</td>\n",
              "      <td>14.65</td>\n",
              "      <td>0.60</td>\n",
              "      <td>1013.5</td>\n",
              "      <td>2.26</td>\n",
              "      <td>0.85</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.49330</td>\n",
              "      <td>Park</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3 rows × 40 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-46fc50c4-124a-40ed-b990-5fcdf49e91da')\"\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-46fc50c4-124a-40ed-b990-5fcdf49e91da 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-46fc50c4-124a-40ed-b990-5fcdf49e91da');\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",
              "  "
            ],
            "text/plain": [
              "              id   timestamp device  PCell_RSRP_max  PCell_RSRQ_max  \\\n",
              "0  Id_ln0e0hfrgx  1624374179    pc2      -97.121875      -12.923750   \n",
              "1  Id_svf7nz9fxv  1624366264    pc4      -65.202500       -9.944375   \n",
              "2  Id_ww2mh07gwj  1624375795    pc2      -79.298125      -17.468750   \n",
              "\n",
              "   PCell_RSSI_max  PCell_SNR_max  PCell_Downlink_Num_RBs  \\\n",
              "0      -87.554375         13.293                 39276.0   \n",
              "1      -35.501875         21.823                 39566.0   \n",
              "2      -74.823750         24.114                 41900.0   \n",
              "\n",
              "   PCell_Downlink_Average_MCS  PCell_Downlink_bandwidth_MHz  ...  \\\n",
              "0                        24.0                            20  ...   \n",
              "1                        25.0                            20  ...   \n",
              "2                        27.0                            20  ...   \n",
              "\n",
              "   apparentTemperature  dewPoint  humidity  pressure  windSpeed  cloudCover  \\\n",
              "0                23.05     14.55      0.59    1013.6       2.27        0.83   \n",
              "1                21.56     13.92      0.62    1013.6       2.42        0.95   \n",
              "2                22.93     14.65      0.60    1013.5       2.26        0.85   \n",
              "\n",
              "   uvIndex  visibility  Traffic Jam Factor    area  \n",
              "0      3.0      16.093             3.07702  Avenue  \n",
              "1      4.0      16.093             0.66000    Park  \n",
              "2      3.0      16.093             2.49330    Park  \n",
              "\n",
              "[3 rows x 40 columns]"
            ]
          },
          "execution_count": 29,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "Test.head(3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ErMajNLspd4G",
        "outputId": "fe145c4c-c288-4125-ec8d-b7d25d8caac6"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Index(['id', 'device', 'area'], dtype='object')"
            ]
          },
          "execution_count": 30,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "Train.select_dtypes(include=['object']).columns\n",
        "Test.select_dtypes(include=['object']).columns"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 696
        },
        "id": "kFW7i6M340j0",
        "outputId": "116e4951-ea17-4c22-ff3b-ce5d48f22dfc"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1000x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "missing_values = Train.isna().sum()/ Train.shape[0]\n",
        "missing_values = missing_values.sort_values(ascending=False)\n",
        "\n",
        "\n",
        "# Normalize the data to be between 0 and 1\n",
        "norm = plt.Normalize(missing_values.values.min(), missing_values.values.max())\n",
        "# Create a colormap\n",
        "cmap = cm.ScalarMappable(norm=norm, cmap='Blues')\n",
        "# Generate colors for each bar\n",
        "colors = [cmap.to_rgba(val) for val in missing_values.values]\n",
        "\n",
        "# Create bar plot of missing values\n",
        "plt.figure(figsize=(10,5))\n",
        "plt.bar(missing_values.index, missing_values.values, color=colors)\n",
        "plt.xticks(rotation=90)\n",
        "plt.xlabel('Column')\n",
        "plt.ylabel('Number of Missing Values')\n",
        "plt.title('Missing Values per Column')\n",
        "plt.show()\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 696
        },
        "id": "7dQ5wi4J5ely",
        "outputId": "8210b5ea-ddd4-45fc-c9a2-bc140caec8f6"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1000x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "missing_values = Test.isna().sum()/ Test.shape[0]\n",
        "missing_values = missing_values.sort_values(ascending=False)\n",
        "import matplotlib.cm as cm\n",
        "\n",
        "# Normalize the data to be between 0 and 1\n",
        "norm = plt.Normalize(missing_values.values.min(), missing_values.values.max())\n",
        "# Create a colormap\n",
        "cmap = cm.ScalarMappable(norm=norm, cmap='Blues')\n",
        "# Generate colors for each bar\n",
        "colors = [cmap.to_rgba(val) for val in missing_values.values]\n",
        "\n",
        "# Create bar plot of missing values\n",
        "plt.figure(figsize=(10,5))\n",
        "plt.bar(missing_values.index, missing_values.values, color=colors)\n",
        "plt.xticks(rotation=90)\n",
        "plt.xlabel('Column')\n",
        "plt.ylabel('Number of Missing Values')\n",
        "plt.title('Missing Values per Column')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1C783Qbr76RJ"
      },
      "source": [
        "SCELLS_ have the most missing values"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FO27LAYe_8Nm"
      },
      "source": [
        "###Baseline Score"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "PaXOhfef_-Gh"
      },
      "outputs": [],
      "source": [
        "from sklearn.metrics import mean_absolute_error, mean_squared_error\n",
        "from sklearn.model_selection import KFold, train_test_split\n",
        "from xgboost import XGBRegressor\n",
        "from lightgbm import LGBMRegressor, Dataset, early_stopping\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.preprocessing import OrdinalEncoder\n",
        "from sklearn.impute import SimpleImputer\n",
        "from sklearn.preprocessing import StandardScaler, MinMaxScaler\n",
        "from abc import ABC, abstractmethod\n",
        "import random"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "dXT99fZIBH8t"
      },
      "outputs": [],
      "source": [
        "!pip install catboost -q\n",
        "from catboost import CatBoostRegressor, Pool"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "adS5r0bfBIj4"
      },
      "outputs": [],
      "source": [
        "Sscaler = StandardScaler()\n",
        "Mmscaler = MinMaxScaler()\n",
        "rfr = RandomForestRegressor()\n",
        "xgb = XGBRegressor()\n",
        "cbc = CatBoostRegressor(eval_metric='RMSE',learning_rate=0.08,depth=9)\n",
        "lgb = LGBMRegressor()\n",
        "mse = mean_squared_error"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "M8l4yOgYDgVw"
      },
      "outputs": [],
      "source": [
        "class Baseline_score():\n",
        "  def __init__(\n",
        "      self,\n",
        "      train: pd.DataFrame,\n",
        "      model: Union[XGBRegressor, LGBMRegressor, CatBoostRegressor,RandomForestRegressor],\n",
        "      scaler: Union[StandardScaler, MinMaxScaler, None] = None,\n",
        "      split_method: Literal[\"KFold\", \"train_test_split\"] = \"KFold\",\n",
        "      return_baseline: Literal[True, False] = True,\n",
        "      num_split = 5,\n",
        "      random_seed = 42,\n",
        "      VERBOSE: int = 0\n",
        "  ):\n",
        "    if not isinstance(model, (XGBRegressor, LGBMRegressor, CatBoostRegressor,RandomForestRegressor)):\n",
        "      raise TypeError(\"model must be a class of either XGBRegressor, LGBMRegressor, CatBoostRegressor, or LinearRegression\")\n",
        "    if scaler is not None and not isinstance(scaler, (StandardScaler, MinMaxScaler)):\n",
        "        raise TypeError(\"scaler must be a class of either StandardScaler, MinMaxScaler or None\")\n",
        "    if not isinstance(train, pd.DataFrame):\n",
        "      raise TypeError(\"train must be a DataFrame\")\n",
        "\n",
        "    self.train = train\n",
        "    self.model = model\n",
        "    self.scaler = scaler\n",
        "    self.split = split_method\n",
        "    self.num_split = num_split\n",
        "    self.seed = random_seed\n",
        "    self.return_baseline = return_baseline\n",
        "    self.VERBOSE = VERBOSE\n",
        "    self.feature_importances = None\n",
        "    self.bsv = {}\n",
        "    random.seed(random_seed)\n",
        "    np.random.seed(random_seed)\n",
        "\n",
        "  def run(self)-> float:\n",
        "    X = self.train.iloc[:,:-1]\n",
        "    self.feature_importances = np.zeros(X.shape[1])\n",
        "    y = self.train.iloc[:,-1]\n",
        "\n",
        "    if self.split == \"train_test_split\":\n",
        "      X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=self.seed)\n",
        "\n",
        "      if self.scaler != None:\n",
        "        X_train, X_test = self.scaler.fit_transform(X_train), self.scaler.fit_transform(X_test)\n",
        "\n",
        "      if isinstance(self.model, LGBMRegressor):\n",
        "        self.model.fit(X_train, y_train, eval_set=(X_test, y_test), callbacks=[early_stopping(stopping_rounds=100)], verbose=self.VERBOSE)\n",
        "\n",
        "      elif isinstance(self.model, CatBoostRegressor):\n",
        "        self.model.fit(X_train,y_train,eval_set=(X_test,y_test),use_best_model=True,early_stopping_rounds=20,verbose=self.VERBOSE)\n",
        "\n",
        "      elif isinstance(self.model, XGBRegressor):\n",
        "        self.model.fit(X_train,y_train, eval_set=[(X_test,y_test)], early_stopping_rounds=20,verbose=self.VERBOSE)\n",
        "\n",
        "      else:\n",
        "        self.model.fit(X_train,y_train)\n",
        "\n",
        "      self.feature_importances += self.model.feature_importances_\n",
        "      if isinstance(self.model, XGBRegressor):\n",
        "        results = self.model.evals_result()\n",
        "        evals_result = self.model.evals_result()\n",
        "        best_scores = {}\n",
        "        for metric in self.model.eval_metric:\n",
        "          best_scores[metric] = min(evals_result['validation_0'][metric])\n",
        "        for metric, score in best_scores.items():\n",
        "          print(f\"Best {metric}: {score}\")\n",
        "      else:\n",
        "        results = self.model.best_score_\n",
        "      preds = self.model.predict(X_test)\n",
        "      train_baseline = self.model.predict(X)\n",
        "      self.bsv.update({'train':train_baseline})\n",
        "\n",
        "      if self.VERBOSE > 0:\n",
        "        print(f\"split: {(self.split).upper()}\")\n",
        "        print(f\"RMSE {mean_squared_error(y_test,preds,squared=False)}\")\n",
        "        print(f\"Model RMSE: {results.values()}\")\n",
        "\n",
        "      score = mean_squared_error(y_test,preds,squared=False)\n",
        "\n",
        "      if self.return_baseline:\n",
        "        return score, self.bsv\n",
        "\n",
        "      return score\n",
        "\n",
        "    else:\n",
        "      fold = KFold(n_splits=self.num_split)\n",
        "      oof_f1 = []\n",
        "      i = 1\n",
        "      for train_index, test_index in fold.split(X, y):\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",
        "        if self.scaler != None:\n",
        "          X_train, X_test = self.scaler.fit_transform(X_train), self.scaler.fit_transform(X_test)\n",
        "\n",
        "        if isinstance(self.model, LGBMRegressor):\n",
        "          self.model.fit(X_train, y_train, eval_set=(X_test, y_test),callbacks=[early_stopping(stopping_rounds=20)],verbose=self.VERBOSE)\n",
        "\n",
        "        elif isinstance(self.model, CatBoostRegressor):\n",
        "          self.model.fit(X_train,y_train,eval_set=(X_test,y_test),use_best_model=True,early_stopping_rounds=20,verbose=self.VERBOSE)\n",
        "\n",
        "        elif isinstance(self.model, XGBRegressor):\n",
        "          self.model.fit(X_train,y_train, eval_set=[(X_test,y_test)], early_stopping_rounds=20,verbose=self.VERBOSE)\n",
        "\n",
        "        else:\n",
        "          self.model.fit(X_train,y_train)\n",
        "\n",
        "        self.feature_importances += self.model.feature_importances_\n",
        "        results = self.model.best_score_\n",
        "        preds=self.model.predict(X_test)\n",
        "        train_baseline = self.model.predict(X)\n",
        "        self.bsv.update({f'{self.split}{i}train':train_baseline})\n",
        "\n",
        "        if self.VERBOSE > 0:\n",
        "          print(f\"split {self.split}{i}\")\n",
        "          print(f\"RMSE {mean_squared_error(y_test,preds,squared=False)}\")\n",
        "          print(f\"Model RMSE: {results.values()}\")\n",
        "        oof_f1.append(mean_squared_error(y_test,preds,squared=False))\n",
        "        i += 1\n",
        "\n",
        "      self.feature_importances = self.feature_importances / self.num_split\n",
        "      score = sum(oof_f1)/self.num_split\n",
        "\n",
        "      if self.return_baseline:\n",
        "        return score, self.bsv\n",
        "\n",
        "      return score\n",
        "\n",
        "  def plot_features(self, n_features=20):\n",
        "    X = self.train.iloc[:,:-1]\n",
        "    features = pd.DataFrame({'feature': list(X.columns),\n",
        "                             'importance': self.feature_importances}).sort_values('importance', ascending = True)\n",
        "    norm = plt.Normalize(features['importance'][:n_features].min(), features['importance'][:n_features].max())\n",
        "    cmap = cm.ScalarMappable(norm=norm, cmap='Reds')\n",
        "    colors = [cmap.to_rgba(val) for val in features['importance'][:n_features].values]\n",
        "    plt.figure(figsize=(25,10))\n",
        "    plt.barh(features['feature'][:n_features],features['importance'][:n_features], color=colors)\n",
        "    plt.xlabel('Importance')\n",
        "    plt.ylabel('Feature')\n",
        "    plt.title('Feature Importances')\n",
        "    plt.show()\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "mfctnAH-VFrD"
      },
      "outputs": [],
      "source": [
        "class Baseline_predict(Baseline_score):\n",
        "  def __init__(self, *args, **kwargs):\n",
        "    super().__init__(*args, **kwargs)\n",
        "    self.preds = None\n",
        "\n",
        "  def predict(self, test):\n",
        "    X = self.train.iloc[:,:-1]\n",
        "    y = self.train.iloc[:,-1]\n",
        "\n",
        "    if self.split == \"train_test_split\":\n",
        "      X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=self.seed)\n",
        "      if self.scaler != None:\n",
        "        X_train, X_test = self.scaler.fit_transform(X_train), self.scaler.fit_transform(X_test)\n",
        "\n",
        "      if isinstance(self.model, LGBMRegressor):\n",
        "        self.model.fit(X_train, y_train, eval_set=(X_test, y_test), eval_metric='rmse', callbacks=[early_stopping(stopping_rounds=100)], verbose=self.VERBOSE)\n",
        "\n",
        "      elif isinstance(self.model, CatBoostRegressor):\n",
        "        self.model.fit(X_train,y_train,eval_set=(X_test,y_test),use_best_model=True,early_stopping_rounds=20,verbose=self.VERBOSE)\n",
        "\n",
        "      elif isinstance(self.model, XGBRegressor):\n",
        "        self.model.fit(X_train,y_train, eval_set=(X_test,y_test), early_stopping_rounds=20,verbose=self.VERBOSE)\n",
        "\n",
        "      else:\n",
        "        self.model.fit(X_train,y_train)\n",
        "\n",
        "      train_baseline = self.model.predict(test)\n",
        "      self.bsv.update({'train':train_baseline})\n",
        "      self.preds = self.model.predict(test)\n",
        "      print(f\"split: {(self.split).upper()}\")\n",
        "\n",
        "      if self.return_baseline:\n",
        "        return self.preds, self.bsv\n",
        "\n",
        "      return self.preds\n",
        "\n",
        "    else:\n",
        "      fold = KFold(n_splits=self.num_split)\n",
        "      oof_preds = []\n",
        "      i = 1\n",
        "      for train_index, test_index in fold.split(X, y):\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",
        "        if self.scaler != None:\n",
        "          X_train, X_test = self.scaler.fit_transform(X_train), self.scaler.fit_transform(X_test)\n",
        "\n",
        "        if isinstance(self.model, LGBMRegressor):\n",
        "          self.model.fit(X_train, y_train, eval_set=(X_test, y_test), eval_metric='rmse', callbacks=[early_stopping(stopping_rounds=100)], verbose=self.VERBOSE)\n",
        "\n",
        "        elif isinstance(self.model, CatBoostRegressor):\n",
        "          self.model.fit(X_train,y_train,eval_set=(X_test,y_test),use_best_model=True,early_stopping_rounds=20,verbose=self.VERBOSE)\n",
        "\n",
        "        elif isinstance(self.model, XGBRegressor):\n",
        "         self.model.fit(X_train,y_train, eval_set=(X_test,y_test), early_stopping_rounds=20,verbose=self.VERBOSE)\n",
        "\n",
        "        else:\n",
        "          self.model.fit(X_train,y_train)\n",
        "\n",
        "        pred=self.model.predict(test)\n",
        "        train_baseline = self.model.predict(test)\n",
        "        self.bsv.update({f'{self.split}{i}train':train_baseline})\n",
        "        if self.VERBOSE > 0:\n",
        "          print(f\"split {self.split}{i}\")\n",
        "        oof_preds.append(pred)\n",
        "        i += 1\n",
        "\n",
        "      self.preds = oof_preds\n",
        "\n",
        "      if self.return_baseline:\n",
        "        return self.preds, self.bsv\n",
        "\n",
        "      return self.preds\n",
        "\n",
        "  def submission(self, test):\n",
        "    pres = self.preds\n",
        "    if self.split == 'KFold':\n",
        "      final_preds = sum(pres)/self.num_split\n",
        "    else:\n",
        "      final_preds = pres\n",
        "    predictions_df = pd.DataFrame({'id': test.id, 'target': final_preds})\n",
        "    return predictions_df\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "AtYuE_l7idjd"
      },
      "outputs": [],
      "source": [
        "TT = Train.drop(['id','device'],axis=1)\n",
        "tests = Test.drop(['id','device'], axis=1)\n",
        "#TT['device'] = TT['device'].map({'pc1':1,'pc3':2})\n",
        "#TT['area'] = TT['area'].map({'Residential':1,'Avenue':2,'Park':3,'Highway':4,'Tunnel':5})\n",
        "categorical_features = ['area']\n",
        "oe = OrdinalEncoder()\n",
        "tests['area'] = oe.fit_transform(tests[['area']])\n",
        "TT['area'] = oe.fit_transform(TT[['area']])\n",
        "tests.fillna(0, inplace=True)\n",
        "TT.fillna(0, inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "TT.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "id": "C68j_a8BlGaz",
        "outputId": "a08d8ea4-bd22-4ffa-9395-1525f4e6bbfa"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    timestamp  PCell_RSRP_max  PCell_RSRQ_max  PCell_RSSI_max  PCell_SNR_max  \\\n",
              "0  1624367008      -84.498750      -14.586875      -49.855625         16.457   \n",
              "1  1624372465      -86.818125      -11.982500      -54.838125         16.674   \n",
              "2  1624371871     -101.916250      -13.257500      -71.241250          9.392   \n",
              "\n",
              "   PCell_Downlink_Num_RBs  PCell_Downlink_Average_MCS  \\\n",
              "0                 35076.0                        23.0   \n",
              "1                 42704.0                        24.0   \n",
              "2                 36017.0                        22.0   \n",
              "\n",
              "   PCell_Downlink_bandwidth_MHz  PCell_Cell_Identity  PCell_freq_MHz  ...  \\\n",
              "0                          20.0           26368258.0          1800.0  ...   \n",
              "1                          20.0            2568970.0          1800.0  ...   \n",
              "2                          20.0            2824458.0          1800.0  ...   \n",
              "\n",
              "   dewPoint  humidity  pressure  windSpeed  cloudCover  uvIndex  visibility  \\\n",
              "0     13.86      0.62    1013.6       2.39        0.95      4.0      16.093   \n",
              "1     14.52      0.59    1013.5       2.29        0.82      3.0      16.093   \n",
              "2     14.71      0.61    1013.8       2.13        0.82      3.0      16.093   \n",
              "\n",
              "   Traffic Jam Factor  area      target  \n",
              "0             0.38979   3.0  51500000.0  \n",
              "1             3.78322   3.0  50300000.0  \n",
              "2             2.08425   3.0  16500000.0  \n",
              "\n",
              "[3 rows x 39 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e20f109d-ea53-46ce-a23b-cc66babdd783\">\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>timestamp</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>PCell_Cell_Identity</th>\n",
              "      <th>PCell_freq_MHz</th>\n",
              "      <th>...</th>\n",
              "      <th>dewPoint</th>\n",
              "      <th>humidity</th>\n",
              "      <th>pressure</th>\n",
              "      <th>windSpeed</th>\n",
              "      <th>cloudCover</th>\n",
              "      <th>uvIndex</th>\n",
              "      <th>visibility</th>\n",
              "      <th>Traffic Jam Factor</th>\n",
              "      <th>area</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1624367008</td>\n",
              "      <td>-84.498750</td>\n",
              "      <td>-14.586875</td>\n",
              "      <td>-49.855625</td>\n",
              "      <td>16.457</td>\n",
              "      <td>35076.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>26368258.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.86</td>\n",
              "      <td>0.62</td>\n",
              "      <td>1013.6</td>\n",
              "      <td>2.39</td>\n",
              "      <td>0.95</td>\n",
              "      <td>4.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>0.38979</td>\n",
              "      <td>3.0</td>\n",
              "      <td>51500000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1624372465</td>\n",
              "      <td>-86.818125</td>\n",
              "      <td>-11.982500</td>\n",
              "      <td>-54.838125</td>\n",
              "      <td>16.674</td>\n",
              "      <td>42704.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>2568970.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.52</td>\n",
              "      <td>0.59</td>\n",
              "      <td>1013.5</td>\n",
              "      <td>2.29</td>\n",
              "      <td>0.82</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>3.78322</td>\n",
              "      <td>3.0</td>\n",
              "      <td>50300000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1624371871</td>\n",
              "      <td>-101.916250</td>\n",
              "      <td>-13.257500</td>\n",
              "      <td>-71.241250</td>\n",
              "      <td>9.392</td>\n",
              "      <td>36017.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>2824458.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.71</td>\n",
              "      <td>0.61</td>\n",
              "      <td>1013.8</td>\n",
              "      <td>2.13</td>\n",
              "      <td>0.82</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.08425</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16500000.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3 rows × 39 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e20f109d-ea53-46ce-a23b-cc66babdd783')\"\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-e20f109d-ea53-46ce-a23b-cc66babdd783 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-e20f109d-ea53-46ce-a23b-cc66babdd783');\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": 140
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "TT_1 = TT.sample(frac=1)\n",
        "TT_1.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "id": "GLihJECAln11",
        "outputId": "1a65a156-f05a-47e5-ce4a-61ac5bdc4992"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        timestamp  PCell_RSRP_max  PCell_RSRQ_max  PCell_RSSI_max  \\\n",
              "28894  1624373796      -74.082500      -12.478125      -40.353750   \n",
              "13917  1624542070      -97.730625      -15.350000      -62.291875   \n",
              "2663   1624354131      -81.581250      -10.125000      -53.066250   \n",
              "\n",
              "       PCell_SNR_max  PCell_Downlink_Num_RBs  PCell_Downlink_Average_MCS  \\\n",
              "28894         21.533                 38435.0                        28.0   \n",
              "13917          7.538                 96188.0                        14.0   \n",
              "2663          10.036                 94606.0                        19.0   \n",
              "\n",
              "       PCell_Downlink_bandwidth_MHz  PCell_Cell_Identity  PCell_freq_MHz  ...  \\\n",
              "28894                          20.0            3428105.0          1800.0  ...   \n",
              "13917                          20.0           26367489.0          1800.0  ...   \n",
              "2663                           20.0           30827264.0          1800.0  ...   \n",
              "\n",
              "       dewPoint  humidity  pressure  windSpeed  cloudCover  uvIndex  \\\n",
              "28894     14.52      0.59    1013.5       2.29        0.82      3.0   \n",
              "13917     13.43      0.71    1017.9       1.97        0.99      3.0   \n",
              "2663      13.90      0.73    1012.2       4.23        0.97      4.0   \n",
              "\n",
              "       visibility  Traffic Jam Factor  area      target  \n",
              "28894      16.093             8.05296   1.0  25600000.0  \n",
              "13917      16.093             8.80702   0.0  33200000.0  \n",
              "2663       16.093             2.61657   2.0  84600000.0  \n",
              "\n",
              "[3 rows x 39 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-85f729e1-e0a6-4558-a065-fe837ec80b10\">\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>timestamp</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>PCell_Cell_Identity</th>\n",
              "      <th>PCell_freq_MHz</th>\n",
              "      <th>...</th>\n",
              "      <th>dewPoint</th>\n",
              "      <th>humidity</th>\n",
              "      <th>pressure</th>\n",
              "      <th>windSpeed</th>\n",
              "      <th>cloudCover</th>\n",
              "      <th>uvIndex</th>\n",
              "      <th>visibility</th>\n",
              "      <th>Traffic Jam Factor</th>\n",
              "      <th>area</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>28894</th>\n",
              "      <td>1624373796</td>\n",
              "      <td>-74.082500</td>\n",
              "      <td>-12.478125</td>\n",
              "      <td>-40.353750</td>\n",
              "      <td>21.533</td>\n",
              "      <td>38435.0</td>\n",
              "      <td>28.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>3428105.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.52</td>\n",
              "      <td>0.59</td>\n",
              "      <td>1013.5</td>\n",
              "      <td>2.29</td>\n",
              "      <td>0.82</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>8.05296</td>\n",
              "      <td>1.0</td>\n",
              "      <td>25600000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13917</th>\n",
              "      <td>1624542070</td>\n",
              "      <td>-97.730625</td>\n",
              "      <td>-15.350000</td>\n",
              "      <td>-62.291875</td>\n",
              "      <td>7.538</td>\n",
              "      <td>96188.0</td>\n",
              "      <td>14.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>26367489.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.43</td>\n",
              "      <td>0.71</td>\n",
              "      <td>1017.9</td>\n",
              "      <td>1.97</td>\n",
              "      <td>0.99</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>8.80702</td>\n",
              "      <td>0.0</td>\n",
              "      <td>33200000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2663</th>\n",
              "      <td>1624354131</td>\n",
              "      <td>-81.581250</td>\n",
              "      <td>-10.125000</td>\n",
              "      <td>-53.066250</td>\n",
              "      <td>10.036</td>\n",
              "      <td>94606.0</td>\n",
              "      <td>19.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>30827264.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.90</td>\n",
              "      <td>0.73</td>\n",
              "      <td>1012.2</td>\n",
              "      <td>4.23</td>\n",
              "      <td>0.97</td>\n",
              "      <td>4.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.61657</td>\n",
              "      <td>2.0</td>\n",
              "      <td>84600000.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3 rows × 39 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-85f729e1-e0a6-4558-a065-fe837ec80b10')\"\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-85f729e1-e0a6-4558-a065-fe837ec80b10 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-85f729e1-e0a6-4558-a065-fe837ec80b10');\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": 141
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "TT_2 = TT.sample(frac=1)\n",
        "TT_2.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "id": "L9oz7SyBlof7",
        "outputId": "f5b9640e-8d05-4dfc-ab55-8ad5e8226bad"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        timestamp  PCell_RSRP_max  PCell_RSRQ_max  PCell_RSSI_max  \\\n",
              "16174  1624356228      -91.250625      -17.616250      -56.321875   \n",
              "24901  1624374131      -76.191875      -12.397500      -45.368125   \n",
              "19241  1624552824      -86.796717      -13.160354      -55.238005   \n",
              "\n",
              "       PCell_SNR_max  PCell_Downlink_Num_RBs  PCell_Downlink_Average_MCS  \\\n",
              "16174       0.817000                 91118.0                        19.0   \n",
              "24901      13.759000                 46681.0                        23.0   \n",
              "19241       6.793939                 11743.0                        25.0   \n",
              "\n",
              "       PCell_Downlink_bandwidth_MHz  PCell_Cell_Identity  PCell_freq_MHz  ...  \\\n",
              "16174                          20.0           40812803.0          1800.0  ...   \n",
              "24901                          20.0            3499534.0          2600.0  ...   \n",
              "19241                          15.0            4051733.0          2000.0  ...   \n",
              "\n",
              "       dewPoint  humidity  pressure  windSpeed  cloudCover  uvIndex  \\\n",
              "16174     13.46      0.69    1012.5       4.09        0.96      4.0   \n",
              "24901     14.55      0.59    1013.6       2.27        0.83      3.0   \n",
              "19241     13.80      0.73    1017.4       2.43        1.00      1.0   \n",
              "\n",
              "       visibility  Traffic Jam Factor  area      target  \n",
              "16174      16.093             3.14159   1.0  46000000.0  \n",
              "24901      16.093             4.50427   0.0  36800000.0  \n",
              "19241      16.093             0.99534   3.0   2720000.0  \n",
              "\n",
              "[3 rows x 39 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-cb65e104-187b-40ab-998c-32cc9d37f218\">\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>timestamp</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>PCell_Cell_Identity</th>\n",
              "      <th>PCell_freq_MHz</th>\n",
              "      <th>...</th>\n",
              "      <th>dewPoint</th>\n",
              "      <th>humidity</th>\n",
              "      <th>pressure</th>\n",
              "      <th>windSpeed</th>\n",
              "      <th>cloudCover</th>\n",
              "      <th>uvIndex</th>\n",
              "      <th>visibility</th>\n",
              "      <th>Traffic Jam Factor</th>\n",
              "      <th>area</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>16174</th>\n",
              "      <td>1624356228</td>\n",
              "      <td>-91.250625</td>\n",
              "      <td>-17.616250</td>\n",
              "      <td>-56.321875</td>\n",
              "      <td>0.817000</td>\n",
              "      <td>91118.0</td>\n",
              "      <td>19.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>40812803.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.46</td>\n",
              "      <td>0.69</td>\n",
              "      <td>1012.5</td>\n",
              "      <td>4.09</td>\n",
              "      <td>0.96</td>\n",
              "      <td>4.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>3.14159</td>\n",
              "      <td>1.0</td>\n",
              "      <td>46000000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24901</th>\n",
              "      <td>1624374131</td>\n",
              "      <td>-76.191875</td>\n",
              "      <td>-12.397500</td>\n",
              "      <td>-45.368125</td>\n",
              "      <td>13.759000</td>\n",
              "      <td>46681.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>3499534.0</td>\n",
              "      <td>2600.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.55</td>\n",
              "      <td>0.59</td>\n",
              "      <td>1013.6</td>\n",
              "      <td>2.27</td>\n",
              "      <td>0.83</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>4.50427</td>\n",
              "      <td>0.0</td>\n",
              "      <td>36800000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19241</th>\n",
              "      <td>1624552824</td>\n",
              "      <td>-86.796717</td>\n",
              "      <td>-13.160354</td>\n",
              "      <td>-55.238005</td>\n",
              "      <td>6.793939</td>\n",
              "      <td>11743.0</td>\n",
              "      <td>25.0</td>\n",
              "      <td>15.0</td>\n",
              "      <td>4051733.0</td>\n",
              "      <td>2000.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.80</td>\n",
              "      <td>0.73</td>\n",
              "      <td>1017.4</td>\n",
              "      <td>2.43</td>\n",
              "      <td>1.00</td>\n",
              "      <td>1.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>0.99534</td>\n",
              "      <td>3.0</td>\n",
              "      <td>2720000.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3 rows × 39 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cb65e104-187b-40ab-998c-32cc9d37f218')\"\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-cb65e104-187b-40ab-998c-32cc9d37f218 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-cb65e104-187b-40ab-998c-32cc9d37f218');\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": 142
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "TT_3 = TT.sample(frac=1)\n",
        "TT_3.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "id": "J6vx5mY6lop5",
        "outputId": "058a4b39-9858-4c80-dbf4-e823927ce3c6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        timestamp  PCell_RSRP_max  PCell_RSRQ_max  PCell_RSSI_max  \\\n",
              "32537  1624348482      -76.865000        -13.8175      -46.056250   \n",
              "1107   1624373140      -85.646875        -12.2125      -52.740000   \n",
              "10247  1624374198      -89.426250        -13.2775      -56.091875   \n",
              "\n",
              "       PCell_SNR_max  PCell_Downlink_Num_RBs  PCell_Downlink_Average_MCS  \\\n",
              "32537         19.504                 80969.0                        26.0   \n",
              "1107          17.522                 41521.0                        25.0   \n",
              "10247          9.816                 32413.0                        17.0   \n",
              "\n",
              "       PCell_Downlink_bandwidth_MHz  PCell_Cell_Identity  PCell_freq_MHz  ...  \\\n",
              "32537                          20.0           33802242.0          1800.0  ...   \n",
              "1107                           20.0            2568970.0          1800.0  ...   \n",
              "10247                          20.0            3499534.0          2600.0  ...   \n",
              "\n",
              "       dewPoint  humidity  pressure  windSpeed  cloudCover  uvIndex  \\\n",
              "32537     14.06      0.77    1011.9       4.07        0.97      3.0   \n",
              "1107      14.52      0.59    1013.5       2.29        0.82      3.0   \n",
              "10247     14.55      0.59    1013.6       2.27        0.83      3.0   \n",
              "\n",
              "       visibility  Traffic Jam Factor  area       target  \n",
              "32537      16.093             2.99305   0.0  141000000.0  \n",
              "1107       16.093             3.00946   3.0   59000000.0  \n",
              "10247      16.093             2.86572   0.0   12800000.0  \n",
              "\n",
              "[3 rows x 39 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3518fccf-b0d8-45a2-9d0b-8d56da56f0b3\">\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>timestamp</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>PCell_Cell_Identity</th>\n",
              "      <th>PCell_freq_MHz</th>\n",
              "      <th>...</th>\n",
              "      <th>dewPoint</th>\n",
              "      <th>humidity</th>\n",
              "      <th>pressure</th>\n",
              "      <th>windSpeed</th>\n",
              "      <th>cloudCover</th>\n",
              "      <th>uvIndex</th>\n",
              "      <th>visibility</th>\n",
              "      <th>Traffic Jam Factor</th>\n",
              "      <th>area</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>32537</th>\n",
              "      <td>1624348482</td>\n",
              "      <td>-76.865000</td>\n",
              "      <td>-13.8175</td>\n",
              "      <td>-46.056250</td>\n",
              "      <td>19.504</td>\n",
              "      <td>80969.0</td>\n",
              "      <td>26.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>33802242.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.06</td>\n",
              "      <td>0.77</td>\n",
              "      <td>1011.9</td>\n",
              "      <td>4.07</td>\n",
              "      <td>0.97</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.99305</td>\n",
              "      <td>0.0</td>\n",
              "      <td>141000000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1107</th>\n",
              "      <td>1624373140</td>\n",
              "      <td>-85.646875</td>\n",
              "      <td>-12.2125</td>\n",
              "      <td>-52.740000</td>\n",
              "      <td>17.522</td>\n",
              "      <td>41521.0</td>\n",
              "      <td>25.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>2568970.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.52</td>\n",
              "      <td>0.59</td>\n",
              "      <td>1013.5</td>\n",
              "      <td>2.29</td>\n",
              "      <td>0.82</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>3.00946</td>\n",
              "      <td>3.0</td>\n",
              "      <td>59000000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10247</th>\n",
              "      <td>1624374198</td>\n",
              "      <td>-89.426250</td>\n",
              "      <td>-13.2775</td>\n",
              "      <td>-56.091875</td>\n",
              "      <td>9.816</td>\n",
              "      <td>32413.0</td>\n",
              "      <td>17.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>3499534.0</td>\n",
              "      <td>2600.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.55</td>\n",
              "      <td>0.59</td>\n",
              "      <td>1013.6</td>\n",
              "      <td>2.27</td>\n",
              "      <td>0.83</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.86572</td>\n",
              "      <td>0.0</td>\n",
              "      <td>12800000.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3 rows × 39 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3518fccf-b0d8-45a2-9d0b-8d56da56f0b3')\"\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-3518fccf-b0d8-45a2-9d0b-8d56da56f0b3 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-3518fccf-b0d8-45a2-9d0b-8d56da56f0b3');\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": 143
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "TT_4 = TT.sample(frac=1)\n",
        "TT_4.head(3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "id": "fGN5jJH2lovq",
        "outputId": "9f2c3e21-7546-4d2a-8eb1-220cc16eb3aa"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "        timestamp  PCell_RSRP_max  PCell_RSRQ_max  PCell_RSSI_max  \\\n",
              "20275  1624552652      -67.468125      -13.708125      -32.853125   \n",
              "2336   1624377851      -91.986250      -12.253125      -59.430000   \n",
              "14864  1624365428      -97.362500      -15.338125      -61.638125   \n",
              "\n",
              "       PCell_SNR_max  PCell_Downlink_Num_RBs  PCell_Downlink_Average_MCS  \\\n",
              "20275         25.419                 81888.0                        28.0   \n",
              "2336          11.900                 40137.0                        19.0   \n",
              "14864          5.629                 47635.0                        13.0   \n",
              "\n",
              "       PCell_Downlink_bandwidth_MHz  PCell_Cell_Identity  PCell_freq_MHz  ...  \\\n",
              "20275                           0.0                  0.0          1800.0  ...   \n",
              "2336                           20.0            3499534.0          2600.0  ...   \n",
              "14864                          20.0           26367489.0          1800.0  ...   \n",
              "\n",
              "       dewPoint  humidity  pressure  windSpeed  cloudCover  uvIndex  \\\n",
              "20275     13.80      0.73    1017.4       2.43        1.00      1.0   \n",
              "2336      14.78      0.57    1013.7       2.21        0.76      2.0   \n",
              "14864     13.91      0.62    1013.6       2.41        0.95      4.0   \n",
              "\n",
              "       visibility  Traffic Jam Factor  area       target  \n",
              "20275      16.093             3.16090   3.0  164000000.0  \n",
              "2336       16.093             2.38578   0.0   21200000.0  \n",
              "14864      16.093             2.82027   0.0   13900000.0  \n",
              "\n",
              "[3 rows x 39 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f6318bf7-bd43-40b8-8372-d3ab24a95237\">\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>timestamp</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>PCell_Cell_Identity</th>\n",
              "      <th>PCell_freq_MHz</th>\n",
              "      <th>...</th>\n",
              "      <th>dewPoint</th>\n",
              "      <th>humidity</th>\n",
              "      <th>pressure</th>\n",
              "      <th>windSpeed</th>\n",
              "      <th>cloudCover</th>\n",
              "      <th>uvIndex</th>\n",
              "      <th>visibility</th>\n",
              "      <th>Traffic Jam Factor</th>\n",
              "      <th>area</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>20275</th>\n",
              "      <td>1624552652</td>\n",
              "      <td>-67.468125</td>\n",
              "      <td>-13.708125</td>\n",
              "      <td>-32.853125</td>\n",
              "      <td>25.419</td>\n",
              "      <td>81888.0</td>\n",
              "      <td>28.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.80</td>\n",
              "      <td>0.73</td>\n",
              "      <td>1017.4</td>\n",
              "      <td>2.43</td>\n",
              "      <td>1.00</td>\n",
              "      <td>1.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>3.16090</td>\n",
              "      <td>3.0</td>\n",
              "      <td>164000000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2336</th>\n",
              "      <td>1624377851</td>\n",
              "      <td>-91.986250</td>\n",
              "      <td>-12.253125</td>\n",
              "      <td>-59.430000</td>\n",
              "      <td>11.900</td>\n",
              "      <td>40137.0</td>\n",
              "      <td>19.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>3499534.0</td>\n",
              "      <td>2600.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.78</td>\n",
              "      <td>0.57</td>\n",
              "      <td>1013.7</td>\n",
              "      <td>2.21</td>\n",
              "      <td>0.76</td>\n",
              "      <td>2.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.38578</td>\n",
              "      <td>0.0</td>\n",
              "      <td>21200000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14864</th>\n",
              "      <td>1624365428</td>\n",
              "      <td>-97.362500</td>\n",
              "      <td>-15.338125</td>\n",
              "      <td>-61.638125</td>\n",
              "      <td>5.629</td>\n",
              "      <td>47635.0</td>\n",
              "      <td>13.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>26367489.0</td>\n",
              "      <td>1800.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.91</td>\n",
              "      <td>0.62</td>\n",
              "      <td>1013.6</td>\n",
              "      <td>2.41</td>\n",
              "      <td>0.95</td>\n",
              "      <td>4.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.82027</td>\n",
              "      <td>0.0</td>\n",
              "      <td>13900000.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3 rows × 39 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f6318bf7-bd43-40b8-8372-d3ab24a95237')\"\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-f6318bf7-bd43-40b8-8372-d3ab24a95237 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-f6318bf7-bd43-40b8-8372-d3ab24a95237');\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": 144
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "cbc = CatBoostRegressor(loss_function='RMSE',custom_metric=['RMSE','MAE','Huber:delta=0.5'],learning_rate=0.08,depth=9,n_estimators=5000)\n",
        "# cbc1 = CatBoostRegressor(learning_rate=0.08,eval_metric='RMSE',n_estimators=5000)"
      ],
      "metadata": {
        "id": "aIUU5dWJArFE"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Baseline = Baseline_score(train=TT,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=200,random_seed=0)\n",
        "lgb_score,bsv_1 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "Baseline = Baseline_score(train=TT_1,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=200,random_seed=0)\n",
        "lgb_score,bsv_2 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "Baseline = Baseline_score(train=TT_2,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=200,random_seed=0)\n",
        "lgb_score,bsv_3 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "Baseline = Baseline_score(train=TT_3,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=200,random_seed=0)\n",
        "lgb_score,bsv_4 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "Baseline = Baseline_score(train=TT_4,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=200,random_seed=0)\n",
        "lgb_score,bsv_5 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l_s9GSEm7wcN",
        "outputId": "9bab7244-03dc-4ecf-f393-454e35c19327"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0:\tlearn: 40442650.1597919\ttest: 40441243.4452648\tbest: 40441243.4452648 (0)\ttotal: 188ms\tremaining: 15m 38s\n",
            "200:\tlearn: 7864657.8762682\ttest: 9831207.3598372\tbest: 9831207.3598372 (200)\ttotal: 15.2s\tremaining: 6m 4s\n",
            "400:\tlearn: 6484143.7447131\ttest: 9445960.8545335\tbest: 9445960.8545335 (400)\ttotal: 24.4s\tremaining: 4m 39s\n",
            "600:\tlearn: 5692807.1389969\ttest: 9320480.4933492\tbest: 9320480.4933492 (600)\ttotal: 35s\tremaining: 4m 16s\n",
            "800:\tlearn: 5112673.1114639\ttest: 9243319.7181046\tbest: 9243319.7181046 (800)\ttotal: 45.7s\tremaining: 3m 59s\n",
            "1000:\tlearn: 4658555.7147118\ttest: 9201158.1505979\tbest: 9201158.1505979 (1000)\ttotal: 54.4s\tremaining: 3m 37s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9186934.688\n",
            "bestIteration = 1099\n",
            "\n",
            "Shrink model to first 1100 iterations.\n",
            "split KFold1\n",
            "RMSE 9186934.688067216\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1579839.4083854882, 'MAE': 3159679.0667699724, 'RMSE': 4433983.505169107}, {'Huber:delta=0.5': 2759037.675235765, 'MAE': 5518075.600470534, 'RMSE': 9186934.688067203}])\n",
            "0:\tlearn: 40473805.7437831\ttest: 40271488.8786360\tbest: 40271488.8786360 (0)\ttotal: 66.3ms\tremaining: 5m 31s\n",
            "200:\tlearn: 7845668.4969783\ttest: 9751838.8026630\tbest: 9751838.8026630 (200)\ttotal: 10.4s\tremaining: 4m 7s\n",
            "400:\tlearn: 6474378.3461882\ttest: 9309507.6012492\tbest: 9309507.6012492 (400)\ttotal: 19.3s\tremaining: 3m 41s\n",
            "600:\tlearn: 5660838.6185916\ttest: 9148955.1385313\tbest: 9148955.1385313 (600)\ttotal: 30s\tremaining: 3m 39s\n",
            "800:\tlearn: 5086168.5881962\ttest: 9090231.6391622\tbest: 9090231.6391622 (800)\ttotal: 40.7s\tremaining: 3m 33s\n",
            "1000:\tlearn: 4654698.7480549\ttest: 9047005.0051045\tbest: 9047005.0051045 (1000)\ttotal: 49.4s\tremaining: 3m 17s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9035868.223\n",
            "bestIteration = 1067\n",
            "\n",
            "Shrink model to first 1068 iterations.\n",
            "split KFold2\n",
            "RMSE 9035868.22254153\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1601618.8126866312, 'MAE': 3203237.8753722645, 'RMSE': 4494526.212280235}, {'Huber:delta=0.5': 2772946.9702646467, 'MAE': 5545894.190528294, 'RMSE': 9035868.222541532}])\n",
            "0:\tlearn: 40360571.9452612\ttest: 40724643.5977347\tbest: 40724643.5977347 (0)\ttotal: 57.9ms\tremaining: 4m 49s\n",
            "200:\tlearn: 7916862.9398097\ttest: 9862129.0972868\tbest: 9862129.0972868 (200)\ttotal: 8.92s\tremaining: 3m 32s\n",
            "400:\tlearn: 6492588.9649827\ttest: 9398217.4812695\tbest: 9398217.4812695 (400)\ttotal: 19.1s\tremaining: 3m 38s\n",
            "600:\tlearn: 5693806.7716452\ttest: 9253671.6029231\tbest: 9251473.8857336 (597)\ttotal: 29.7s\tremaining: 3m 37s\n",
            "800:\tlearn: 5120544.3123558\ttest: 9187178.2239650\tbest: 9187101.9294249 (799)\ttotal: 40.4s\tremaining: 3m 31s\n",
            "1000:\tlearn: 4693404.3102289\ttest: 9145352.8459478\tbest: 9145352.8459478 (1000)\ttotal: 49.5s\tremaining: 3m 17s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9143670.115\n",
            "bestIteration = 1012\n",
            "\n",
            "Shrink model to first 1013 iterations.\n",
            "split KFold3\n",
            "RMSE 9143670.114888072\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1643793.5439462366, 'MAE': 3287587.337891472, 'RMSE': 4630517.769028484}, {'Huber:delta=0.5': 2767389.3498237724, 'MAE': 5534778.949646539, 'RMSE': 9143670.114888074}])\n",
            "0:\tlearn: 40620857.8749762\ttest: 39663467.2972414\tbest: 39663467.2972414 (0)\ttotal: 127ms\tremaining: 10m 36s\n",
            "200:\tlearn: 7877663.3269849\ttest: 9639095.1919404\tbest: 9639095.1919404 (200)\ttotal: 10.5s\tremaining: 4m 10s\n",
            "400:\tlearn: 6514678.1743145\ttest: 9170935.9640069\tbest: 9170935.9640069 (400)\ttotal: 21s\tremaining: 4m\n",
            "600:\tlearn: 5699539.0330499\ttest: 9013435.6113557\tbest: 9013435.6113557 (600)\ttotal: 31.6s\tremaining: 3m 51s\n",
            "800:\tlearn: 5105806.2524377\ttest: 8936606.3872422\tbest: 8936606.3872422 (800)\ttotal: 40.1s\tremaining: 3m 30s\n",
            "1000:\tlearn: 4654491.2220602\ttest: 8893342.7524512\tbest: 8892910.4842908 (993)\ttotal: 50.7s\tremaining: 3m 22s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8885340.555\n",
            "bestIteration = 1043\n",
            "\n",
            "Shrink model to first 1044 iterations.\n",
            "split KFold4\n",
            "RMSE 8885340.555365104\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1613023.3669727312, 'MAE': 3226046.983944453, 'RMSE': 4536047.703017956}, {'Huber:delta=0.5': 2693325.223429604, 'MAE': 5386650.696858199, 'RMSE': 8885340.555365087}])\n",
            "0:\tlearn: 40274489.2668160\ttest: 41116263.2092984\tbest: 41116263.2092984 (0)\ttotal: 57.9ms\tremaining: 4m 49s\n",
            "200:\tlearn: 7853189.6085108\ttest: 9406824.4769731\tbest: 9406824.4769731 (200)\ttotal: 11.1s\tremaining: 4m 25s\n",
            "400:\tlearn: 6514001.8982748\ttest: 9085197.6645798\tbest: 9084728.7221287 (399)\ttotal: 21.7s\tremaining: 4m 9s\n",
            "600:\tlearn: 5698246.0355791\ttest: 8981526.2872340\tbest: 8980933.9890382 (599)\ttotal: 30.5s\tremaining: 3m 43s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8951620.628\n",
            "bestIteration = 676\n",
            "\n",
            "Shrink model to first 677 iterations.\n",
            "split KFold5\n",
            "RMSE 8951620.628251718\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1889875.905841212, 'MAE': 3779752.0616814187, 'RMSE': 5400037.609116212}, {'Huber:delta=0.5': 2799611.4518940956, 'MAE': 5599223.153787169, 'RMSE': 8951620.62825171}])\n",
            "CV_RMSE_Score:9040686.841822729\n",
            "0:\tlearn: 40397387.1231714\ttest: 40577592.9061134\tbest: 40577592.9061134 (0)\ttotal: 65.1ms\tremaining: 5m 25s\n",
            "200:\tlearn: 7900878.7331912\ttest: 9520176.5287704\tbest: 9520176.5287704 (200)\ttotal: 9.56s\tremaining: 3m 48s\n",
            "400:\tlearn: 6489912.1599611\ttest: 9079773.0030051\tbest: 9079505.8721727 (399)\ttotal: 19.2s\tremaining: 3m 40s\n",
            "600:\tlearn: 5702287.7732067\ttest: 8934904.6650278\tbest: 8933863.2783608 (595)\ttotal: 29.9s\tremaining: 3m 38s\n",
            "800:\tlearn: 5144317.7944286\ttest: 8856709.4114540\tbest: 8855774.3629081 (797)\ttotal: 40.5s\tremaining: 3m 32s\n",
            "1000:\tlearn: 4693492.8809785\ttest: 8807570.8410454\tbest: 8807366.1927769 (999)\ttotal: 49s\tremaining: 3m 15s\n",
            "1200:\tlearn: 4318461.2483400\ttest: 8783280.7126664\tbest: 8782274.3164174 (1195)\ttotal: 59.7s\tremaining: 3m 9s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8775600.152\n",
            "bestIteration = 1260\n",
            "\n",
            "Shrink model to first 1261 iterations.\n",
            "split KFold1\n",
            "RMSE 8775600.152148586\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1497273.8041114623, 'MAE': 2994547.8582219095, 'RMSE': 4184996.684387541}, {'Huber:delta=0.5': 2707646.499419678, 'MAE': 5415293.248838367, 'RMSE': 8775600.152148576}])\n",
            "0:\tlearn: 40569342.0720277\ttest: 39874075.2844631\tbest: 39874075.2844631 (0)\ttotal: 56.9ms\tremaining: 4m 44s\n",
            "200:\tlearn: 7887726.7094864\ttest: 9446590.2840994\tbest: 9446590.2840994 (200)\ttotal: 10.5s\tremaining: 4m 11s\n",
            "400:\tlearn: 6511306.9772906\ttest: 9088928.6994914\tbest: 9088834.3088597 (398)\ttotal: 21.2s\tremaining: 4m 3s\n",
            "600:\tlearn: 5723310.8951611\ttest: 8991999.7222585\tbest: 8991999.7222585 (600)\ttotal: 30.2s\tremaining: 3m 41s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8943095.765\n",
            "bestIteration = 755\n",
            "\n",
            "Shrink model to first 756 iterations.\n",
            "split KFold2\n",
            "RMSE 8943095.764688136\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1824509.1604847894, 'MAE': 3649018.5709685674, 'RMSE': 5199947.502603887}, {'Huber:delta=0.5': 2741047.8443942005, 'MAE': 5482095.938787399, 'RMSE': 8943095.764688132}])\n",
            "0:\tlearn: 40241989.4629697\ttest: 41258202.5218665\tbest: 41258202.5218665 (0)\ttotal: 58.3ms\tremaining: 4m 51s\n",
            "200:\tlearn: 7837048.8893029\ttest: 9940983.1118280\tbest: 9940983.1118280 (200)\ttotal: 10.8s\tremaining: 4m 16s\n",
            "400:\tlearn: 6440553.8777348\ttest: 9530057.1771207\tbest: 9527518.9853182 (397)\ttotal: 20.6s\tremaining: 3m 56s\n",
            "600:\tlearn: 5661749.8988646\ttest: 9404523.2023007\tbest: 9404523.2023007 (600)\ttotal: 30s\tremaining: 3m 39s\n",
            "800:\tlearn: 5075968.0818963\ttest: 9322942.1029411\tbest: 9322814.2783138 (791)\ttotal: 40.6s\tremaining: 3m 33s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9316837.448\n",
            "bestIteration = 811\n",
            "\n",
            "Shrink model to first 812 iterations.\n",
            "split KFold3\n",
            "RMSE 9316837.44785439\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1768367.3647095456, 'MAE': 3536734.9794180766, 'RMSE': 5001732.772514034}, {'Huber:delta=0.5': 2823275.4664873434, 'MAE': 5646551.182973675, 'RMSE': 9316837.447854389}])\n",
            "0:\tlearn: 40304457.5996275\ttest: 40971927.2801196\tbest: 40971927.2801196 (0)\ttotal: 64.2ms\tremaining: 5m 20s\n",
            "200:\tlearn: 7828630.9760642\ttest: 9692443.2922162\tbest: 9692443.2922162 (200)\ttotal: 10.8s\tremaining: 4m 18s\n",
            "400:\tlearn: 6500567.7276977\ttest: 9290458.6618418\tbest: 9290458.6618418 (400)\ttotal: 20.3s\tremaining: 3m 53s\n",
            "600:\tlearn: 5683671.0941200\ttest: 9139199.0934056\tbest: 9137243.7636454 (591)\ttotal: 29.8s\tremaining: 3m 38s\n",
            "800:\tlearn: 5100997.3916554\ttest: 9072397.1035870\tbest: 9071599.4635013 (794)\ttotal: 40.4s\tremaining: 3m 31s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9045075.306\n",
            "bestIteration = 970\n",
            "\n",
            "Shrink model to first 971 iterations.\n",
            "split KFold4\n",
            "RMSE 9045075.305833943\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1654287.210770301, 'MAE': 3308574.671539603, 'RMSE': 4662990.0368740745}, {'Huber:delta=0.5': 2713675.462580256, 'MAE': 5427351.175159512, 'RMSE': 9045075.305833943}])\n",
            "0:\tlearn: 40654467.6810110\ttest: 39516532.1859034\tbest: 39516532.1859034 (0)\ttotal: 62.1ms\tremaining: 5m 10s\n",
            "200:\tlearn: 7855876.2487323\ttest: 9811109.0650266\tbest: 9811109.0650266 (200)\ttotal: 8.47s\tremaining: 3m 22s\n",
            "400:\tlearn: 6493326.0902492\ttest: 9392675.8336583\tbest: 9392675.8336583 (400)\ttotal: 19.3s\tremaining: 3m 41s\n",
            "600:\tlearn: 5661122.4843626\ttest: 9251631.3871388\tbest: 9251294.2054815 (598)\ttotal: 29.8s\tremaining: 3m 38s\n",
            "800:\tlearn: 5103437.4664509\ttest: 9180463.5047920\tbest: 9180463.5047920 (800)\ttotal: 38.5s\tremaining: 3m 21s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9156782.38\n",
            "bestIteration = 903\n",
            "\n",
            "Shrink model to first 904 iterations.\n",
            "split KFold5\n",
            "RMSE 9156782.379771348\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1708407.6909034913, 'MAE': 3416815.631805987, 'RMSE': 4814885.202816549}, {'Huber:delta=0.5': 2802109.7066727434, 'MAE': 5604219.663344495, 'RMSE': 9156782.379771344}])\n",
            "CV_RMSE_Score:9047478.210059281\n",
            "0:\tlearn: 40374571.5228432\ttest: 40684704.3564019\tbest: 40684704.3564019 (0)\ttotal: 54.5ms\tremaining: 4m 32s\n",
            "200:\tlearn: 7870075.2602916\ttest: 9951077.6818092\tbest: 9951077.6818092 (200)\ttotal: 10.7s\tremaining: 4m 14s\n",
            "400:\tlearn: 6506120.8433796\ttest: 9550364.3827762\tbest: 9550364.3827762 (400)\ttotal: 19.4s\tremaining: 3m 41s\n",
            "600:\tlearn: 5682699.9684635\ttest: 9400112.3690913\tbest: 9399007.7719743 (595)\ttotal: 30s\tremaining: 3m 39s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9388992.383\n",
            "bestIteration = 629\n",
            "\n",
            "Shrink model to first 630 iterations.\n",
            "split KFold1\n",
            "RMSE 9388992.383303683\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1934534.699206503, 'MAE': 3869069.648412, 'RMSE': 5526778.169947332}, {'Huber:delta=0.5': 2825411.2876326665, 'MAE': 5650822.825264289, 'RMSE': 9388992.383303687}])\n",
            "0:\tlearn: 40431393.0328191\ttest: 40476102.4539179\tbest: 40476102.4539179 (0)\ttotal: 53.7ms\tremaining: 4m 28s\n",
            "200:\tlearn: 7778562.8875023\ttest: 9852490.0096928\tbest: 9852490.0096928 (200)\ttotal: 10.7s\tremaining: 4m 14s\n",
            "400:\tlearn: 6436898.6216825\ttest: 9453326.5528720\tbest: 9452823.5762875 (398)\ttotal: 21.1s\tremaining: 4m 2s\n",
            "600:\tlearn: 5686316.1801849\ttest: 9297465.3778840\tbest: 9297465.3778840 (600)\ttotal: 29.7s\tremaining: 3m 37s\n",
            "800:\tlearn: 5118114.8242339\ttest: 9222336.9545353\tbest: 9220700.2374544 (797)\ttotal: 40.7s\tremaining: 3m 33s\n",
            "1000:\tlearn: 4662117.0494845\ttest: 9189103.7685997\tbest: 9188400.1304619 (996)\ttotal: 51.4s\tremaining: 3m 25s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9183376.546\n",
            "bestIteration = 1029\n",
            "\n",
            "Shrink model to first 1030 iterations.\n",
            "split KFold2\n",
            "RMSE 9183376.545760632\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1627471.1931729265, 'MAE': 3254942.636344837, 'RMSE': 4565939.187823711}, {'Huber:delta=0.5': 2746711.4172580517, 'MAE': 5493423.084515095, 'RMSE': 9183376.545760611}])\n",
            "0:\tlearn: 40481017.3751953\ttest: 40243069.0305636\tbest: 40243069.0305636 (0)\ttotal: 63.2ms\tremaining: 5m 15s\n",
            "200:\tlearn: 7889059.3909557\ttest: 9572638.9806788\tbest: 9572638.9806788 (200)\ttotal: 10.8s\tremaining: 4m 18s\n",
            "400:\tlearn: 6518132.7116996\ttest: 9195328.1910619\tbest: 9195328.1910619 (400)\ttotal: 19s\tremaining: 3m 37s\n",
            "600:\tlearn: 5755054.0501691\ttest: 9094770.9863206\tbest: 9094289.8485833 (599)\ttotal: 29.3s\tremaining: 3m 34s\n",
            "800:\tlearn: 5186323.9531334\ttest: 9051114.8139486\tbest: 9050577.6221200 (799)\ttotal: 39.7s\tremaining: 3m 28s\n",
            "1000:\tlearn: 4711783.6996713\ttest: 9011490.9974426\tbest: 9011032.7916230 (997)\ttotal: 48.6s\tremaining: 3m 14s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9002322.64\n",
            "bestIteration = 1086\n",
            "\n",
            "Shrink model to first 1087 iterations.\n",
            "split KFold3\n",
            "RMSE 9002322.63961902\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1604386.021997066, 'MAE': 3208772.293993109, 'RMSE': 4499686.576645502}, {'Huber:delta=0.5': 2736467.141610294, 'MAE': 5472934.5332196, 'RMSE': 9002322.63961902}])\n",
            "0:\tlearn: 40285109.5700607\ttest: 41044384.4644482\tbest: 41044384.4644482 (0)\ttotal: 55.8ms\tremaining: 4m 39s\n",
            "200:\tlearn: 7809427.6633877\ttest: 9584328.0663771\tbest: 9584328.0663771 (200)\ttotal: 10.1s\tremaining: 4m 2s\n",
            "400:\tlearn: 6453130.8752292\ttest: 9203186.7519598\tbest: 9203186.7519598 (400)\ttotal: 18.9s\tremaining: 3m 36s\n",
            "600:\tlearn: 5671468.3440313\ttest: 9059424.2181634\tbest: 9059083.9595903 (594)\ttotal: 29.6s\tremaining: 3m 36s\n",
            "800:\tlearn: 5111355.1538284\ttest: 8982676.6352893\tbest: 8982676.6352893 (800)\ttotal: 40.2s\tremaining: 3m 30s\n",
            "1000:\tlearn: 4673304.5346885\ttest: 8947343.0949098\tbest: 8947146.0674495 (997)\ttotal: 48.7s\tremaining: 3m 14s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8947146.067\n",
            "bestIteration = 997\n",
            "\n",
            "Shrink model to first 998 iterations.\n",
            "split KFold4\n",
            "RMSE 8947146.067449542\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1641729.3974674875, 'MAE': 3283459.0449339626, 'RMSE': 4638523.4047751995}, {'Huber:delta=0.5': 2781025.533412525, 'MAE': 5562051.316824047, 'RMSE': 8947146.067449545}])\n",
            "0:\tlearn: 40594556.8881456\ttest: 39747845.2634996\tbest: 39747845.2634996 (0)\ttotal: 114ms\tremaining: 9m 31s\n",
            "200:\tlearn: 7915060.0608358\ttest: 9434578.3322733\tbest: 9434578.3322733 (200)\ttotal: 9.91s\tremaining: 3m 56s\n",
            "400:\tlearn: 6515941.4305977\ttest: 9033640.9558429\tbest: 9033640.9558429 (400)\ttotal: 20.3s\tremaining: 3m 53s\n",
            "600:\tlearn: 5680155.0104434\ttest: 8907760.6425615\tbest: 8907760.6425615 (600)\ttotal: 30.6s\tremaining: 3m 44s\n",
            "800:\tlearn: 5087320.5432135\ttest: 8846898.4451486\tbest: 8846125.1978853 (795)\ttotal: 39.4s\tremaining: 3m 26s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8818292.279\n",
            "bestIteration = 949\n",
            "\n",
            "Shrink model to first 950 iterations.\n",
            "split KFold5\n",
            "RMSE 8818292.278757341\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1662759.451722597, 'MAE': 3325519.153444208, 'RMSE': 4710173.094579195}, {'Huber:delta=0.5': 2732496.103113063, 'MAE': 5464992.45622513, 'RMSE': 8818292.278757334}])\n",
            "CV_RMSE_Score:9068025.982978042\n",
            "0:\tlearn: 40536259.8367656\ttest: 40008513.6549640\tbest: 40008513.6549640 (0)\ttotal: 66.2ms\tremaining: 5m 30s\n",
            "200:\tlearn: 7927194.5783676\ttest: 9452715.6857740\tbest: 9452715.6857740 (200)\ttotal: 10.8s\tremaining: 4m 17s\n",
            "400:\tlearn: 6523599.0149516\ttest: 9092587.3926262\tbest: 9092587.3926262 (400)\ttotal: 19.3s\tremaining: 3m 41s\n",
            "600:\tlearn: 5733451.6052665\ttest: 8980008.6448310\tbest: 8980008.6448310 (600)\ttotal: 29.8s\tremaining: 3m 38s\n",
            "800:\tlearn: 5161269.6611996\ttest: 8932465.2624994\tbest: 8932240.4223395 (799)\ttotal: 40.4s\tremaining: 3m 31s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8924052.322\n",
            "bestIteration = 864\n",
            "\n",
            "Shrink model to first 865 iterations.\n",
            "split KFold1\n",
            "RMSE 8924052.321504807\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1749081.1011118495, 'MAE': 3498162.4522226844, 'RMSE': 4952852.971638502}, {'Huber:delta=0.5': 2702180.1772532514, 'MAE': 5404360.604505506, 'RMSE': 8924052.321504805}])\n",
            "0:\tlearn: 40491903.8438834\ttest: 40192556.4430164\tbest: 40192556.4430164 (0)\ttotal: 54ms\tremaining: 4m 29s\n",
            "200:\tlearn: 7791527.5704455\ttest: 9624600.1005182\tbest: 9624600.1005182 (200)\ttotal: 10.4s\tremaining: 4m 8s\n",
            "400:\tlearn: 6442385.4919666\ttest: 9257542.7003323\tbest: 9257542.7003323 (400)\ttotal: 18.8s\tremaining: 3m 36s\n",
            "600:\tlearn: 5645546.7096457\ttest: 9121820.6754977\tbest: 9120750.0572590 (598)\ttotal: 28.9s\tremaining: 3m 31s\n",
            "800:\tlearn: 5065119.7122517\ttest: 9063421.0941348\tbest: 9063421.0941348 (800)\ttotal: 39.7s\tremaining: 3m 28s\n",
            "1000:\tlearn: 4624621.1915414\ttest: 9021419.1689379\tbest: 9021419.1689379 (1000)\ttotal: 49.8s\tremaining: 3m 18s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9019426.511\n",
            "bestIteration = 1021\n",
            "\n",
            "Shrink model to first 1022 iterations.\n",
            "split KFold2\n",
            "RMSE 9019426.510613741\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1609084.1111567037, 'MAE': 3218168.472312417, 'RMSE': 4538902.080898853}, {'Huber:delta=0.5': 2782207.937776689, 'MAE': 5564416.125552399, 'RMSE': 9019426.51061374}])\n",
            "0:\tlearn: 40443993.7680348\ttest: 40425319.0528236\tbest: 40425319.0528236 (0)\ttotal: 74ms\tremaining: 6m 9s\n",
            "200:\tlearn: 7909404.4905367\ttest: 9758413.3120842\tbest: 9758413.3120842 (200)\ttotal: 8.77s\tremaining: 3m 29s\n",
            "400:\tlearn: 6487087.3408415\ttest: 9334209.1172766\tbest: 9333050.7918048 (397)\ttotal: 19.1s\tremaining: 3m 39s\n",
            "600:\tlearn: 5684773.5422339\ttest: 9190635.4970990\tbest: 9190503.9649258 (599)\ttotal: 29.7s\tremaining: 3m 37s\n",
            "800:\tlearn: 5112779.0043521\ttest: 9112340.4676386\tbest: 9112340.4676386 (800)\ttotal: 39.8s\tremaining: 3m 28s\n",
            "1000:\tlearn: 4656616.5017043\ttest: 9060747.8189012\tbest: 9060747.8189012 (1000)\ttotal: 49s\tremaining: 3m 15s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9034837.022\n",
            "bestIteration = 1163\n",
            "\n",
            "Shrink model to first 1164 iterations.\n",
            "split KFold3\n",
            "RMSE 9034837.021719694\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1546740.3888469222, 'MAE': 3093481.027692846, 'RMSE': 4321750.986858463}, {'Huber:delta=0.5': 2739654.0762610044, 'MAE': 5479308.402520984, 'RMSE': 9034837.0217197}])\n",
            "0:\tlearn: 40375960.0743673\ttest: 40704479.5495943\tbest: 40704479.5495943 (0)\ttotal: 63.1ms\tremaining: 5m 15s\n",
            "200:\tlearn: 7851684.0655969\ttest: 9631497.1787017\tbest: 9631497.1787017 (200)\ttotal: 10.5s\tremaining: 4m 11s\n",
            "400:\tlearn: 6518692.6949578\ttest: 9257267.5320652\tbest: 9257267.5320652 (400)\ttotal: 19s\tremaining: 3m 37s\n",
            "600:\tlearn: 5703800.1520951\ttest: 9104342.8511545\tbest: 9104342.8511545 (600)\ttotal: 29.5s\tremaining: 3m 35s\n",
            "800:\tlearn: 5109892.3757727\ttest: 9033374.4042090\tbest: 9032995.8247834 (799)\ttotal: 40s\tremaining: 3m 29s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9007833.933\n",
            "bestIteration = 898\n",
            "\n",
            "Shrink model to first 899 iterations.\n",
            "split KFold4\n",
            "RMSE 9007833.933406075\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1711161.3683253108, 'MAE': 3422322.9866496255, 'RMSE': 4827735.606661734}, {'Huber:delta=0.5': 2752147.6802837104, 'MAE': 5504295.610566401, 'RMSE': 9007833.93340607}])\n",
            "0:\tlearn: 40320877.1476780\ttest: 40897425.7163923\tbest: 40897425.7163923 (0)\ttotal: 58.3ms\tremaining: 4m 51s\n",
            "200:\tlearn: 7866713.7504729\ttest: 9822803.1447542\tbest: 9822803.1447542 (200)\ttotal: 10.6s\tremaining: 4m 12s\n",
            "400:\tlearn: 6493395.2451357\ttest: 9358486.8737335\tbest: 9358486.8737335 (400)\ttotal: 20.9s\tremaining: 3m 59s\n",
            "600:\tlearn: 5725245.0696428\ttest: 9212398.0211185\tbest: 9212398.0211185 (600)\ttotal: 29.1s\tremaining: 3m 32s\n",
            "800:\tlearn: 5129778.5525951\ttest: 9130634.4667051\tbest: 9127983.7928234 (795)\ttotal: 39.6s\tremaining: 3m 27s\n",
            "1000:\tlearn: 4673130.1591730\ttest: 9074062.4714217\tbest: 9074062.4714217 (1000)\ttotal: 50.1s\tremaining: 3m 20s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9056891.087\n",
            "bestIteration = 1140\n",
            "\n",
            "Shrink model to first 1141 iterations.\n",
            "split KFold5\n",
            "RMSE 9056891.086603716\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1561458.7877352738, 'MAE': 3122917.82546954, 'RMSE': 4369821.177713359}, {'Huber:delta=0.5': 2782297.7710385, 'MAE': 5564595.792075981, 'RMSE': 9056891.086603703}])\n",
            "CV_RMSE_Score:9008608.174769606\n",
            "0:\tlearn: 40351529.3055456\ttest: 40794888.2892370\tbest: 40794888.2892370 (0)\ttotal: 55.5ms\tremaining: 4m 37s\n",
            "20:\tlearn: 15539161.6780626\ttest: 15462940.5854241\tbest: 15462940.5854241 (20)\ttotal: 913ms\tremaining: 3m 36s\n",
            "40:\tlearn: 11485397.2647095\ttest: 11452287.9098242\tbest: 11452287.9098242 (40)\ttotal: 1.98s\tremaining: 3m 59s\n",
            "60:\tlearn: 10391944.2608352\ttest: 10570234.6991139\tbest: 10570234.6991139 (60)\ttotal: 3.73s\tremaining: 5m 2s\n",
            "80:\tlearn: 9813968.8632720\ttest: 10233959.8161350\tbest: 10233959.8161350 (80)\ttotal: 5.41s\tremaining: 5m 28s\n",
            "100:\tlearn: 9341271.2068387\ttest: 9965848.5834162\tbest: 9965848.5834162 (100)\ttotal: 6.42s\tremaining: 5m 11s\n",
            "120:\tlearn: 8972870.0050934\ttest: 9759780.1924200\tbest: 9759780.1924200 (120)\ttotal: 7.27s\tremaining: 4m 53s\n",
            "140:\tlearn: 8684877.6423408\ttest: 9636577.7172088\tbest: 9636577.7172088 (140)\ttotal: 8.12s\tremaining: 4m 39s\n",
            "160:\tlearn: 8394296.9722569\ttest: 9504214.7253726\tbest: 9504214.7253726 (160)\ttotal: 9.05s\tremaining: 4m 31s\n",
            "180:\tlearn: 8126524.1223510\ttest: 9393864.5166941\tbest: 9393864.5166941 (180)\ttotal: 9.91s\tremaining: 4m 23s\n",
            "200:\tlearn: 7882849.4695564\ttest: 9311178.3949038\tbest: 9311178.3949038 (200)\ttotal: 10.8s\tremaining: 4m 16s\n",
            "220:\tlearn: 7686870.5781888\ttest: 9236777.3299919\tbest: 9236777.3299919 (220)\ttotal: 11.6s\tremaining: 4m 10s\n",
            "240:\tlearn: 7501818.3116562\ttest: 9175170.3665798\tbest: 9175170.3665798 (240)\ttotal: 12.4s\tremaining: 4m 5s\n",
            "260:\tlearn: 7343888.1524866\ttest: 9120684.6311141\tbest: 9120684.6311141 (260)\ttotal: 13.2s\tremaining: 4m\n",
            "280:\tlearn: 7203427.7044043\ttest: 9077864.2684842\tbest: 9077864.2684842 (280)\ttotal: 14.1s\tremaining: 3m 56s\n",
            "300:\tlearn: 7064499.8816079\ttest: 9045167.1773947\tbest: 9045167.1773947 (300)\ttotal: 14.9s\tremaining: 3m 52s\n",
            "320:\tlearn: 6935335.3747001\ttest: 9022946.4842430\tbest: 9022946.4842430 (320)\ttotal: 15.8s\tremaining: 3m 50s\n",
            "340:\tlearn: 6821861.6050050\ttest: 8998866.7303686\tbest: 8998866.7303686 (340)\ttotal: 17.5s\tremaining: 3m 59s\n",
            "360:\tlearn: 6712931.9632367\ttest: 8973058.6993547\tbest: 8972455.2122431 (359)\ttotal: 19.3s\tremaining: 4m 7s\n",
            "380:\tlearn: 6607562.4919615\ttest: 8956922.3404464\tbest: 8956922.3404464 (380)\ttotal: 20.4s\tremaining: 4m 7s\n",
            "400:\tlearn: 6509172.7619914\ttest: 8939437.7582300\tbest: 8939437.7582300 (400)\ttotal: 21.2s\tremaining: 4m 3s\n",
            "420:\tlearn: 6408697.4169238\ttest: 8919335.3873476\tbest: 8919335.3873476 (420)\ttotal: 22.1s\tremaining: 3m 59s\n",
            "440:\tlearn: 6327122.9260861\ttest: 8903782.0597143\tbest: 8903782.0597143 (440)\ttotal: 23s\tremaining: 3m 57s\n",
            "460:\tlearn: 6240966.2901999\ttest: 8886275.4601800\tbest: 8886275.4601800 (460)\ttotal: 23.9s\tremaining: 3m 54s\n",
            "480:\tlearn: 6162449.5329878\ttest: 8869330.3924872\tbest: 8869253.7849253 (479)\ttotal: 24.8s\tremaining: 3m 52s\n",
            "500:\tlearn: 6085418.3771718\ttest: 8852319.1384429\tbest: 8852319.1384429 (500)\ttotal: 25.7s\tremaining: 3m 50s\n",
            "520:\tlearn: 6007731.9859733\ttest: 8846972.6303496\tbest: 8846972.6303496 (520)\ttotal: 26.6s\tremaining: 3m 48s\n",
            "540:\tlearn: 5941469.1456881\ttest: 8837822.9963280\tbest: 8837822.9963280 (540)\ttotal: 27.4s\tremaining: 3m 45s\n",
            "560:\tlearn: 5878105.7839678\ttest: 8828626.5190069\tbest: 8828626.5190069 (560)\ttotal: 28.2s\tremaining: 3m 43s\n",
            "580:\tlearn: 5805557.4853811\ttest: 8815313.9255139\tbest: 8815313.9255139 (580)\ttotal: 29.1s\tremaining: 3m 41s\n",
            "600:\tlearn: 5744474.2016416\ttest: 8814730.4374098\tbest: 8814488.6137330 (598)\ttotal: 29.9s\tremaining: 3m 39s\n",
            "620:\tlearn: 5674855.9143822\ttest: 8807506.6369833\tbest: 8807506.6369833 (620)\ttotal: 31.6s\tremaining: 3m 43s\n",
            "640:\tlearn: 5609714.2785073\ttest: 8796756.8688740\tbest: 8796645.5308507 (639)\ttotal: 33.3s\tremaining: 3m 46s\n",
            "660:\tlearn: 5548111.4264733\ttest: 8790933.4730052\tbest: 8789013.7679442 (656)\ttotal: 34.5s\tremaining: 3m 46s\n",
            "680:\tlearn: 5483904.9560055\ttest: 8781909.0147227\tbest: 8781909.0147227 (680)\ttotal: 35.3s\tremaining: 3m 43s\n",
            "700:\tlearn: 5422894.1298184\ttest: 8780620.9020084\tbest: 8780620.9020084 (700)\ttotal: 36.1s\tremaining: 3m 41s\n",
            "720:\tlearn: 5359607.2624200\ttest: 8779840.3146984\tbest: 8779750.4423302 (719)\ttotal: 37s\tremaining: 3m 39s\n",
            "740:\tlearn: 5293997.1475580\ttest: 8770726.1883944\tbest: 8770127.9332901 (736)\ttotal: 37.8s\tremaining: 3m 37s\n",
            "760:\tlearn: 5241156.0503433\ttest: 8765555.4103524\tbest: 8765555.4103524 (760)\ttotal: 38.7s\tremaining: 3m 35s\n",
            "780:\tlearn: 5190372.8624950\ttest: 8760480.9914880\tbest: 8760480.9914880 (780)\ttotal: 39.6s\tremaining: 3m 33s\n",
            "800:\tlearn: 5137257.3554759\ttest: 8755627.2490623\tbest: 8755627.2490623 (800)\ttotal: 40.4s\tremaining: 3m 32s\n",
            "820:\tlearn: 5086972.2319999\ttest: 8754529.1432202\tbest: 8754529.1432202 (820)\ttotal: 41.3s\tremaining: 3m 30s\n",
            "840:\tlearn: 5040488.3274139\ttest: 8750046.9757528\tbest: 8750046.9757528 (840)\ttotal: 42.1s\tremaining: 3m 28s\n",
            "860:\tlearn: 4995522.1404798\ttest: 8744575.8683929\tbest: 8744575.8683929 (860)\ttotal: 43s\tremaining: 3m 26s\n",
            "880:\tlearn: 4951116.9302273\ttest: 8740790.2076425\tbest: 8740140.8865178 (877)\ttotal: 43.8s\tremaining: 3m 24s\n",
            "900:\tlearn: 4911596.4743786\ttest: 8739667.2970068\tbest: 8739496.0253644 (898)\ttotal: 45.3s\tremaining: 3m 26s\n",
            "920:\tlearn: 4863102.3457862\ttest: 8737262.6929605\tbest: 8737262.6929605 (920)\ttotal: 47s\tremaining: 3m 28s\n",
            "940:\tlearn: 4823213.7020358\ttest: 8734331.6491147\tbest: 8734258.8939676 (938)\ttotal: 48.4s\tremaining: 3m 28s\n",
            "960:\tlearn: 4782017.9984509\ttest: 8730358.6424114\tbest: 8730358.6424114 (960)\ttotal: 49.2s\tremaining: 3m 26s\n",
            "980:\tlearn: 4740363.4791649\ttest: 8728471.9495187\tbest: 8728471.9495187 (980)\ttotal: 50.1s\tremaining: 3m 25s\n",
            "1000:\tlearn: 4693483.0672176\ttest: 8727216.5504241\tbest: 8727216.5504241 (1000)\ttotal: 50.9s\tremaining: 3m 23s\n",
            "1020:\tlearn: 4653639.6370962\ttest: 8726329.2800330\tbest: 8726089.8667920 (1007)\ttotal: 51.8s\tremaining: 3m 21s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8726089.867\n",
            "bestIteration = 1007\n",
            "\n",
            "Shrink model to first 1008 iterations.\n",
            "split KFold1\n",
            "RMSE 8726089.866792042\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1652298.7262196466, 'MAE': 3304597.7024383005, 'RMSE': 4637273.61580642}, {'Huber:delta=0.5': 2690590.306151762, 'MAE': 5381180.862302523, 'RMSE': 8726089.86679204}])\n",
            "0:\tlearn: 40465199.5567477\ttest: 40314363.7083832\tbest: 40314363.7083832 (0)\ttotal: 54.3ms\tremaining: 4m 31s\n",
            "20:\tlearn: 15361941.7822050\ttest: 15846716.8592389\tbest: 15846716.8592389 (20)\ttotal: 924ms\tremaining: 3m 39s\n",
            "40:\tlearn: 11257861.3097000\ttest: 12188613.4925194\tbest: 12188613.4925194 (40)\ttotal: 1.78s\tremaining: 3m 34s\n",
            "60:\tlearn: 10203300.1092540\ttest: 11360298.7867599\tbest: 11360298.7867599 (60)\ttotal: 2.61s\tremaining: 3m 31s\n",
            "80:\tlearn: 9656600.2539478\ttest: 10966525.5176167\tbest: 10966525.5176167 (80)\ttotal: 3.43s\tremaining: 3m 28s\n",
            "100:\tlearn: 9209575.4489196\ttest: 10686104.2679859\tbest: 10686104.2679859 (100)\ttotal: 4.33s\tremaining: 3m 30s\n",
            "120:\tlearn: 8837070.0766402\ttest: 10494795.6272747\tbest: 10494795.6272747 (120)\ttotal: 5.16s\tremaining: 3m 28s\n",
            "140:\tlearn: 8513777.1679986\ttest: 10299145.5912934\tbest: 10299145.5912934 (140)\ttotal: 6.39s\tremaining: 3m 40s\n",
            "160:\tlearn: 8247704.7449992\ttest: 10181300.8807604\tbest: 10181300.8807604 (160)\ttotal: 8.09s\tremaining: 4m 3s\n",
            "180:\tlearn: 8008253.2635162\ttest: 10078513.3219399\tbest: 10078513.3219399 (180)\ttotal: 9.81s\tremaining: 4m 21s\n",
            "200:\tlearn: 7791471.4068169\ttest: 9980940.4873005\tbest: 9980940.4873005 (200)\ttotal: 10.7s\tremaining: 4m 14s\n",
            "220:\tlearn: 7598112.8434730\ttest: 9910773.9739131\tbest: 9910773.9739131 (220)\ttotal: 11.5s\tremaining: 4m 8s\n",
            "240:\tlearn: 7422148.4873217\ttest: 9846267.8855221\tbest: 9846267.8855221 (240)\ttotal: 12.3s\tremaining: 4m 3s\n",
            "260:\tlearn: 7276741.2346676\ttest: 9792099.6980370\tbest: 9792099.6980370 (260)\ttotal: 13.2s\tremaining: 3m 59s\n",
            "280:\tlearn: 7122634.3270468\ttest: 9748166.9278808\tbest: 9748166.9278808 (280)\ttotal: 14.1s\tremaining: 3m 57s\n",
            "300:\tlearn: 6988804.0608963\ttest: 9706397.7665340\tbest: 9706397.7665340 (300)\ttotal: 15.1s\tremaining: 3m 55s\n",
            "320:\tlearn: 6884231.6523437\ttest: 9672874.9669723\tbest: 9672874.9669723 (320)\ttotal: 15.9s\tremaining: 3m 52s\n",
            "340:\tlearn: 6782306.3988729\ttest: 9640609.8209814\tbest: 9640550.4529989 (339)\ttotal: 16.8s\tremaining: 3m 48s\n",
            "360:\tlearn: 6667062.3051798\ttest: 9621192.4016802\tbest: 9621192.4016802 (360)\ttotal: 17.6s\tremaining: 3m 46s\n",
            "380:\tlearn: 6556113.1931124\ttest: 9604315.6228883\tbest: 9602927.8112067 (379)\ttotal: 18.5s\tremaining: 3m 43s\n",
            "400:\tlearn: 6462516.5948938\ttest: 9578216.5661730\tbest: 9578216.5661730 (400)\ttotal: 19.3s\tremaining: 3m 41s\n",
            "420:\tlearn: 6372167.8401634\ttest: 9559187.4697277\tbest: 9559162.2036633 (418)\ttotal: 20.6s\tremaining: 3m 44s\n",
            "440:\tlearn: 6280773.1301569\ttest: 9539159.3068637\tbest: 9539159.3068637 (440)\ttotal: 22.3s\tremaining: 3m 50s\n",
            "460:\tlearn: 6197024.6868987\ttest: 9524967.2158425\tbest: 9524967.2158425 (460)\ttotal: 24s\tremaining: 3m 56s\n",
            "480:\tlearn: 6119185.2226876\ttest: 9508007.4623739\tbest: 9508007.4623739 (480)\ttotal: 24.9s\tremaining: 3m 53s\n",
            "500:\tlearn: 6048478.8073596\ttest: 9497549.9381395\tbest: 9497549.9381395 (500)\ttotal: 25.8s\tremaining: 3m 51s\n",
            "520:\tlearn: 5971977.2819078\ttest: 9481526.2509329\tbest: 9480837.1303378 (517)\ttotal: 26.7s\tremaining: 3m 49s\n",
            "540:\tlearn: 5901747.3813403\ttest: 9470011.1443495\tbest: 9467332.2388657 (539)\ttotal: 27.5s\tremaining: 3m 47s\n",
            "560:\tlearn: 5831950.7865477\ttest: 9462206.3496753\tbest: 9462206.3496753 (560)\ttotal: 28.4s\tremaining: 3m 44s\n",
            "580:\tlearn: 5762783.4129692\ttest: 9453081.2860962\tbest: 9453081.2860962 (580)\ttotal: 29.3s\tremaining: 3m 42s\n",
            "600:\tlearn: 5692206.9673241\ttest: 9443428.5373513\tbest: 9443428.5373513 (600)\ttotal: 30.2s\tremaining: 3m 40s\n",
            "620:\tlearn: 5627522.0986688\ttest: 9434590.7882045\tbest: 9434590.7882045 (620)\ttotal: 31s\tremaining: 3m 38s\n",
            "640:\tlearn: 5570494.7890750\ttest: 9423922.8879894\tbest: 9422673.0670471 (638)\ttotal: 31.9s\tremaining: 3m 36s\n",
            "660:\tlearn: 5508167.0808640\ttest: 9415781.4879514\tbest: 9415639.4080094 (659)\ttotal: 32.7s\tremaining: 3m 34s\n",
            "680:\tlearn: 5455974.2811607\ttest: 9410519.9756027\tbest: 9408481.6454108 (675)\ttotal: 33.6s\tremaining: 3m 33s\n",
            "700:\tlearn: 5397371.7947954\ttest: 9400527.7589151\tbest: 9400527.7589151 (700)\ttotal: 34.9s\tremaining: 3m 33s\n",
            "720:\tlearn: 5343515.2655063\ttest: 9393761.5736502\tbest: 9393761.5736502 (720)\ttotal: 36.5s\tremaining: 3m 36s\n",
            "740:\tlearn: 5291028.8910999\ttest: 9391513.7303046\tbest: 9390670.1417232 (729)\ttotal: 38.3s\tremaining: 3m 39s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9390670.142\n",
            "bestIteration = 729\n",
            "\n",
            "Shrink model to first 730 iterations.\n",
            "split KFold2\n",
            "RMSE 9390670.141723238\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1845765.4059715816, 'MAE': 3691531.061942159, 'RMSE': 5267791.721771447}, {'Huber:delta=0.5': 2843951.3297490617, 'MAE': 5687902.90949712, 'RMSE': 9390670.141723244}])\n",
            "0:\tlearn: 40430268.2488647\ttest: 40478690.2501771\tbest: 40478690.2501771 (0)\ttotal: 59.2ms\tremaining: 4m 56s\n",
            "20:\tlearn: 15558155.9892102\ttest: 15779858.9906410\tbest: 15779858.9906410 (20)\ttotal: 988ms\tremaining: 3m 54s\n",
            "40:\tlearn: 11484030.7666260\ttest: 11899480.9171496\tbest: 11899480.9171496 (40)\ttotal: 1.85s\tremaining: 3m 43s\n",
            "60:\tlearn: 10410288.5537929\ttest: 11042851.8289151\tbest: 11042851.8289151 (60)\ttotal: 2.69s\tremaining: 3m 38s\n",
            "80:\tlearn: 9805298.7150293\ttest: 10642454.9822780\tbest: 10642454.9822780 (80)\ttotal: 3.53s\tremaining: 3m 34s\n",
            "100:\tlearn: 9332399.6668344\ttest: 10328444.7405280\tbest: 10328444.7405280 (100)\ttotal: 4.39s\tremaining: 3m 32s\n",
            "120:\tlearn: 8982095.3450309\ttest: 10141007.7628489\tbest: 10141007.7628489 (120)\ttotal: 5.23s\tremaining: 3m 30s\n",
            "140:\tlearn: 8660328.3079217\ttest: 9983771.8800594\tbest: 9983771.8800594 (140)\ttotal: 6.07s\tremaining: 3m 29s\n",
            "160:\tlearn: 8372233.6061010\ttest: 9830323.8598666\tbest: 9830323.8598666 (160)\ttotal: 6.92s\tremaining: 3m 27s\n",
            "180:\tlearn: 8116238.8832282\ttest: 9695937.6707222\tbest: 9695937.6707222 (180)\ttotal: 7.78s\tremaining: 3m 27s\n",
            "200:\tlearn: 7898732.8624059\ttest: 9599488.1563725\tbest: 9599488.1563725 (200)\ttotal: 8.65s\tremaining: 3m 26s\n",
            "220:\tlearn: 7712762.5243676\ttest: 9520308.8077462\tbest: 9520308.8077462 (220)\ttotal: 9.87s\tremaining: 3m 33s\n",
            "240:\tlearn: 7536556.4753005\ttest: 9466787.5420077\tbest: 9466787.5420077 (240)\ttotal: 11.6s\tremaining: 3m 48s\n",
            "260:\tlearn: 7379836.4183766\ttest: 9420182.9299160\tbest: 9420182.9299160 (260)\ttotal: 13.4s\tremaining: 4m 3s\n",
            "280:\tlearn: 7240853.8988102\ttest: 9375714.9659581\tbest: 9375714.9659581 (280)\ttotal: 14.3s\tremaining: 3m 59s\n",
            "300:\tlearn: 7097857.6624799\ttest: 9327068.2074197\tbest: 9327068.2074197 (300)\ttotal: 15.1s\tremaining: 3m 56s\n",
            "320:\tlearn: 6961427.0431149\ttest: 9295656.2204667\tbest: 9295656.2204667 (320)\ttotal: 16.1s\tremaining: 3m 54s\n",
            "340:\tlearn: 6840074.5516309\ttest: 9270710.8637972\tbest: 9269997.6164049 (339)\ttotal: 16.9s\tremaining: 3m 51s\n",
            "360:\tlearn: 6726612.0704079\ttest: 9243581.2120693\tbest: 9243581.2120693 (360)\ttotal: 17.8s\tremaining: 3m 48s\n",
            "380:\tlearn: 6616811.4596845\ttest: 9216807.1345452\tbest: 9216807.1345452 (380)\ttotal: 18.6s\tremaining: 3m 45s\n",
            "400:\tlearn: 6514870.2184630\ttest: 9191955.0865787\tbest: 9191955.0865787 (400)\ttotal: 19.5s\tremaining: 3m 43s\n",
            "420:\tlearn: 6413920.5876848\ttest: 9174852.2380868\tbest: 9174852.2380868 (420)\ttotal: 20.3s\tremaining: 3m 41s\n",
            "440:\tlearn: 6316448.7759358\ttest: 9159470.1941983\tbest: 9159470.1941983 (440)\ttotal: 21.2s\tremaining: 3m 39s\n",
            "460:\tlearn: 6225730.9403923\ttest: 9137798.9112240\tbest: 9137798.9112240 (460)\ttotal: 22s\tremaining: 3m 36s\n",
            "480:\tlearn: 6140885.8198283\ttest: 9125038.7103258\tbest: 9124392.4620978 (479)\ttotal: 22.9s\tremaining: 3m 35s\n",
            "500:\tlearn: 6057151.0857145\ttest: 9106951.8228947\tbest: 9106089.1272175 (498)\ttotal: 24s\tremaining: 3m 35s\n",
            "520:\tlearn: 5976700.3006211\ttest: 9088424.7022185\tbest: 9088424.7022185 (520)\ttotal: 25.7s\tremaining: 3m 41s\n",
            "540:\tlearn: 5894559.7934127\ttest: 9071371.1419615\tbest: 9071371.1419615 (540)\ttotal: 27.4s\tremaining: 3m 46s\n",
            "560:\tlearn: 5816844.6155560\ttest: 9062367.7979019\tbest: 9060537.8193108 (558)\ttotal: 28.5s\tremaining: 3m 45s\n",
            "580:\tlearn: 5740895.8770690\ttest: 9047494.4204520\tbest: 9047494.4204520 (580)\ttotal: 29.3s\tremaining: 3m 42s\n",
            "600:\tlearn: 5673681.5971978\ttest: 9036509.5414422\tbest: 9036015.6272537 (599)\ttotal: 30.1s\tremaining: 3m 40s\n",
            "620:\tlearn: 5607850.0336210\ttest: 9031119.5106152\tbest: 9031119.5106152 (620)\ttotal: 31s\tremaining: 3m 38s\n",
            "640:\tlearn: 5538197.9068890\ttest: 9025661.9839858\tbest: 9025661.9839858 (640)\ttotal: 31.9s\tremaining: 3m 37s\n",
            "660:\tlearn: 5475677.6088585\ttest: 9018711.1635516\tbest: 9018711.1635516 (660)\ttotal: 32.8s\tremaining: 3m 35s\n",
            "680:\tlearn: 5414463.0121506\ttest: 9008560.6585137\tbest: 9008560.6585137 (680)\ttotal: 33.7s\tremaining: 3m 33s\n",
            "700:\tlearn: 5355064.7537085\ttest: 9004726.7871279\tbest: 9003828.8410073 (699)\ttotal: 34.5s\tremaining: 3m 31s\n",
            "720:\tlearn: 5298081.3865161\ttest: 8998894.0054840\tbest: 8998736.2137733 (718)\ttotal: 35.3s\tremaining: 3m 29s\n",
            "740:\tlearn: 5240483.1796225\ttest: 8995679.6915507\tbest: 8994560.3233425 (736)\ttotal: 36.2s\tremaining: 3m 28s\n",
            "760:\tlearn: 5184518.9014702\ttest: 8990738.1216059\tbest: 8990738.1216059 (760)\ttotal: 37s\tremaining: 3m 26s\n",
            "780:\tlearn: 5128479.0969553\ttest: 8982119.7742885\tbest: 8982119.7742885 (780)\ttotal: 38.1s\tremaining: 3m 25s\n",
            "800:\tlearn: 5078905.3783709\ttest: 8979450.6589751\tbest: 8979403.5752364 (796)\ttotal: 39.7s\tremaining: 3m 28s\n",
            "820:\tlearn: 5027756.9624633\ttest: 8974607.5752362\tbest: 8974607.5752362 (820)\ttotal: 41.5s\tremaining: 3m 31s\n",
            "840:\tlearn: 4979079.4307115\ttest: 8972733.3765603\tbest: 8971784.5707744 (838)\ttotal: 42.6s\tremaining: 3m 30s\n",
            "860:\tlearn: 4932143.7477907\ttest: 8967506.4308181\tbest: 8967506.4308181 (860)\ttotal: 43.5s\tremaining: 3m 29s\n",
            "880:\tlearn: 4884092.7332395\ttest: 8965638.7946301\tbest: 8964498.7559063 (874)\ttotal: 44.4s\tremaining: 3m 27s\n",
            "900:\tlearn: 4845283.0028003\ttest: 8961003.6829236\tbest: 8960973.6779552 (899)\ttotal: 45.2s\tremaining: 3m 25s\n",
            "920:\tlearn: 4799180.6153919\ttest: 8958402.9571439\tbest: 8957215.3650113 (919)\ttotal: 46.1s\tremaining: 3m 24s\n",
            "940:\tlearn: 4752235.6998069\ttest: 8954919.3568168\tbest: 8954486.4095452 (938)\ttotal: 46.9s\tremaining: 3m 22s\n",
            "960:\tlearn: 4706840.3760033\ttest: 8952971.8001182\tbest: 8952971.8001182 (960)\ttotal: 47.8s\tremaining: 3m 20s\n",
            "980:\tlearn: 4661425.5312963\ttest: 8949206.5607886\tbest: 8949206.5607886 (980)\ttotal: 48.6s\tremaining: 3m 19s\n",
            "1000:\tlearn: 4619756.6081517\ttest: 8945142.0716175\tbest: 8945062.8504825 (999)\ttotal: 49.5s\tremaining: 3m 17s\n",
            "1020:\tlearn: 4581204.6403503\ttest: 8940925.5809231\tbest: 8940925.5809231 (1020)\ttotal: 50.3s\tremaining: 3m 16s\n",
            "1040:\tlearn: 4538404.2319294\ttest: 8938328.7801571\tbest: 8938328.7801571 (1040)\ttotal: 51.2s\tremaining: 3m 14s\n",
            "1060:\tlearn: 4493646.1713713\ttest: 8937397.8158477\tbest: 8936414.7853237 (1057)\ttotal: 52.1s\tremaining: 3m 13s\n",
            "1080:\tlearn: 4456291.6302964\ttest: 8935814.3608044\tbest: 8935551.4852707 (1074)\ttotal: 53.8s\tremaining: 3m 15s\n",
            "1100:\tlearn: 4419201.4347324\ttest: 8929754.7681446\tbest: 8929754.7681446 (1100)\ttotal: 55.6s\tremaining: 3m 16s\n",
            "1120:\tlearn: 4382232.3653978\ttest: 8930095.4469569\tbest: 8929009.0465812 (1101)\ttotal: 56.8s\tremaining: 3m 16s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8929009.047\n",
            "bestIteration = 1101\n",
            "\n",
            "Shrink model to first 1102 iterations.\n",
            "split KFold3\n",
            "RMSE 8929009.04658116\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1562874.5586322648, 'MAE': 3125749.3672635406, 'RMSE': 4380422.127164087}, {'Huber:delta=0.5': 2734028.247893336, 'MAE': 5468056.745785687, 'RMSE': 8929009.046581184}])\n",
            "0:\tlearn: 40546278.8916683\ttest: 39974039.5903832\tbest: 39974039.5903832 (0)\ttotal: 54.6ms\tremaining: 4m 32s\n",
            "20:\tlearn: 15379961.9916387\ttest: 15979760.9779728\tbest: 15979760.9779728 (20)\ttotal: 916ms\tremaining: 3m 37s\n",
            "40:\tlearn: 11231651.1287368\ttest: 12246945.8618433\tbest: 12246945.8618433 (40)\ttotal: 1.83s\tremaining: 3m 41s\n",
            "60:\tlearn: 10221214.7698631\ttest: 11384786.3535750\tbest: 11384786.3535750 (60)\ttotal: 2.69s\tremaining: 3m 37s\n",
            "80:\tlearn: 9653796.2012996\ttest: 10927844.0048340\tbest: 10927844.0048340 (80)\ttotal: 3.55s\tremaining: 3m 35s\n",
            "100:\tlearn: 9240818.8514642\ttest: 10639517.2292796\tbest: 10639517.2292796 (100)\ttotal: 4.37s\tremaining: 3m 31s\n",
            "120:\tlearn: 8904699.9838234\ttest: 10406416.9832452\tbest: 10406416.9832452 (120)\ttotal: 5.21s\tremaining: 3m 29s\n",
            "140:\tlearn: 8588953.8212695\ttest: 10194067.9356434\tbest: 10194067.9356434 (140)\ttotal: 6.02s\tremaining: 3m 27s\n",
            "160:\tlearn: 8314758.9290009\ttest: 10031676.4962889\tbest: 10031676.4962889 (160)\ttotal: 6.86s\tremaining: 3m 26s\n",
            "180:\tlearn: 8087970.5113847\ttest: 9914734.9901834\tbest: 9914734.9901834 (180)\ttotal: 7.68s\tremaining: 3m 24s\n",
            "200:\tlearn: 7858792.3458485\ttest: 9791113.3389553\tbest: 9791113.3389553 (200)\ttotal: 8.52s\tremaining: 3m 23s\n",
            "220:\tlearn: 7674350.4589723\ttest: 9717500.2056980\tbest: 9717500.2056980 (220)\ttotal: 9.65s\tremaining: 3m 28s\n",
            "240:\tlearn: 7489244.0647822\ttest: 9641709.3222153\tbest: 9641709.3222153 (240)\ttotal: 11.4s\tremaining: 3m 44s\n",
            "260:\tlearn: 7320664.3547749\ttest: 9571684.0659136\tbest: 9571238.9963742 (259)\ttotal: 13.1s\tremaining: 3m 58s\n",
            "280:\tlearn: 7179798.5677573\ttest: 9525416.2494533\tbest: 9525416.2494533 (280)\ttotal: 14.1s\tremaining: 3m 55s\n",
            "300:\tlearn: 7044325.1887717\ttest: 9474223.2473560\tbest: 9474223.2473560 (300)\ttotal: 14.9s\tremaining: 3m 52s\n",
            "320:\tlearn: 6923736.0893189\ttest: 9435242.5902374\tbest: 9435242.5902374 (320)\ttotal: 15.8s\tremaining: 3m 49s\n",
            "340:\tlearn: 6805008.8944856\ttest: 9398154.0685909\tbest: 9398154.0685909 (340)\ttotal: 16.6s\tremaining: 3m 46s\n",
            "360:\tlearn: 6693854.0479736\ttest: 9351978.8140917\tbest: 9351978.8140917 (360)\ttotal: 17.4s\tremaining: 3m 43s\n",
            "380:\tlearn: 6596715.0454089\ttest: 9324971.0206375\tbest: 9324504.6926066 (379)\ttotal: 18.3s\tremaining: 3m 41s\n",
            "400:\tlearn: 6511914.3520191\ttest: 9300872.5216971\tbest: 9300872.5216971 (400)\ttotal: 19.1s\tremaining: 3m 39s\n",
            "420:\tlearn: 6406188.4938962\ttest: 9274367.8480662\tbest: 9274367.8480662 (420)\ttotal: 20s\tremaining: 3m 37s\n",
            "440:\tlearn: 6321522.0902413\ttest: 9253460.3847497\tbest: 9253460.3847497 (440)\ttotal: 20.8s\tremaining: 3m 34s\n",
            "460:\tlearn: 6236959.2373666\ttest: 9235329.5024957\tbest: 9235329.5024957 (460)\ttotal: 21.6s\tremaining: 3m 32s\n",
            "480:\tlearn: 6149438.7991246\ttest: 9217080.0348996\tbest: 9217080.0348996 (480)\ttotal: 22.4s\tremaining: 3m 30s\n",
            "500:\tlearn: 6072391.7495637\ttest: 9200552.5915836\tbest: 9200552.5915836 (500)\ttotal: 23.3s\tremaining: 3m 29s\n",
            "520:\tlearn: 5993931.8133898\ttest: 9187679.2916275\tbest: 9187679.2916275 (520)\ttotal: 24.9s\tremaining: 3m 34s\n",
            "540:\tlearn: 5917208.9725652\ttest: 9177566.0177944\tbest: 9177226.3263329 (539)\ttotal: 26.7s\tremaining: 3m 39s\n",
            "560:\tlearn: 5839452.3208766\ttest: 9162427.4859693\tbest: 9162224.0516019 (559)\ttotal: 28s\tremaining: 3m 41s\n",
            "580:\tlearn: 5765792.8977802\ttest: 9153024.7452642\tbest: 9153024.7452642 (580)\ttotal: 28.8s\tremaining: 3m 39s\n",
            "600:\tlearn: 5696713.4219199\ttest: 9143917.9325042\tbest: 9143917.9325042 (600)\ttotal: 29.7s\tremaining: 3m 37s\n",
            "620:\tlearn: 5635303.1913419\ttest: 9131566.8165458\tbest: 9131441.0915201 (619)\ttotal: 30.5s\tremaining: 3m 35s\n",
            "640:\tlearn: 5576577.9521510\ttest: 9126223.0852405\tbest: 9126223.0852405 (640)\ttotal: 31.4s\tremaining: 3m 33s\n",
            "660:\tlearn: 5514855.2118224\ttest: 9116654.3773148\tbest: 9116654.3773148 (660)\ttotal: 32.3s\tremaining: 3m 31s\n",
            "680:\tlearn: 5454372.4261102\ttest: 9111413.2016270\tbest: 9111413.2016270 (680)\ttotal: 33.2s\tremaining: 3m 30s\n",
            "700:\tlearn: 5399624.8660193\ttest: 9103355.1413445\tbest: 9103355.1413445 (700)\ttotal: 34.1s\tremaining: 3m 29s\n",
            "720:\tlearn: 5342746.1674377\ttest: 9095766.4947702\tbest: 9095766.4947702 (720)\ttotal: 35s\tremaining: 3m 27s\n",
            "740:\tlearn: 5290122.1020429\ttest: 9090775.7182195\tbest: 9090775.7182195 (740)\ttotal: 35.8s\tremaining: 3m 26s\n",
            "760:\tlearn: 5243480.2096272\ttest: 9083035.8464592\tbest: 9083035.8464592 (760)\ttotal: 36.7s\tremaining: 3m 24s\n",
            "780:\tlearn: 5189626.4027506\ttest: 9074807.2320265\tbest: 9074644.2945888 (776)\ttotal: 37.5s\tremaining: 3m 22s\n",
            "800:\tlearn: 5139166.7529034\ttest: 9067920.3941315\tbest: 9067920.3941315 (800)\ttotal: 39.2s\tremaining: 3m 25s\n",
            "820:\tlearn: 5095703.5876795\ttest: 9063516.1447146\tbest: 9063048.5224561 (819)\ttotal: 40.9s\tremaining: 3m 27s\n",
            "840:\tlearn: 5047038.1440330\ttest: 9054096.3621064\tbest: 9054096.3621064 (840)\ttotal: 42.1s\tremaining: 3m 28s\n",
            "860:\tlearn: 4998916.9683502\ttest: 9047310.6521313\tbest: 9047282.0739873 (858)\ttotal: 43s\tremaining: 3m 26s\n",
            "880:\tlearn: 4952229.0229019\ttest: 9040948.5592192\tbest: 9040926.2524534 (879)\ttotal: 43.8s\tremaining: 3m 25s\n",
            "900:\tlearn: 4911345.7372803\ttest: 9039139.9621831\tbest: 9038233.2808653 (896)\ttotal: 44.7s\tremaining: 3m 23s\n",
            "920:\tlearn: 4868089.6271616\ttest: 9033587.2959523\tbest: 9033587.2959523 (920)\ttotal: 45.6s\tremaining: 3m 21s\n",
            "940:\tlearn: 4820174.5363033\ttest: 9030024.1479551\tbest: 9028950.1769362 (934)\ttotal: 46.5s\tremaining: 3m 20s\n",
            "960:\tlearn: 4780421.8485847\ttest: 9028275.8835358\tbest: 9027511.3231194 (957)\ttotal: 47.4s\tremaining: 3m 19s\n",
            "980:\tlearn: 4739338.8978565\ttest: 9021612.0779485\tbest: 9021612.0779485 (980)\ttotal: 48.3s\tremaining: 3m 18s\n",
            "1000:\tlearn: 4698481.9620775\ttest: 9017685.5144160\tbest: 9017603.7631131 (999)\ttotal: 49.2s\tremaining: 3m 16s\n",
            "1020:\tlearn: 4659323.5611399\ttest: 9011323.5573825\tbest: 9011323.5573825 (1020)\ttotal: 50s\tremaining: 3m 14s\n",
            "1040:\tlearn: 4614560.5181100\ttest: 9008927.0721736\tbest: 9008452.9642593 (1038)\ttotal: 50.9s\tremaining: 3m 13s\n",
            "1060:\tlearn: 4578791.4201479\ttest: 9004290.2967825\tbest: 9004191.8771286 (1058)\ttotal: 51.7s\tremaining: 3m 12s\n",
            "1080:\tlearn: 4540184.2578871\ttest: 9001962.0987314\tbest: 9001837.3138810 (1070)\ttotal: 53.5s\tremaining: 3m 13s\n",
            "1100:\tlearn: 4497356.6788031\ttest: 8998987.6629908\tbest: 8998269.3133162 (1093)\ttotal: 55.2s\tremaining: 3m 15s\n",
            "1120:\tlearn: 4457762.4321813\ttest: 8997397.4245483\tbest: 8997397.4245483 (1120)\ttotal: 56.5s\tremaining: 3m 15s\n",
            "1140:\tlearn: 4425109.4452328\ttest: 8993368.7218823\tbest: 8993368.7218823 (1140)\ttotal: 57.4s\tremaining: 3m 13s\n",
            "1160:\tlearn: 4391524.1673868\ttest: 8990531.4082663\tbest: 8990315.3503258 (1159)\ttotal: 58.2s\tremaining: 3m 12s\n",
            "1180:\tlearn: 4357312.1203355\ttest: 8989913.0966501\tbest: 8989529.5560574 (1174)\ttotal: 59.1s\tremaining: 3m 10s\n",
            "1200:\tlearn: 4323102.0253102\ttest: 8987764.7965372\tbest: 8987764.7965372 (1200)\ttotal: 59.9s\tremaining: 3m 9s\n",
            "1220:\tlearn: 4291516.0277012\ttest: 8983068.3417408\tbest: 8982591.4770808 (1217)\ttotal: 1m\tremaining: 3m 7s\n",
            "1240:\tlearn: 4259815.0279769\ttest: 8978404.3769298\tbest: 8978316.4479311 (1239)\ttotal: 1m 1s\tremaining: 3m 6s\n",
            "1260:\tlearn: 4227136.7524429\ttest: 8975069.6710501\tbest: 8974865.2426930 (1257)\ttotal: 1m 2s\tremaining: 3m 5s\n",
            "1280:\tlearn: 4193960.3102809\ttest: 8973914.9318979\tbest: 8972871.2449982 (1273)\ttotal: 1m 3s\tremaining: 3m 3s\n",
            "1300:\tlearn: 4165429.5628506\ttest: 8970383.7272079\tbest: 8970383.7272079 (1300)\ttotal: 1m 4s\tremaining: 3m 2s\n",
            "1320:\tlearn: 4138936.6742104\ttest: 8969001.2325449\tbest: 8968718.5519789 (1307)\ttotal: 1m 5s\tremaining: 3m 1s\n",
            "1340:\tlearn: 4111142.3340897\ttest: 8967528.3413452\tbest: 8967341.4137477 (1339)\ttotal: 1m 5s\tremaining: 2m 59s\n",
            "1360:\tlearn: 4081014.9131762\ttest: 8964081.5068261\tbest: 8964081.5068261 (1360)\ttotal: 1m 7s\tremaining: 3m\n",
            "1380:\tlearn: 4049805.1139626\ttest: 8963174.3367329\tbest: 8962383.5705674 (1377)\ttotal: 1m 9s\tremaining: 3m 1s\n",
            "1400:\tlearn: 4023302.4930140\ttest: 8961722.0821657\tbest: 8961530.0844328 (1396)\ttotal: 1m 10s\tremaining: 3m 1s\n",
            "1420:\tlearn: 3993393.2406893\ttest: 8959429.4073920\tbest: 8958467.0386494 (1410)\ttotal: 1m 11s\tremaining: 2m 59s\n",
            "1440:\tlearn: 3965984.5558190\ttest: 8957663.0400669\tbest: 8957253.4194543 (1438)\ttotal: 1m 12s\tremaining: 2m 58s\n",
            "1460:\tlearn: 3934309.8010507\ttest: 8955116.1129753\tbest: 8955116.1129753 (1460)\ttotal: 1m 13s\tremaining: 2m 57s\n",
            "1480:\tlearn: 3905285.6412732\ttest: 8955036.8798502\tbest: 8954426.9909410 (1470)\ttotal: 1m 13s\tremaining: 2m 55s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 8954426.991\n",
            "bestIteration = 1470\n",
            "\n",
            "Shrink model to first 1471 iterations.\n",
            "split KFold4\n",
            "RMSE 8954426.990941033\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1402758.4717981433, 'MAE': 2805517.1935952906, 'RMSE': 3894226.8057420133}, {'Huber:delta=0.5': 2695581.8700242257, 'MAE': 5391163.990047434, 'RMSE': 8954426.990941022}])\n",
            "0:\tlearn: 40383496.1985870\ttest: 40659982.1764329\tbest: 40659982.1764329 (0)\ttotal: 59.2ms\tremaining: 4m 55s\n",
            "20:\tlearn: 15466831.2791970\ttest: 15947577.3780889\tbest: 15947577.3780889 (20)\ttotal: 960ms\tremaining: 3m 47s\n",
            "40:\tlearn: 11311584.6809282\ttest: 12073076.7071795\tbest: 12073076.7071795 (40)\ttotal: 1.82s\tremaining: 3m 40s\n",
            "60:\tlearn: 10247588.8272839\ttest: 11235121.1193883\tbest: 11235121.1193883 (60)\ttotal: 2.67s\tremaining: 3m 36s\n",
            "80:\tlearn: 9715975.8332232\ttest: 10859376.4088736\tbest: 10859376.4088736 (80)\ttotal: 3.52s\tremaining: 3m 33s\n",
            "100:\tlearn: 9280142.0894392\ttest: 10603114.6213585\tbest: 10603114.6213585 (100)\ttotal: 4.4s\tremaining: 3m 33s\n",
            "120:\tlearn: 8922711.9863953\ttest: 10392060.9062953\tbest: 10392060.9062953 (120)\ttotal: 5.28s\tremaining: 3m 32s\n",
            "140:\tlearn: 8610272.3231537\ttest: 10221972.1335362\tbest: 10221972.1335362 (140)\ttotal: 6.99s\tremaining: 4m 1s\n",
            "160:\tlearn: 8301123.7924927\ttest: 10076804.1375098\tbest: 10076804.1375098 (160)\ttotal: 8.65s\tremaining: 4m 20s\n",
            "180:\tlearn: 8055154.2414553\ttest: 9971520.4534913\tbest: 9971520.4534913 (180)\ttotal: 9.98s\tremaining: 4m 25s\n",
            "200:\tlearn: 7840720.7764685\ttest: 9877244.1137261\tbest: 9877244.1137261 (200)\ttotal: 10.8s\tremaining: 4m 18s\n",
            "220:\tlearn: 7632328.9722283\ttest: 9780054.7281825\tbest: 9780054.7281825 (220)\ttotal: 11.7s\tremaining: 4m 12s\n",
            "240:\tlearn: 7468390.4916284\ttest: 9724618.2706852\tbest: 9724618.2706852 (240)\ttotal: 12.5s\tremaining: 4m 7s\n",
            "260:\tlearn: 7308515.5965885\ttest: 9678694.1781046\tbest: 9678694.1781046 (260)\ttotal: 13.4s\tremaining: 4m 4s\n",
            "280:\tlearn: 7164448.1653746\ttest: 9629574.4469375\tbest: 9629574.4469375 (280)\ttotal: 14.4s\tremaining: 4m 1s\n",
            "300:\tlearn: 7022623.9569531\ttest: 9587873.5106969\tbest: 9587873.5106969 (300)\ttotal: 15.2s\tremaining: 3m 57s\n",
            "320:\tlearn: 6904750.1688986\ttest: 9552789.5317026\tbest: 9552789.5317026 (320)\ttotal: 16.4s\tremaining: 3m 58s\n",
            "340:\tlearn: 6793420.4141239\ttest: 9519631.0977208\tbest: 9519631.0977208 (340)\ttotal: 17.3s\tremaining: 3m 56s\n",
            "360:\tlearn: 6670942.5494093\ttest: 9490469.5392930\tbest: 9490469.5392930 (360)\ttotal: 18.2s\tremaining: 3m 54s\n",
            "380:\tlearn: 6560285.2390768\ttest: 9468744.5206931\tbest: 9467374.4898497 (378)\ttotal: 19.4s\tremaining: 3m 54s\n",
            "400:\tlearn: 6471762.4664360\ttest: 9450280.4658541\tbest: 9450280.4658541 (400)\ttotal: 21.1s\tremaining: 4m 2s\n",
            "420:\tlearn: 6369768.8375320\ttest: 9429495.5365069\tbest: 9429495.5365069 (420)\ttotal: 22.8s\tremaining: 4m 8s\n",
            "440:\tlearn: 6269980.5255337\ttest: 9406879.7235234\tbest: 9406462.9238411 (439)\ttotal: 24.1s\tremaining: 4m 9s\n",
            "460:\tlearn: 6186236.8351994\ttest: 9395255.5141769\tbest: 9395255.5141769 (460)\ttotal: 25s\tremaining: 4m 6s\n",
            "480:\tlearn: 6107756.5063780\ttest: 9375014.2993689\tbest: 9375014.2993689 (480)\ttotal: 25.8s\tremaining: 4m 2s\n",
            "500:\tlearn: 6028845.2667620\ttest: 9363384.9942453\tbest: 9363384.9942453 (500)\ttotal: 26.7s\tremaining: 3m 59s\n",
            "520:\tlearn: 5958430.5819244\ttest: 9349379.7962347\tbest: 9349379.7962347 (520)\ttotal: 27.8s\tremaining: 3m 58s\n",
            "540:\tlearn: 5893575.1536156\ttest: 9344782.9788534\tbest: 9344782.9788534 (540)\ttotal: 28.6s\tremaining: 3m 55s\n",
            "560:\tlearn: 5823155.0771095\ttest: 9331822.9839282\tbest: 9331822.9839282 (560)\ttotal: 29.5s\tremaining: 3m 53s\n",
            "580:\tlearn: 5762716.8014195\ttest: 9326853.7684111\tbest: 9326853.7684111 (580)\ttotal: 30.3s\tremaining: 3m 50s\n",
            "600:\tlearn: 5696989.2555049\ttest: 9317772.3348591\tbest: 9317554.0363470 (596)\ttotal: 31.1s\tremaining: 3m 47s\n",
            "620:\tlearn: 5639675.4205285\ttest: 9315779.5460031\tbest: 9315779.5460031 (620)\ttotal: 32s\tremaining: 3m 45s\n",
            "640:\tlearn: 5574461.2661288\ttest: 9307037.3928894\tbest: 9306701.4851371 (639)\ttotal: 32.8s\tremaining: 3m 43s\n",
            "660:\tlearn: 5511843.9519658\ttest: 9296886.8410730\tbest: 9296886.8410730 (660)\ttotal: 33.8s\tremaining: 3m 41s\n",
            "680:\tlearn: 5456723.2279809\ttest: 9285797.2985112\tbest: 9284681.4199115 (678)\ttotal: 35.5s\tremaining: 3m 45s\n",
            "700:\tlearn: 5399750.8818580\ttest: 9275054.2607614\tbest: 9275054.2607614 (700)\ttotal: 37.1s\tremaining: 3m 47s\n",
            "720:\tlearn: 5343775.7917803\ttest: 9272001.3688129\tbest: 9270926.9929277 (714)\ttotal: 38.4s\tremaining: 3m 47s\n",
            "740:\tlearn: 5284003.6729729\ttest: 9263312.3090999\tbest: 9263312.3090999 (740)\ttotal: 39.3s\tremaining: 3m 45s\n",
            "760:\tlearn: 5232189.2769184\ttest: 9257516.3056103\tbest: 9255357.2946750 (758)\ttotal: 40.1s\tremaining: 3m 43s\n",
            "780:\tlearn: 5176539.1728794\ttest: 9248963.3376721\tbest: 9248963.3376721 (780)\ttotal: 41s\tremaining: 3m 41s\n",
            "800:\tlearn: 5129845.8197438\ttest: 9247645.8598003\tbest: 9246792.4519928 (796)\ttotal: 41.9s\tremaining: 3m 39s\n",
            "820:\tlearn: 5081939.1875990\ttest: 9238543.8572955\tbest: 9238543.8572955 (820)\ttotal: 42.7s\tremaining: 3m 37s\n",
            "840:\tlearn: 5034528.7470734\ttest: 9238264.9002391\tbest: 9237684.9991038 (826)\ttotal: 43.6s\tremaining: 3m 35s\n",
            "860:\tlearn: 4986462.1649661\ttest: 9234986.6941230\tbest: 9234587.3773673 (855)\ttotal: 44.5s\tremaining: 3m 34s\n",
            "880:\tlearn: 4941440.9890649\ttest: 9234406.8507768\tbest: 9233020.5362190 (865)\ttotal: 45.4s\tremaining: 3m 32s\n",
            "Stopped by overfitting detector  (20 iterations wait)\n",
            "\n",
            "bestTest = 9233020.536\n",
            "bestIteration = 865\n",
            "\n",
            "Shrink model to first 866 iterations.\n",
            "split KFold5\n",
            "RMSE 9233020.536219014\n",
            "Model RMSE: dict_values([{'Huber:delta=0.5': 1737086.7892954715, 'MAE': 3474173.8285899246, 'RMSE': 4929532.737747312}, {'Huber:delta=0.5': 2828146.883308257, 'MAE': 5656294.016615511, 'RMSE': 9233020.536219014}])\n",
            "CV_RMSE_Score:9046643.316451298\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "KFold_preds = pd.DataFrame(bsv_1).join(TT['target'])\n",
        "KFold_preds_1 = pd.DataFrame(bsv_2).join(TT_1['target'].reset_index(drop=True))\n",
        "KFold_preds_2 = pd.DataFrame(bsv_3).join(TT_2['target'].reset_index(drop=True))\n",
        "KFold_preds_3 = pd.DataFrame(bsv_4).join(TT_3['target'].reset_index(drop=True))\n",
        "KFold_preds_4 = pd.DataFrame(bsv_5).join(TT_4['target'].reset_index(drop=True))"
      ],
      "metadata": {
        "id": "CGzyW2Y4dn6x"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Baseline = Baseline_predict(train=TT,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_1,bsv_test1 = Baseline.predict(tests)\n",
        "\n",
        "\n",
        "Baseline = Baseline_predict(train=TT_1,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_2,bsv_test2 = Baseline.predict(tests)\n",
        "\n",
        "\n",
        "Baseline = Baseline_predict(train=TT_2,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_3,bsv_test3 = Baseline.predict(tests)\n",
        "\n",
        "\n",
        "Baseline = Baseline_predict(train=TT_3,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_4,bsv_test4 = Baseline.predict(tests)\n",
        "\n",
        "\n",
        "Baseline = Baseline_predict(train=TT_4,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_5,bsv_test5 = Baseline.predict(tests)\n"
      ],
      "metadata": {
        "id": "Snh-ZsHuGKUC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_preds = pd.DataFrame(bsv_test1)\n",
        "test_preds_1 = pd.DataFrame(bsv_test2)\n",
        "test_preds_2 = pd.DataFrame(bsv_test3)\n",
        "test_preds_3 = pd.DataFrame(bsv_test4)\n",
        "test_preds_4 = pd.DataFrame(bsv_test5)"
      ],
      "metadata": {
        "id": "w3S8Kolljl3l"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Baseline = Baseline_predict(train=KFold_preds,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_11,bsv_test11 = Baseline.predict(test_preds)\n",
        "\n",
        "\n",
        "Baseline = Baseline_predict(train=KFold_preds_1,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_12,bsv_test12 = Baseline.predict(test_preds_1)\n",
        "\n",
        "\n",
        "Baseline = Baseline_predict(train=KFold_preds_2,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_13,bsv_test13 = Baseline.predict(test_preds_2)\n",
        "\n",
        "\n",
        "Baseline = Baseline_predict(train=KFold_preds_3,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_14,bsv_test14 = Baseline.predict(test_preds_3)\n",
        "\n",
        "\n",
        "Baseline = Baseline_predict(train=KFold_preds_4,model=cbc,split_method='KFold',num_split=5,return_baseline=True,VERBOSE=0,random_seed=0)\n",
        "predict_15,bsv_test15 = Baseline.predict(test_preds_4)"
      ],
      "metadata": {
        "id": "5FBfUVzQQgz7"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ensemble_pred = sum(predict_1)/5 + sum(predict_2)/5 + sum(predict_3)/5 + sum(predict_4)/5  + sum(predict_5)/5 + sum(predict_11)/5 + sum(predict_12)/5 + sum(predict_13)/5  + sum(predict_14)/5 + sum(predict_15)/5"
      ],
      "metadata": {
        "id": "V6Z1iAtXkwcP"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ensemble_pred/10"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0GfjoBK_lOd3",
        "outputId": "ee17c95f-58af-4895-8897-460a32ead3cb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([22962739.74011078, 61687235.422408  , 82657596.30170731, ...,\n",
              "       20643754.58625842, 14300308.32036489, 33680426.76940422])"
            ]
          },
          "metadata": {},
          "execution_count": 168
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictions_df = pd.DataFrame({'id': Test.id, 'target': ensemble_pred/10})\n",
        "predictions_df.to_csv('40th_sub.csv',index=False)"
      ],
      "metadata": {
        "id": "HYf872jRQZ_R"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Baseline = Baseline_score(train=TT_4,model=cbc,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "lgb_score,bsv_1 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "KFold_preds = pd.DataFrame(bsv_1)\n",
        "KFold_preds = KFold_preds.join(TT_4['target'].reset_index(drop=True))\n",
        "\n",
        "Baseline = Baseline_score(train=TT_4,model=cbc1,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "lgb_score,bsv_2 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "KFold_preds_1 = pd.DataFrame(bsv_2)\n",
        "KFold_preds_1 = KFold_preds_1.join(TT_4['target'].reset_index(drop=True))\n",
        "\n",
        "predictions = Baseline_predict(train=TT_4,model=cbc,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "predict_1,bsv_test1 = predictions.predict(tests)\n",
        "\n",
        "predictions = Baseline_predict(train=TT_4,model=cbc1,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "predict_2,bsv_test2 = predictions.predict(tests)\n",
        "\n",
        "testpreds_1 = pd.DataFrame(bsv_test1)\n",
        "testpreds_2 = pd.DataFrame(bsv_test2)\n",
        "\n",
        "predictions = Baseline_predict(train=KFold_preds,model=cbc,split_method='KFold',return_baseline=False,VERBOSE=0,random_seed=0)\n",
        "predict_3 = predictions.predict(testpreds_1)\n",
        "\n",
        "predictions = Baseline_predict(train=KFold_preds_1,model=cbc1,split_method='KFold',return_baseline=False,VERBOSE=0,random_seed=0)\n",
        "predict_4 = predictions.predict(testpreds_2)\n",
        "\n",
        "ensemble_pred = sum(predict_1)/15 + sum(predict_2)/15 + sum(predict_3)/5 + sum(predict_4)/5\n",
        "final_preds = ensemble_pred/4"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QXcZGO5Pr4Y_",
        "outputId": "82b6dbda-0e4a-42a0-a553-55fb9cb6d982"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CV_RMSE_Score:9204054.26349729\n",
            "CV_RMSE_Score:9180576.968995651\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "Baseline = Baseline_score(train=TT_3,model=cbc,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "lgb_score,bsv_1 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "KFold_preds = pd.DataFrame(bsv_1)\n",
        "KFold_preds = KFold_preds.join(TT_3['target'].reset_index(drop=True))\n",
        "\n",
        "Baseline = Baseline_score(train=TT_3,model=cbc1,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "lgb_score,bsv_2 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "KFold_preds_1 = pd.DataFrame(bsv_2)\n",
        "KFold_preds_1 = KFold_preds_1.join(TT_3['target'].reset_index(drop=True))\n",
        "\n",
        "predictions = Baseline_predict(train=TT_3,model=cbc,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "predict_1,bsv_test1 = predictions.predict(tests)\n",
        "\n",
        "predictions = Baseline_predict(train=TT_3,model=cbc1,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "predict_2,bsv_test2 = predictions.predict(tests)\n",
        "\n",
        "testpreds_1 = pd.DataFrame(bsv_test1)\n",
        "testpreds_2 = pd.DataFrame(bsv_test2)\n",
        "\n",
        "predictions = Baseline_predict(train=KFold_preds,model=cbc,split_method='KFold',return_baseline=False,VERBOSE=0,random_seed=0)\n",
        "predict_3 = predictions.predict(testpreds_1)\n",
        "\n",
        "predictions = Baseline_predict(train=KFold_preds_1,model=cbc1,split_method='KFold',return_baseline=False,VERBOSE=0,random_seed=0)\n",
        "predict_4 = predictions.predict(testpreds_2)\n",
        "\n",
        "ensemble_pred = sum(predict_1)/15 + sum(predict_2)/15 + sum(predict_3)/5 + sum(predict_4)/5\n",
        "final_preds_3 = ensemble_pred/4"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lTochCBpe3g_",
        "outputId": "6746317c-4f3a-42fa-a73a-683e5d434e64"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CV_RMSE_Score:9170087.915600324\n",
            "CV_RMSE_Score:9130755.388979923\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tpMTyWrmieTN"
      },
      "source": [
        "###Data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-Itz-Fq-6r2r"
      },
      "outputs": [],
      "source": [
        "Test['target'] = 'Test'\n",
        "data = pd.concat([Train,Test]).reset_index(drop=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kvqeG7348Zy9"
      },
      "source": [
        "###Data Visualization\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 272
        },
        "id": "FnIIVy_W8kmh",
        "outputId": "38db92ad-3e5a-451e-cb56-956db60b0cb4"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-f2a36b07-55b8-4963-bed2-8dd1df60bcab\">\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>id</th>\n",
              "      <th>timestamp</th>\n",
              "      <th>device</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>...</th>\n",
              "      <th>dewPoint</th>\n",
              "      <th>humidity</th>\n",
              "      <th>pressure</th>\n",
              "      <th>windSpeed</th>\n",
              "      <th>cloudCover</th>\n",
              "      <th>uvIndex</th>\n",
              "      <th>visibility</th>\n",
              "      <th>Traffic Jam Factor</th>\n",
              "      <th>area</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Id_qt2qn56050</td>\n",
              "      <td>1624367008</td>\n",
              "      <td>pc1</td>\n",
              "      <td>-84.498750</td>\n",
              "      <td>-14.586875</td>\n",
              "      <td>-49.855625</td>\n",
              "      <td>16.457</td>\n",
              "      <td>35076.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>...</td>\n",
              "      <td>13.86</td>\n",
              "      <td>0.62</td>\n",
              "      <td>1013.6</td>\n",
              "      <td>2.39</td>\n",
              "      <td>0.95</td>\n",
              "      <td>4.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>0.38979</td>\n",
              "      <td>Residential</td>\n",
              "      <td>51500000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Id_f1zf07cwb6</td>\n",
              "      <td>1624372465</td>\n",
              "      <td>pc3</td>\n",
              "      <td>-86.818125</td>\n",
              "      <td>-11.982500</td>\n",
              "      <td>-54.838125</td>\n",
              "      <td>16.674</td>\n",
              "      <td>42704.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.52</td>\n",
              "      <td>0.59</td>\n",
              "      <td>1013.5</td>\n",
              "      <td>2.29</td>\n",
              "      <td>0.82</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>3.78322</td>\n",
              "      <td>Residential</td>\n",
              "      <td>50300000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Id_uoxdz7di5b</td>\n",
              "      <td>1624371871</td>\n",
              "      <td>pc3</td>\n",
              "      <td>-101.916250</td>\n",
              "      <td>-13.257500</td>\n",
              "      <td>-71.241250</td>\n",
              "      <td>9.392</td>\n",
              "      <td>36017.0</td>\n",
              "      <td>22.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>...</td>\n",
              "      <td>14.71</td>\n",
              "      <td>0.61</td>\n",
              "      <td>1013.8</td>\n",
              "      <td>2.13</td>\n",
              "      <td>0.82</td>\n",
              "      <td>3.0</td>\n",
              "      <td>16.093</td>\n",
              "      <td>2.08425</td>\n",
              "      <td>Residential</td>\n",
              "      <td>16500000.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>3 rows × 41 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f2a36b07-55b8-4963-bed2-8dd1df60bcab')\"\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-f2a36b07-55b8-4963-bed2-8dd1df60bcab 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-f2a36b07-55b8-4963-bed2-8dd1df60bcab');\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",
              "  "
            ],
            "text/plain": [
              "              id   timestamp device  PCell_RSRP_max  PCell_RSRQ_max  \\\n",
              "0  Id_qt2qn56050  1624367008    pc1      -84.498750      -14.586875   \n",
              "1  Id_f1zf07cwb6  1624372465    pc3      -86.818125      -11.982500   \n",
              "2  Id_uoxdz7di5b  1624371871    pc3     -101.916250      -13.257500   \n",
              "\n",
              "   PCell_RSSI_max  PCell_SNR_max  PCell_Downlink_Num_RBs  \\\n",
              "0      -49.855625         16.457                 35076.0   \n",
              "1      -54.838125         16.674                 42704.0   \n",
              "2      -71.241250          9.392                 36017.0   \n",
              "\n",
              "   PCell_Downlink_Average_MCS  PCell_Downlink_bandwidth_MHz  ...  dewPoint  \\\n",
              "0                        23.0                          20.0  ...     13.86   \n",
              "1                        24.0                          20.0  ...     14.52   \n",
              "2                        22.0                          20.0  ...     14.71   \n",
              "\n",
              "   humidity  pressure  windSpeed  cloudCover  uvIndex  visibility  \\\n",
              "0      0.62    1013.6       2.39        0.95      4.0      16.093   \n",
              "1      0.59    1013.5       2.29        0.82      3.0      16.093   \n",
              "2      0.61    1013.8       2.13        0.82      3.0      16.093   \n",
              "\n",
              "   Traffic Jam Factor         area      target  \n",
              "0             0.38979  Residential  51500000.0  \n",
              "1             3.78322  Residential  50300000.0  \n",
              "2             2.08425  Residential  16500000.0  \n",
              "\n",
              "[3 rows x 41 columns]"
            ]
          },
          "execution_count": 343,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "data.head(3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fjK3iHoF-Ibp"
      },
      "outputs": [],
      "source": [
        "Vard = Variables.set_index('VARIABLENAME').T"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 232
        },
        "id": "PsU-AkUP68n2",
        "outputId": "2de04859-b583-402a-d08c-84be82d679e4"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-e3b2d288-4ab7-4255-8a87-94da53a166fc\">\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>VARIABLENAME</th>\n",
              "      <th>PCell_RSRP_max</th>\n",
              "      <th>PCell_RSRQ_max</th>\n",
              "      <th>PCell_RSSI_max</th>\n",
              "      <th>PCell_SNR_max</th>\n",
              "      <th>PCell_Downlink_Num_RBs</th>\n",
              "      <th>PCell_Downlink_Average_MCS</th>\n",
              "      <th>PCell_Downlink_bandwidth_MHz</th>\n",
              "      <th>PCell_Cell_Identity</th>\n",
              "      <th>PCell_freq_MHz</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>VARIABLE DESCRIPTION</th>\n",
              "      <td>Maximum reference signal received power in dBm from the primary cell</td>\n",
              "      <td>Maximum reference signal received quality in dB from the primary cell</td>\n",
              "      <td>Maximum received signal strength indicator in dBm from the primary cell</td>\n",
              "      <td>Maximum signal to noise ratio in dB from the primary cell</td>\n",
              "      <td>Aggregated number of received resource blocks in downlink from the primary cell</td>\n",
              "      <td>Average modulation and coding scheme weighted by received resource blocks from the primary cell</td>\n",
              "      <td>Downlink bandwidth in MHz from the primary cell</td>\n",
              "      <td>Cell ID of the primary cell</td>\n",
              "      <td>Carrier frequency in MHz of the primary cell</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e3b2d288-4ab7-4255-8a87-94da53a166fc')\"\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-e3b2d288-4ab7-4255-8a87-94da53a166fc 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-e3b2d288-4ab7-4255-8a87-94da53a166fc');\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",
              "  "
            ],
            "text/plain": [
              "VARIABLENAME                                                                PCell_RSRP_max  \\\n",
              "VARIABLE DESCRIPTION  Maximum reference signal received power in dBm from the primary cell   \n",
              "\n",
              "VARIABLENAME                                                                 PCell_RSRQ_max  \\\n",
              "VARIABLE DESCRIPTION  Maximum reference signal received quality in dB from the primary cell   \n",
              "\n",
              "VARIABLENAME                                                                   PCell_RSSI_max  \\\n",
              "VARIABLE DESCRIPTION  Maximum received signal strength indicator in dBm from the primary cell   \n",
              "\n",
              "VARIABLENAME                                                      PCell_SNR_max  \\\n",
              "VARIABLE DESCRIPTION  Maximum signal to noise ratio in dB from the primary cell   \n",
              "\n",
              "VARIABLENAME                                                                   PCell_Downlink_Num_RBs  \\\n",
              "VARIABLE DESCRIPTION  Aggregated number of received resource blocks in downlink from the primary cell   \n",
              "\n",
              "VARIABLENAME                                                                               PCell_Downlink_Average_MCS  \\\n",
              "VARIABLE DESCRIPTION  Average modulation and coding scheme weighted by received resource blocks from the primary cell   \n",
              "\n",
              "VARIABLENAME                             PCell_Downlink_bandwidth_MHz  \\\n",
              "VARIABLE DESCRIPTION  Downlink bandwidth in MHz from the primary cell   \n",
              "\n",
              "VARIABLENAME                   PCell_Cell_Identity  \\\n",
              "VARIABLE DESCRIPTION  Cell ID of the primary cell    \n",
              "\n",
              "VARIABLENAME                                        PCell_freq_MHz  \n",
              "VARIABLE DESCRIPTION  Carrier frequency in MHz of the primary cell  "
            ]
          },
          "execution_count": 16,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "Vard[data.filter(like=\"PCell\").columns]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Iot3QN2AVmsN"
      },
      "outputs": [],
      "source": [
        "data[['PCell_Downlink_Average_MCS','PCell_SNR_max','PCell_Downlink_Num_RBs']].describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "iHR85-bY1Ee5"
      },
      "outputs": [],
      "source": [
        "g = sns.PairGrid(data[['PCell_Downlink_Average_MCS','PCell_SNR_max','PCell_Downlink_Num_RBs']])\n",
        "g.map_diag(sns.histplot)\n",
        "g.map_offdiag(sns.scatterplot)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "SmxLQvP2z0bv"
      },
      "outputs": [],
      "source": [
        "scatterplot(x=Train['PCell_Downlink_Average_MCS'],y=Train['PCell_SNR_max'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "d8q2xkUi3j1K"
      },
      "outputs": [],
      "source": [
        "g = sns.PairGrid(data[['PCell_Downlink_Average_MCS','PCell_RSRP_max','PCell_RSRQ_max','PCell_RSSI_max','PCell_SNR_max']])\n",
        "g.map_diag(sns.histplot)\n",
        "g.map_offdiag(sns.scatterplot)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Arndm51853mI"
      },
      "outputs": [],
      "source": [
        "data[['PCell_RSRP_max','PCell_RSRQ_max','PCell_RSSI_max']].describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "or9JJ7Z5-gs0"
      },
      "outputs": [],
      "source": [
        "data[['PCell_Downlink_Average_MCS','PCell_RSRP_max','PCell_RSRQ_max','PCell_RSSI_max','PCell_SNR_max']].head(3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "p0WdpdFkeNHU"
      },
      "outputs": [],
      "source": [
        "data[['SCell_SNR_max','SCell_Downlink_Num_RBs','SCell_Cell_Identity','SCell_RSRQ_max']].describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "EhKecWbTxvcH"
      },
      "outputs": [],
      "source": [
        "g = sns.PairGrid(data[['SCell_SNR_max','SCell_Downlink_Num_RBs','SCell_Cell_Identity','SCell_RSRQ_max']])\n",
        "g.map_diag(sns.histplot)\n",
        "g.map_offdiag(sns.scatterplot)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "uR411L-nxPOy"
      },
      "outputs": [],
      "source": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "h5BYgVGGAV7-"
      },
      "source": [
        "###Feature Engineering"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "U5VMZS-p8UeF",
        "outputId": "ed319619-9616-4e45-8a0e-53f90d258906"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                    VARIABLENAME  \\\n",
              "0                             id   \n",
              "1                      timestamp   \n",
              "2                         device   \n",
              "3                 PCell_RSRP_max   \n",
              "4                 PCell_RSRQ_max   \n",
              "5                 PCell_RSSI_max   \n",
              "6                  PCell_SNR_max   \n",
              "7         PCell_Downlink_Num_RBs   \n",
              "8         PCell_Downlink_TB_Size   \n",
              "9     PCell_Downlink_Average_MCS   \n",
              "10  PCell_Downlink_bandwidth_MHz   \n",
              "11           PCell_Cell_Identity   \n",
              "12                PCell_freq_MHz   \n",
              "13                SCell_RSRP_max   \n",
              "14                SCell_RSRQ_max   \n",
              "15                SCell_RSSI_max   \n",
              "16                 SCell_SNR_max   \n",
              "17        SCell_Downlink_Num_RBs   \n",
              "18        SCell_Downlink_TB_Size   \n",
              "19    SCell_Downlink_Average_MCS   \n",
              "20  SCell_Downlink_bandwidth_MHz   \n",
              "21           SCell_Cell_Identity   \n",
              "22                SCell_freq_MHz   \n",
              "23                      operator   \n",
              "24                      Latitude   \n",
              "25                     Longitude   \n",
              "26                      Altitude   \n",
              "27                     speed_kmh   \n",
              "28                           COG   \n",
              "29               precipIntensity   \n",
              "30             precipProbability   \n",
              "31                   temperature   \n",
              "32           apparentTemperature   \n",
              "33                      dewPoint   \n",
              "34                      humidity   \n",
              "35                      pressure   \n",
              "36                     windSpeed   \n",
              "37                    cloudCover   \n",
              "38                       uvIndex   \n",
              "39                    visibility   \n",
              "40            Traffic Jam Factor   \n",
              "41                          area   \n",
              "42                        target   \n",
              "\n",
              "                                                                                 VARIABLE DESCRIPTION  \n",
              "0                                                                              Unique ID for each row  \n",
              "1                                                                           Unix timestamp in seconds  \n",
              "2                                                                   ID of the vehicle carrying the UE  \n",
              "3                                Maximum reference signal received power in dBm from the primary cell  \n",
              "4                               Maximum reference signal received quality in dB from the primary cell  \n",
              "5                             Maximum received signal strength indicator in dBm from the primary cell  \n",
              "6                                           Maximum signal to noise ratio in dB from the primary cell  \n",
              "7                     Aggregated number of received resource blocks in downlink from the primary cell  \n",
              "8                                   Aggregated transport block size in downlink from the primary cell  \n",
              "9     Average modulation and coding scheme weighted by received resource blocks from the primary cell  \n",
              "10                                                    Downlink bandwidth in MHz from the primary cell  \n",
              "11                                                                       Cell ID of the primary cell   \n",
              "12                                                       Carrier frequency in MHz of the primary cell  \n",
              "13                             Maximum reference signal received power in dBm from the secondary cell  \n",
              "14                            Maximum reference signal received quality in dB from the secondary cell  \n",
              "15                          Maximum received signal strength indicator in dBm from the secondary cell  \n",
              "16                                        Maximum signal to noise ratio in dB from the secondary cell  \n",
              "17                  Aggregated number of received resource blocks in downlink from the secondary cell  \n",
              "18                                Aggregated transport block size in downlink from the secondary cell  \n",
              "19  Average modulation and coding scheme weighted by received resource blocks from the secondary cell  \n",
              "20                                                  Downlink bandwidth in MHz from the secondary cell  \n",
              "21                                                                     Cell ID of the secondary cell   \n",
              "22                                                     Carrier frequency in MHz of the secondary cell  \n",
              "23                                                                                 ID of the operator  \n",
              "24                                                                        GPS latitude of the vehicle  \n",
              "25                                                                       GPS longitude of the vehicle  \n",
              "26                                                                        GPS altitude of the vehicle  \n",
              "27                                                             GPS-based speed of the vehicle in km/h  \n",
              "28                                                        GPS-based course over ground of the vehicle  \n",
              "29                                                                            Precipitation intensity  \n",
              "30                                                                          Precipitation probability  \n",
              "31                                                                                        Temperature  \n",
              "32                                                                               Apparent temperature  \n",
              "33                                                                                          Dew point  \n",
              "34                                                                                           Humidity  \n",
              "35                                                                               Atmospheric pressure  \n",
              "36                                                                                         Wind speed  \n",
              "37                                                                                        Cloud cover  \n",
              "38                                                                                  Ultraviolet index  \n",
              "39                                                               Visibility due to weather conditions  \n",
              "40                      Factor of traffic congestion (a higher factor corresponds to heavier traffic)  \n",
              "41                                Categorical urban area among Avenue/Park/Residential/Highway/Tunnel  \n",
              "42                                            Prediction target: downlink datarate in bits per second  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-d352507f-9445-4d91-9cae-ca1598d986d8\">\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>VARIABLENAME</th>\n",
              "      <th>VARIABLE DESCRIPTION</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>id</td>\n",
              "      <td>Unique ID for each row</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>timestamp</td>\n",
              "      <td>Unix timestamp in seconds</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>device</td>\n",
              "      <td>ID of the vehicle carrying the UE</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>PCell_RSRP_max</td>\n",
              "      <td>Maximum reference signal received power in dBm from the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>PCell_RSRQ_max</td>\n",
              "      <td>Maximum reference signal received quality in dB from the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>PCell_RSSI_max</td>\n",
              "      <td>Maximum received signal strength indicator in dBm from the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>PCell_SNR_max</td>\n",
              "      <td>Maximum signal to noise ratio in dB from the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>PCell_Downlink_Num_RBs</td>\n",
              "      <td>Aggregated number of received resource blocks in downlink from the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>PCell_Downlink_TB_Size</td>\n",
              "      <td>Aggregated transport block size in downlink from the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>PCell_Downlink_Average_MCS</td>\n",
              "      <td>Average modulation and coding scheme weighted by received resource blocks from the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>PCell_Downlink_bandwidth_MHz</td>\n",
              "      <td>Downlink bandwidth in MHz from the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>PCell_Cell_Identity</td>\n",
              "      <td>Cell ID of the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>PCell_freq_MHz</td>\n",
              "      <td>Carrier frequency in MHz of the primary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>SCell_RSRP_max</td>\n",
              "      <td>Maximum reference signal received power in dBm from the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>SCell_RSRQ_max</td>\n",
              "      <td>Maximum reference signal received quality in dB from the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>SCell_RSSI_max</td>\n",
              "      <td>Maximum received signal strength indicator in dBm from the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>SCell_SNR_max</td>\n",
              "      <td>Maximum signal to noise ratio in dB from the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>SCell_Downlink_Num_RBs</td>\n",
              "      <td>Aggregated number of received resource blocks in downlink from the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>SCell_Downlink_TB_Size</td>\n",
              "      <td>Aggregated transport block size in downlink from the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>SCell_Downlink_Average_MCS</td>\n",
              "      <td>Average modulation and coding scheme weighted by received resource blocks from the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>SCell_Downlink_bandwidth_MHz</td>\n",
              "      <td>Downlink bandwidth in MHz from the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>SCell_Cell_Identity</td>\n",
              "      <td>Cell ID of the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>SCell_freq_MHz</td>\n",
              "      <td>Carrier frequency in MHz of the secondary cell</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>operator</td>\n",
              "      <td>ID of the operator</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>Latitude</td>\n",
              "      <td>GPS latitude of the vehicle</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>Longitude</td>\n",
              "      <td>GPS longitude of the vehicle</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>Altitude</td>\n",
              "      <td>GPS altitude of the vehicle</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>speed_kmh</td>\n",
              "      <td>GPS-based speed of the vehicle in km/h</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>COG</td>\n",
              "      <td>GPS-based course over ground of the vehicle</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>precipIntensity</td>\n",
              "      <td>Precipitation intensity</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>precipProbability</td>\n",
              "      <td>Precipitation probability</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>temperature</td>\n",
              "      <td>Temperature</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>apparentTemperature</td>\n",
              "      <td>Apparent temperature</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>dewPoint</td>\n",
              "      <td>Dew point</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>humidity</td>\n",
              "      <td>Humidity</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>pressure</td>\n",
              "      <td>Atmospheric pressure</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>windSpeed</td>\n",
              "      <td>Wind speed</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>cloudCover</td>\n",
              "      <td>Cloud cover</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>uvIndex</td>\n",
              "      <td>Ultraviolet index</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39</th>\n",
              "      <td>visibility</td>\n",
              "      <td>Visibility due to weather conditions</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>Traffic Jam Factor</td>\n",
              "      <td>Factor of traffic congestion (a higher factor corresponds to heavier traffic)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>area</td>\n",
              "      <td>Categorical urban area among Avenue/Park/Residential/Highway/Tunnel</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42</th>\n",
              "      <td>target</td>\n",
              "      <td>Prediction target: downlink datarate in bits per second</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d352507f-9445-4d91-9cae-ca1598d986d8')\"\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-d352507f-9445-4d91-9cae-ca1598d986d8 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-d352507f-9445-4d91-9cae-ca1598d986d8');\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": 467
        }
      ],
      "source": [
        "Variables"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OhI1-HoBAYVG"
      },
      "outputs": [],
      "source": [
        "data['N_PCell_RSRQ'] = data['PCell_RSRP_max'] / data['PCell_RSSI_max']\n",
        "data['N_PCell_RSRP_Noise'] = data['PCell_RSRP_max'] / data['PCell_SNR_max']\n",
        "data['N_PCell_RSRQ_Noise'] = data['PCell_RSRQ_max'] / data['PCell_SNR_max']\n",
        "data['N_PCell_RSSI_Noise'] = data['PCell_RSSI_max'] / data['PCell_SNR_max']\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "gTIQ09dBVOX5"
      },
      "outputs": [],
      "source": [
        "data['N_SCell_RSRQ'] = data['SCell_RSRP_max'] / data['SCell_RSSI_max']\n",
        "data['N_SCell_RSRP_Noise'] = data['SCell_RSRP_max'] / data['SCell_SNR_max']\n",
        "data['N_SCell_RSRQ_Noise'] = data['SCell_RSRQ_max'] / data['SCell_SNR_max']\n",
        "data['N_SCell_RSSI_Noise'] = data['SCell_RSSI_max'] / data['SCell_SNR_max']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OBkEeCF_AZI8"
      },
      "outputs": [],
      "source": [
        "def RSRmapper(rsr_col: pd.Series, h_bar: float, l_bar: float, colval: float) -> int:\n",
        "  \"\"\"\n",
        "  This Function maps the values of a Referenced_Signal_Received(RSR) column\n",
        "  to five different integer values between the intervals of h_bar(Highest possible value) and\n",
        "  l_bar(Lowest possible Value) and returns the label for the colval\n",
        "\n",
        "  Args:\n",
        "   rsr_col : Referenced signal received column\n",
        "   h_bar : Highest possible column value\n",
        "   l_bar : Lowest possible column value\n",
        "   colval : Column Value to turn into a interval label\n",
        "\n",
        "  Returns:\n",
        "   The Interval Label of the Column Value\n",
        "\n",
        "  \"\"\"\n",
        "\n",
        "  common_diff = abs(h_bar - l_bar)/5\n",
        "  if (h_bar-common_diff) <= colval <= h_bar :\n",
        "    return 6\n",
        "  elif (h_bar - 2*common_diff) <= colval <= (h_bar - common_diff):\n",
        "    return 5\n",
        "  elif (h_bar - 3*common_diff) <= colval <= (h_bar - 2*common_diff):\n",
        "    return 4\n",
        "  elif (h_bar - 4*common_diff) <= colval <= (h_bar - 3*common_diff):\n",
        "    return 3\n",
        "  elif (h_bar - 5*common_diff) <= colval <= (h_bar - 4*common_diff):\n",
        "    return 2\n",
        "  elif (h_bar - 6*common_diff) <= colval <= (h_bar - 5*common_diff):\n",
        "    return 1\n",
        "  else:\n",
        "    return 0\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kjj1qDaIJC6E"
      },
      "outputs": [],
      "source": [
        "# data['N_PCell_RSRP_max_bin'] = data['PCell_RSRP_max'].apply(lambda x: RSRmapper(rsr_col = data['PCell_RSRP_max'], h_bar=-40, l_bar=-140, colval=x))\n",
        "# data['N_PCell_RSRQ_max_bin'] = data['PCell_RSRQ_max'].apply(lambda x: RSRmapper(rsr_col=data['PCell_RSRQ_max'], h_bar=-3, l_bar=-19.5, colval=x))\n",
        "# data['N_PCell_RSSI_max_bin'] = data['PCell_RSSI_max'].apply(lambda x: RSRmapper(rsr_col=data['PCell_RSSI_max'], h_bar=0, l_bar=-120, colval=x))\n",
        "\n",
        "# data['N_SCell_RSRP_max_bin'] = data['SCell_RSRP_max'].apply(lambda x: RSRmapper(rsr_col = data['SCell_RSRP_max'], h_bar=-40, l_bar=-140, colval=x))\n",
        "# data['N_SCell_RSRQ_max_bin'] = data['SCell_RSRQ_max'].apply(lambda x: RSRmapper(rsr_col=data['SCell_RSRQ_max'], h_bar=-3, l_bar=-19.5, colval=x))\n",
        "# data['N_SCell_RSSI_max_bin'] = data['SCell_RSSI_max'].apply(lambda x: RSRmapper(rsr_col=data['SCell_RSSI_max'], h_bar=0, l_bar=-120, colval=x))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "d4qFcf2Qhk-1"
      },
      "outputs": [],
      "source": [
        "data['N_PCell_RSRP_max_bin'] = pd.qcut(data['PCell_RSRP_max'],q=10,labels=False)\n",
        "data['N_PCell_RSRQ_max_bin'] = pd.qcut(data['PCell_RSRQ_max'],q=10,labels=False)\n",
        "data['N_PCell_RSSI_max_bin'] = pd.qcut(data['PCell_RSSI_max'],q=10,labels=False)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "P_hpLqLTh5XY"
      },
      "outputs": [],
      "source": [
        "data['N_SCell_RSRP_max_bin'] = pd.qcut(data['SCell_RSRP_max'],q=10,labels=False)\n",
        "data['N_SCell_RSRQ_max_bin'] = pd.qcut(data['SCell_RSRQ_max'],q=10,labels=False)\n",
        "data['N_SCell_RSSI_max_bin'] = pd.qcut(data['SCell_RSSI_max'],q=10,labels=False)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_OZASs5swaQo"
      },
      "outputs": [],
      "source": [
        "from sklearn.cluster import KMeans\n",
        "def kmeans_clustering(df: pd.DataFrame, n_clusters: int) -> List:\n",
        "  df = df.copy()\n",
        "  df.fillna(0, inplace=True)\n",
        "  kmeans = KMeans(n_init='auto',n_clusters=n_clusters)\n",
        "  kmeans.fit(df)\n",
        "  labels = kmeans.labels_\n",
        "  centroids = kmeans.cluster_centers_\n",
        "  return labels, centroids"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-YQ8I8f5xVAK"
      },
      "outputs": [],
      "source": [
        "# PCell_RS_max_labels, PCell_RS_max_centroids = kmeans_clustering(data[['PCell_RSRP_max','PCell_RSRQ_max','PCell_RSSI_max']], n_clusters=3)\n",
        "# SCell_RS_max_labels, SCell_RS_max_centroids = kmeans_clustering(data[['SCell_RSRP_max','SCell_RSRQ_max','SCell_RSSI_max']], n_clusters=3)\n",
        "# PCell_SI_max_labels, SCell_RS_max_centroids = kmeans_clustering(data[['PCell_RSSI_max','PCell_Downlink_Average_MCS']], n_clusters=3)\n",
        "# PCell_RQ_max_labels, SCell_RS_max_centroids = kmeans_clustering(data[['PCell_RSRQ_max','PCell_Downlink_Average_MCS']], n_clusters=3)\n",
        "# SCell_SI_max_labels, SCell_RS_max_centroids = kmeans_clustering(data[['SCell_RSRP_max','SCell_Downlink_Average_MCS']], n_clusters=3)\n",
        "# SCell_RQ_max_labels, SCell_RS_max_centroids = kmeans_clustering(data[['SCell_RSRP_max','SCell_Downlink_Average_MCS']], n_clusters=3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "E5D-IDWxZN68"
      },
      "outputs": [],
      "source": [
        "from datetime import datetime\n",
        "def Timestampfeatures(timestamp)-> List:\n",
        "  dt = datetime.fromtimestamp(timestamp)\n",
        "  hour = dt.hour\n",
        "  minute = dt.minute\n",
        "  second = dt.second\n",
        "\n",
        "  days = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday']\n",
        "  day_of_week = days[dt.weekday()]\n",
        "\n",
        "  hour = dt.hour\n",
        "  if 6 <= hour < 12:\n",
        "    time_of_day = 'Morning'\n",
        "  elif 12 <= hour < 18:\n",
        "    time_of_day = 'Afternoon'\n",
        "  elif 18 <= hour < 24:\n",
        "    time_of_day = 'Evening'\n",
        "  else:\n",
        "    time_of_day = 'Night'\n",
        "\n",
        "  return hour, minute, second, day_of_week, time_of_day\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "GuxmtPjJezn8"
      },
      "outputs": [],
      "source": [
        "data['hour'] = data['timestamp'].apply(lambda x:Timestampfeatures(x)[0])\n",
        "data['minute'] = data['timestamp'].apply(lambda x:Timestampfeatures(x)[1])\n",
        "data['second'] = data['timestamp'].apply(lambda x:Timestampfeatures(x)[2])\n",
        "data['day_of_week'] = data['timestamp'].apply(lambda x:Timestampfeatures(x)[3])\n",
        "data['time_of_day'] = data['timestamp'].apply(lambda x:Timestampfeatures(x)[4])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WChYpI5UDf6r"
      },
      "outputs": [],
      "source": [
        "data['original_index'] = data.index"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-YzYCc3qjwOQ"
      },
      "outputs": [],
      "source": [
        "days = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday']\n",
        "data['day_of_week'] = pd.Categorical(data['day_of_week'], categories=days, ordered=True)\n",
        "\n",
        "# Sort the DataFrame by the appropriate columns\n",
        "data = data.sort_values(by=['day_of_week', 'hour', 'minute', 'second'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "AZysFFXx1kl1"
      },
      "outputs": [],
      "source": [
        "PCell_cols = ['PCell_RSRP_max', 'PCell_RSRQ_max', 'PCell_RSSI_max', 'PCell_SNR_max',\n",
        "       'PCell_Downlink_Num_RBs', 'PCell_Downlink_Average_MCS',]\n",
        "SCell_cols = ['SCell_RSRP_max', 'SCell_RSRQ_max', 'SCell_RSSI_max', 'SCell_SNR_max',\n",
        "       'SCell_Downlink_Num_RBs', 'SCell_Downlink_Average_MCS']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "XNoM4TcN9NdT"
      },
      "outputs": [],
      "source": [
        "mask = data['operator'] == 1\n",
        "shifted_values = data[PCell_cols].shift().where(mask)\n",
        "PCell_result = data[PCell_cols].where(mask) - shifted_values\n",
        "PCell_result.fillna(0, inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "VHKQTsHQ98hr"
      },
      "outputs": [],
      "source": [
        "mask = data['operator'] == 1\n",
        "shifted_values = data[SCell_cols].shift().where(mask)\n",
        "SCell_result = data[SCell_cols].where(mask) - shifted_values\n",
        "SCell_result.fillna(0, inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "GD9rNgjZ9eZ5"
      },
      "outputs": [],
      "source": [
        "PCell_result = PCell_result.rename(columns={'PCell_RSRP_max':'N_PCell_operator_RSRP_diff','PCell_RSRQ_max':'N_PCell_operator_RSRQ_diff',\n",
        "                             'PCell_RSSI_max':'N_PCell_operator_RSSI_diff','PCell_SNR_max':'N_PCell_operator_SNR_diff',\n",
        "                             'PCell_Downlink_Num_RBs':'N_PCell_operator_Downlink_Num_RBs_diff',\n",
        "                             'PCell_Downlink_Average_MCS':'N_PCell_operator_Downlink_Average_MCS_diff'})\n",
        "\n",
        "SCell_result = SCell_result.rename(columns={'SCell_RSRP_max':'N_SCell_operator_RSRP_diff','SCell_RSRQ_max':'N_SCell_operator_RSRQ_diff',\n",
        "                             'SCell_RSSI_max':'N_SCell_operator_RSSI_diff','SCell_SNR_max':'N_SCell_operator_SNR_diff',\n",
        "                             'SCell_Downlink_Num_RBs':'N_SCell_operator_Downlink_Num_RBs_diff',\n",
        "                             'SCell_Downlink_Average_MCS':'N_SCell_operator_Downlink_Average_MCS_diff'})\n",
        "\n",
        "df = pd.concat([SCell_result,PCell_result],axis=1)\n",
        "data = pd.concat([data,df],axis=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "L26zMu3nDnlq"
      },
      "outputs": [],
      "source": [
        "data = data.sort_values(by='original_index')\n",
        "\n",
        "# Drop the original_index column\n",
        "data = data.drop(columns='original_index')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "q8L_51BOtdFz"
      },
      "outputs": [],
      "source": [
        "cols = ['temperature','dewPoint','Altitude']\n",
        "gcols = ['operator']\n",
        "\n",
        "for gcol in gcols:\n",
        "  for col in cols:\n",
        "    data[f'N_{gcol}_{col}_feature_sum'] = data.groupby([gcol])[col].transform('sum')\n",
        "    data[f'N_{gcol}_{col}_feature_max'] = data.groupby([gcol])[col].transform('max')\n",
        "    data[f'N_{gcol}_{col}_feature_min'] = data.groupby([gcol])[col].transform('min')\n",
        "    data[f'N_{gcol}_{col}_feature_mean'] = data.groupby([gcol])[col].transform('mean')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "sdqsQnQu-FBQ"
      },
      "outputs": [],
      "source": [
        "# data['N_PCell_Feature_1'] = data['PCell_Downlink_Average_MCS']/data['PCell_Downlink_Num_RBs']\n",
        "# data['N_SCell_Feature_1'] = data['SCell_Downlink_Average_MCS']/data['SCell_Downlink_Num_RBs']"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "qzg3KpfJNwhb"
      },
      "outputs": [],
      "source": [
        "train = data[data['target'] != 'Test'].copy()\n",
        "test = data[data['target'] == 'Test'].copy()\n",
        "test = test.drop(['target'],axis=1)\n",
        "train['target'] = np.float64(train['target'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "E_kSfbIeQyg5"
      },
      "outputs": [],
      "source": [
        "TT = train.drop(['id','device'],axis=1)\n",
        "tests = test.drop(['id','device'], axis=1)\n",
        "\n",
        "cat_features = ['area','day_of_week','time_of_day']\n",
        "oe = OrdinalEncoder()\n",
        "for col in cat_features:\n",
        "  tests[col] = oe.fit_transform(tests[[col]])\n",
        "  TT[col] = oe.fit_transform(TT[[col]])\n",
        "\n",
        "\n",
        "TT.fillna(0, inplace=True)\n",
        "tests.fillna(0, inplace=True)\n",
        "\n",
        "TT = TT[[col for col in TT.columns if col != 'target'] + [\"target\"]]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ICLcGzVLXIVX"
      },
      "outputs": [],
      "source": [
        "TT= TT.drop(['hour'],axis=1)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "eq-59JCjimDI",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "676d8919-2ace-4e7b-c339-389bf75d3da6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CV_RMSE_Score:9204135.828065008\n",
            "CV_RMSE_Score:9171013.012532929\n"
          ]
        }
      ],
      "source": [
        "Baseline = Baseline_score(train=TT,model=cbc,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "lgb_score,bsv_1 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "KFold_preds = pd.DataFrame(bsv_1)\n",
        "KFold_preds = KFold_preds.join(TT['target'])\n",
        "\n",
        "Baseline = Baseline_score(train=TT,model=cbc1,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "lgb_score,bsv_2 = Baseline.run()\n",
        "print(f\"CV_RMSE_Score:{lgb_score}\")\n",
        "\n",
        "KFold_preds_1 = pd.DataFrame(bsv_2)\n",
        "KFold_preds_1 = KFold_preds_1.join(TT['target'])\n",
        "\n",
        "predictions = Baseline_predict(train=TT,model=cbc,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "predict_1,bsv_test1 = predictions.predict(tests)\n",
        "\n",
        "predictions = Baseline_predict(train=TT,model=cbc1,split_method='KFold',return_baseline=True,num_split=15,VERBOSE=0,random_seed=0)\n",
        "predict_2,bsv_test2 = predictions.predict(tests)\n",
        "\n",
        "testpreds_1 = pd.DataFrame(bsv_test1)\n",
        "testpreds_2 = pd.DataFrame(bsv_test2)\n",
        "\n",
        "predictions = Baseline_predict(train=KFold_preds,model=cbc,split_method='KFold',return_baseline=False,VERBOSE=0,random_seed=0)\n",
        "predict_3 = predictions.predict(testpreds_1)\n",
        "\n",
        "predictions = Baseline_predict(train=KFold_preds_1,model=cbc1,split_method='KFold',return_baseline=False,VERBOSE=0,random_seed=0)\n",
        "predict_4 = predictions.predict(testpreds_2)\n",
        "\n",
        "ensemble_pred = sum(predict_1)/15 + sum(predict_2)/15 + sum(predict_3)/5 + sum(predict_4)/5\n",
        "final_preds = ensemble_pred/4"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "final_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MQGWnMFYY1ko",
        "outputId": "4f1da55b-f6d4-4129-eecb-97cf3966bd96"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([22667347.81289839, 56464256.74563297, 78430796.4174026 , ...,\n",
              "       20596929.17270317, 13817500.89916121, 33133917.77315626])"
            ]
          },
          "metadata": {},
          "execution_count": 484
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "predictions_df = pd.DataFrame({'id': Test.id, 'target': final_preds})\n",
        "predictions_df.to_csv('34th_sub.csv',index=False)"
      ],
      "metadata": {
        "id": "YyzwuIH1Y9bB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 521
        },
        "id": "aa8l62A-Gavc",
        "outputId": "1b86a5db-f7c7-4d72-e913-ae5dad75a7a6"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 2500x1000 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "Baseline.plot_features(n_features=40)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "OZvpuibLgvy6"
      },
      "outputs": [],
      "source": []
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [
        "seUcMl_BqUyz",
        "kvqeG7348Zy9"
      ],
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}