2019uch0031+2019ucs0092_ML_lab.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "2019uch0031+2019ucs0092_ML_lab.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/yash-clear/Anomaly_Detection/blob/main/2019uch0031%2B2019ucs0092_ML_lab.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "\n",
        "\n",
        "```\n",
        "`# This is formatted as code`\n",
        "```\n",
        "\n",
        "# ASSIGNMENT -1\n",
        "\n",
        "The goal of this assignment is\n",
        "to take new cases as target variable and predict them from mobility data.\n"
      ],
      "metadata": {
        "id": "YvnevlwAN8rL"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## TASK PERFORMED BY : YASHASWITA SINGH (2019UCH0031)\n",
        "##                     ANKUSH GARG (2019UCS0092)"
      ],
      "metadata": {
        "id": "a4VhpmZWYkuA"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Language: Python"
      ],
      "metadata": {
        "id": "BZIRcZbeZwj6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#Importing essential libraries \n",
        "\n",
        "import pandas as pd\n",
        "import plotly.express as px\n",
        "import seaborn as sns\n",
        "from copy import copy\n",
        "from scipy import stats\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n"
      ],
      "metadata": {
        "id": "NMX6NldQN8vB"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "IMPORTING DATASET"
      ],
      "metadata": {
        "id": "Ri2EeNvcRW16"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "df1 = pd.read_csv('/content/changes-visitors-covid (1).csv')\n",
        "df2 = pd.read_csv('/content/owid-covid-data (1).csv')"
      ],
      "metadata": {
        "id": "-9C8rlFzN8y_"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Take “India” as the\n",
        "case study for this assignment."
      ],
      "metadata": {
        "id": "wsFvRb4ZRcNW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "india_df1 = df1[df1['Entity'] == \"India\"]\n",
        "india_df2 = df2[df2['iso_code']=='IND']"
      ],
      "metadata": {
        "id": "U0wtKmUya2st"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "india_df1.reset_index(inplace = True)\n",
        "india_df2.reset_index(inplace = True)"
      ],
      "metadata": {
        "id": "id3gmfUIaHn9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "india_df1"
      ],
      "metadata": {
        "id": "x1nCmdmq0ofx",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "outputId": "d195065f-04b8-4698-869f-97f34e5d4bd9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     index Entity Code         Day  retail_and_recreation  \\\n",
              "0    23897  India  IND  2020-02-17                  0.667   \n",
              "1    23898  India  IND  2020-02-18                  0.500   \n",
              "2    23899  India  IND  2020-02-19                  0.400   \n",
              "3    23900  India  IND  2020-02-20                  0.500   \n",
              "4    23901  India  IND  2020-02-21                 -0.143   \n",
              "..     ...    ...  ...         ...                    ...   \n",
              "466  24363  India  IND  2021-05-28                -61.714   \n",
              "467  24364  India  IND  2021-05-29                -61.286   \n",
              "468  24365  India  IND  2021-05-30                -61.143   \n",
              "469  24366  India  IND  2021-05-31                -60.143   \n",
              "470  24367  India  IND  2021-06-01                -58.714   \n",
              "\n",
              "     grocery_and_pharmacy  residential  transit_stations   parks  workplaces  \n",
              "0                   1.667        0.000             2.000   3.000       3.000  \n",
              "1                   1.750        0.000             2.000   3.250       3.000  \n",
              "2                   1.800        0.200             1.800   2.800       3.200  \n",
              "3                   2.000        0.000             2.333   3.167       3.333  \n",
              "4                   1.714        0.714             1.429   3.571       0.143  \n",
              "..                    ...          ...               ...     ...         ...  \n",
              "466               -25.000       24.143           -49.143 -41.000     -45.429  \n",
              "467               -24.429       23.714           -48.714 -40.000     -44.571  \n",
              "468               -24.714       23.714           -49.000 -39.143     -44.286  \n",
              "469               -23.429       23.286           -48.286 -38.000     -43.429  \n",
              "470               -21.143       22.714           -47.000 -36.571     -42.429  \n",
              "\n",
              "[471 rows x 10 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5fa98457-efd6-4067-95a2-8240ae31a6dd\">\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>index</th>\n",
              "      <th>Entity</th>\n",
              "      <th>Code</th>\n",
              "      <th>Day</th>\n",
              "      <th>retail_and_recreation</th>\n",
              "      <th>grocery_and_pharmacy</th>\n",
              "      <th>residential</th>\n",
              "      <th>transit_stations</th>\n",
              "      <th>parks</th>\n",
              "      <th>workplaces</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>23897</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2020-02-17</td>\n",
              "      <td>0.667</td>\n",
              "      <td>1.667</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>3.000</td>\n",
              "      <td>3.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>23898</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2020-02-18</td>\n",
              "      <td>0.500</td>\n",
              "      <td>1.750</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>3.250</td>\n",
              "      <td>3.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>23899</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2020-02-19</td>\n",
              "      <td>0.400</td>\n",
              "      <td>1.800</td>\n",
              "      <td>0.200</td>\n",
              "      <td>1.800</td>\n",
              "      <td>2.800</td>\n",
              "      <td>3.200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>23900</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2020-02-20</td>\n",
              "      <td>0.500</td>\n",
              "      <td>2.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.333</td>\n",
              "      <td>3.167</td>\n",
              "      <td>3.333</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>23901</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2020-02-21</td>\n",
              "      <td>-0.143</td>\n",
              "      <td>1.714</td>\n",
              "      <td>0.714</td>\n",
              "      <td>1.429</td>\n",
              "      <td>3.571</td>\n",
              "      <td>0.143</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>466</th>\n",
              "      <td>24363</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2021-05-28</td>\n",
              "      <td>-61.714</td>\n",
              "      <td>-25.000</td>\n",
              "      <td>24.143</td>\n",
              "      <td>-49.143</td>\n",
              "      <td>-41.000</td>\n",
              "      <td>-45.429</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>467</th>\n",
              "      <td>24364</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2021-05-29</td>\n",
              "      <td>-61.286</td>\n",
              "      <td>-24.429</td>\n",
              "      <td>23.714</td>\n",
              "      <td>-48.714</td>\n",
              "      <td>-40.000</td>\n",
              "      <td>-44.571</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>468</th>\n",
              "      <td>24365</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2021-05-30</td>\n",
              "      <td>-61.143</td>\n",
              "      <td>-24.714</td>\n",
              "      <td>23.714</td>\n",
              "      <td>-49.000</td>\n",
              "      <td>-39.143</td>\n",
              "      <td>-44.286</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>469</th>\n",
              "      <td>24366</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2021-05-31</td>\n",
              "      <td>-60.143</td>\n",
              "      <td>-23.429</td>\n",
              "      <td>23.286</td>\n",
              "      <td>-48.286</td>\n",
              "      <td>-38.000</td>\n",
              "      <td>-43.429</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>470</th>\n",
              "      <td>24367</td>\n",
              "      <td>India</td>\n",
              "      <td>IND</td>\n",
              "      <td>2021-06-01</td>\n",
              "      <td>-58.714</td>\n",
              "      <td>-21.143</td>\n",
              "      <td>22.714</td>\n",
              "      <td>-47.000</td>\n",
              "      <td>-36.571</td>\n",
              "      <td>-42.429</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>471 rows × 10 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5fa98457-efd6-4067-95a2-8240ae31a6dd')\"\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-5fa98457-efd6-4067-95a2-8240ae31a6dd 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-5fa98457-efd6-4067-95a2-8240ae31a6dd');\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": 521
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "india_df1.drop(['index'], axis=1, inplace = True)\n",
        "india_df2.drop(['index'], axis=1, inplace = True)\n",
        "\n",
        "india_df2.drop(india_df2.columns[6:], axis=1, inplace = True)"
      ],
      "metadata": {
        "id": "VFJKX7PkaHsn",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8ad31dcd-204d-42c2-fa9b-26086f3210db"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/pandas/core/frame.py:4913: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  errors=errors,\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "RESET THE INDEXES"
      ],
      "metadata": {
        "id": "rTIq7Si5Ro4J"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "india_df2"
      ],
      "metadata": {
        "id": "Ag_vegYh1fe1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "outputId": "c718171d-4e35-4db4-ec42-147b65aaf439"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    iso_code continent location        date  total_cases  new_cases\n",
              "0        IND      Asia    India  2020-01-30          1.0        1.0\n",
              "1        IND      Asia    India  2020-01-31          1.0        0.0\n",
              "2        IND      Asia    India  2020-02-01          1.0        0.0\n",
              "3        IND      Asia    India  2020-02-02          2.0        1.0\n",
              "4        IND      Asia    India  2020-02-03          3.0        1.0\n",
              "..       ...       ...      ...         ...          ...        ...\n",
              "486      IND      Asia    India  2021-05-30   28047534.0   152734.0\n",
              "487      IND      Asia    India  2021-05-31   28175044.0   127510.0\n",
              "488      IND      Asia    India  2021-06-01   28307832.0   132788.0\n",
              "489      IND      Asia    India  2021-06-02   28441986.0   134154.0\n",
              "490      IND      Asia    India  2021-06-03   28574350.0   132364.0\n",
              "\n",
              "[491 rows x 6 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-998974fb-f5fe-46d1-af18-d4b6928d2585\">\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>iso_code</th>\n",
              "      <th>continent</th>\n",
              "      <th>location</th>\n",
              "      <th>date</th>\n",
              "      <th>total_cases</th>\n",
              "      <th>new_cases</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-01-30</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-01-31</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-02-01</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-02-02</td>\n",
              "      <td>2.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-02-03</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>486</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-05-30</td>\n",
              "      <td>28047534.0</td>\n",
              "      <td>152734.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>487</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-05-31</td>\n",
              "      <td>28175044.0</td>\n",
              "      <td>127510.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>488</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-06-01</td>\n",
              "      <td>28307832.0</td>\n",
              "      <td>132788.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>489</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-06-02</td>\n",
              "      <td>28441986.0</td>\n",
              "      <td>134154.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>490</th>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-06-03</td>\n",
              "      <td>28574350.0</td>\n",
              "      <td>132364.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>491 rows × 6 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-998974fb-f5fe-46d1-af18-d4b6928d2585')\"\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-998974fb-f5fe-46d1-af18-d4b6928d2585 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-998974fb-f5fe-46d1-af18-d4b6928d2585');\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": 523
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "india_df2 = india_df2[18: -2]\n",
        "india_df2.reset_index(inplace = True)"
      ],
      "metadata": {
        "id": "S9pHytneaHw2"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "india_df2"
      ],
      "metadata": {
        "id": "-DLeOnZwaH1H",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "outputId": "58ee8142-6f26-47bd-a2a3-5f431942639b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     index iso_code continent location        date  total_cases  new_cases\n",
              "0       18      IND      Asia    India  2020-02-17          3.0        0.0\n",
              "1       19      IND      Asia    India  2020-02-18          3.0        0.0\n",
              "2       20      IND      Asia    India  2020-02-19          3.0        0.0\n",
              "3       21      IND      Asia    India  2020-02-20          3.0        0.0\n",
              "4       22      IND      Asia    India  2020-02-21          3.0        0.0\n",
              "..     ...      ...       ...      ...         ...          ...        ...\n",
              "466    484      IND      Asia    India  2021-05-28   27729247.0   173790.0\n",
              "467    485      IND      Asia    India  2021-05-29   27894800.0   165553.0\n",
              "468    486      IND      Asia    India  2021-05-30   28047534.0   152734.0\n",
              "469    487      IND      Asia    India  2021-05-31   28175044.0   127510.0\n",
              "470    488      IND      Asia    India  2021-06-01   28307832.0   132788.0\n",
              "\n",
              "[471 rows x 7 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-427c422f-f837-42f5-aaa3-e1f386d32de8\">\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>index</th>\n",
              "      <th>iso_code</th>\n",
              "      <th>continent</th>\n",
              "      <th>location</th>\n",
              "      <th>date</th>\n",
              "      <th>total_cases</th>\n",
              "      <th>new_cases</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>18</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-02-17</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>19</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-02-18</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>20</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-02-19</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>21</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-02-20</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>22</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2020-02-21</td>\n",
              "      <td>3.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>466</th>\n",
              "      <td>484</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-05-28</td>\n",
              "      <td>27729247.0</td>\n",
              "      <td>173790.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>467</th>\n",
              "      <td>485</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-05-29</td>\n",
              "      <td>27894800.0</td>\n",
              "      <td>165553.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>468</th>\n",
              "      <td>486</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-05-30</td>\n",
              "      <td>28047534.0</td>\n",
              "      <td>152734.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>469</th>\n",
              "      <td>487</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-05-31</td>\n",
              "      <td>28175044.0</td>\n",
              "      <td>127510.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>470</th>\n",
              "      <td>488</td>\n",
              "      <td>IND</td>\n",
              "      <td>Asia</td>\n",
              "      <td>India</td>\n",
              "      <td>2021-06-01</td>\n",
              "      <td>28307832.0</td>\n",
              "      <td>132788.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>471 rows × 7 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-427c422f-f837-42f5-aaa3-e1f386d32de8')\"\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-427c422f-f837-42f5-aaa3-e1f386d32de8 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-427c422f-f837-42f5-aaa3-e1f386d32de8');\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": 525
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Creating a new data frame "
      ],
      "metadata": {
        "id": "bVqflTp6aH5U"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "final_df = pd.concat([india_df1, india_df2['new_cases']], axis=1, join='inner')"
      ],
      "metadata": {
        "id": "YipbVVoq2EYo"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "final_df = final_df.iloc[:, 3:]"
      ],
      "metadata": {
        "id": "8bjwds1kRe7l"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "final_df"
      ],
      "metadata": {
        "id": "AOHCMsoK2EdN",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "outputId": "3cd77583-13d2-435f-d456-a096dfcd444a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     retail_and_recreation  grocery_and_pharmacy  residential  \\\n",
              "0                    0.667                 1.667        0.000   \n",
              "1                    0.500                 1.750        0.000   \n",
              "2                    0.400                 1.800        0.200   \n",
              "3                    0.500                 2.000        0.000   \n",
              "4                   -0.143                 1.714        0.714   \n",
              "..                     ...                   ...          ...   \n",
              "466                -61.714               -25.000       24.143   \n",
              "467                -61.286               -24.429       23.714   \n",
              "468                -61.143               -24.714       23.714   \n",
              "469                -60.143               -23.429       23.286   \n",
              "470                -58.714               -21.143       22.714   \n",
              "\n",
              "     transit_stations   parks  workplaces  new_cases  \n",
              "0               2.000   3.000       3.000        0.0  \n",
              "1               2.000   3.250       3.000        0.0  \n",
              "2               1.800   2.800       3.200        0.0  \n",
              "3               2.333   3.167       3.333        0.0  \n",
              "4               1.429   3.571       0.143        0.0  \n",
              "..                ...     ...         ...        ...  \n",
              "466           -49.143 -41.000     -45.429   173790.0  \n",
              "467           -48.714 -40.000     -44.571   165553.0  \n",
              "468           -49.000 -39.143     -44.286   152734.0  \n",
              "469           -48.286 -38.000     -43.429   127510.0  \n",
              "470           -47.000 -36.571     -42.429   132788.0  \n",
              "\n",
              "[471 rows x 7 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-ceea89f6-fe06-4091-8220-45c9fc033907\">\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>retail_and_recreation</th>\n",
              "      <th>grocery_and_pharmacy</th>\n",
              "      <th>residential</th>\n",
              "      <th>transit_stations</th>\n",
              "      <th>parks</th>\n",
              "      <th>workplaces</th>\n",
              "      <th>new_cases</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.667</td>\n",
              "      <td>1.667</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>3.000</td>\n",
              "      <td>3.000</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.500</td>\n",
              "      <td>1.750</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>3.250</td>\n",
              "      <td>3.000</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.400</td>\n",
              "      <td>1.800</td>\n",
              "      <td>0.200</td>\n",
              "      <td>1.800</td>\n",
              "      <td>2.800</td>\n",
              "      <td>3.200</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.500</td>\n",
              "      <td>2.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.333</td>\n",
              "      <td>3.167</td>\n",
              "      <td>3.333</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-0.143</td>\n",
              "      <td>1.714</td>\n",
              "      <td>0.714</td>\n",
              "      <td>1.429</td>\n",
              "      <td>3.571</td>\n",
              "      <td>0.143</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>466</th>\n",
              "      <td>-61.714</td>\n",
              "      <td>-25.000</td>\n",
              "      <td>24.143</td>\n",
              "      <td>-49.143</td>\n",
              "      <td>-41.000</td>\n",
              "      <td>-45.429</td>\n",
              "      <td>173790.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>467</th>\n",
              "      <td>-61.286</td>\n",
              "      <td>-24.429</td>\n",
              "      <td>23.714</td>\n",
              "      <td>-48.714</td>\n",
              "      <td>-40.000</td>\n",
              "      <td>-44.571</td>\n",
              "      <td>165553.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>468</th>\n",
              "      <td>-61.143</td>\n",
              "      <td>-24.714</td>\n",
              "      <td>23.714</td>\n",
              "      <td>-49.000</td>\n",
              "      <td>-39.143</td>\n",
              "      <td>-44.286</td>\n",
              "      <td>152734.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>469</th>\n",
              "      <td>-60.143</td>\n",
              "      <td>-23.429</td>\n",
              "      <td>23.286</td>\n",
              "      <td>-48.286</td>\n",
              "      <td>-38.000</td>\n",
              "      <td>-43.429</td>\n",
              "      <td>127510.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>470</th>\n",
              "      <td>-58.714</td>\n",
              "      <td>-21.143</td>\n",
              "      <td>22.714</td>\n",
              "      <td>-47.000</td>\n",
              "      <td>-36.571</td>\n",
              "      <td>-42.429</td>\n",
              "      <td>132788.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>471 rows × 7 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ceea89f6-fe06-4091-8220-45c9fc033907')\"\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-ceea89f6-fe06-4091-8220-45c9fc033907 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-ceea89f6-fe06-4091-8220-45c9fc033907');\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": 528
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "y = final_df['new_cases']\n",
        "y\n",
        "\n",
        "\n",
        "X = final_df.iloc[:,:-1]\n",
        "X"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "6HQfwt-_smB4",
        "outputId": "be2532fa-2d7c-495d-8557-266192b576d0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     retail_and_recreation  grocery_and_pharmacy  residential  \\\n",
              "0                    0.667                 1.667        0.000   \n",
              "1                    0.500                 1.750        0.000   \n",
              "2                    0.400                 1.800        0.200   \n",
              "3                    0.500                 2.000        0.000   \n",
              "4                   -0.143                 1.714        0.714   \n",
              "..                     ...                   ...          ...   \n",
              "466                -61.714               -25.000       24.143   \n",
              "467                -61.286               -24.429       23.714   \n",
              "468                -61.143               -24.714       23.714   \n",
              "469                -60.143               -23.429       23.286   \n",
              "470                -58.714               -21.143       22.714   \n",
              "\n",
              "     transit_stations   parks  workplaces  \n",
              "0               2.000   3.000       3.000  \n",
              "1               2.000   3.250       3.000  \n",
              "2               1.800   2.800       3.200  \n",
              "3               2.333   3.167       3.333  \n",
              "4               1.429   3.571       0.143  \n",
              "..                ...     ...         ...  \n",
              "466           -49.143 -41.000     -45.429  \n",
              "467           -48.714 -40.000     -44.571  \n",
              "468           -49.000 -39.143     -44.286  \n",
              "469           -48.286 -38.000     -43.429  \n",
              "470           -47.000 -36.571     -42.429  \n",
              "\n",
              "[471 rows x 6 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5c1f42fd-557f-4223-b700-9512e877d9e0\">\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>retail_and_recreation</th>\n",
              "      <th>grocery_and_pharmacy</th>\n",
              "      <th>residential</th>\n",
              "      <th>transit_stations</th>\n",
              "      <th>parks</th>\n",
              "      <th>workplaces</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.667</td>\n",
              "      <td>1.667</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>3.000</td>\n",
              "      <td>3.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.500</td>\n",
              "      <td>1.750</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>3.250</td>\n",
              "      <td>3.000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.400</td>\n",
              "      <td>1.800</td>\n",
              "      <td>0.200</td>\n",
              "      <td>1.800</td>\n",
              "      <td>2.800</td>\n",
              "      <td>3.200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.500</td>\n",
              "      <td>2.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.333</td>\n",
              "      <td>3.167</td>\n",
              "      <td>3.333</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-0.143</td>\n",
              "      <td>1.714</td>\n",
              "      <td>0.714</td>\n",
              "      <td>1.429</td>\n",
              "      <td>3.571</td>\n",
              "      <td>0.143</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>466</th>\n",
              "      <td>-61.714</td>\n",
              "      <td>-25.000</td>\n",
              "      <td>24.143</td>\n",
              "      <td>-49.143</td>\n",
              "      <td>-41.000</td>\n",
              "      <td>-45.429</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>467</th>\n",
              "      <td>-61.286</td>\n",
              "      <td>-24.429</td>\n",
              "      <td>23.714</td>\n",
              "      <td>-48.714</td>\n",
              "      <td>-40.000</td>\n",
              "      <td>-44.571</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>468</th>\n",
              "      <td>-61.143</td>\n",
              "      <td>-24.714</td>\n",
              "      <td>23.714</td>\n",
              "      <td>-49.000</td>\n",
              "      <td>-39.143</td>\n",
              "      <td>-44.286</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>469</th>\n",
              "      <td>-60.143</td>\n",
              "      <td>-23.429</td>\n",
              "      <td>23.286</td>\n",
              "      <td>-48.286</td>\n",
              "      <td>-38.000</td>\n",
              "      <td>-43.429</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>470</th>\n",
              "      <td>-58.714</td>\n",
              "      <td>-21.143</td>\n",
              "      <td>22.714</td>\n",
              "      <td>-47.000</td>\n",
              "      <td>-36.571</td>\n",
              "      <td>-42.429</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>471 rows × 6 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5c1f42fd-557f-4223-b700-9512e877d9e0')\"\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-5c1f42fd-557f-4223-b700-9512e877d9e0 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-5c1f42fd-557f-4223-b700-9512e877d9e0');\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": 529
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from random import seed\n",
        "def train_test_split(X,y,test_size=0.2):\n",
        "\n",
        "    seed(5)\n",
        "    ind = np.random.permutation(X.index)\n",
        "    X=X.reindex(ind)\n",
        "    y=y.reindex(ind)\n",
        "\n",
        "    l=X.shape[0]\n",
        "    train_X,train_y,test_X,test_Y=[],[],[],[]\n",
        "\n",
        "    criterion=int((1-test_size)*l)\n",
        "    train_X=X.iloc[:criterion]\n",
        "    train_y=y.iloc[:criterion]\n",
        "    test_X=X.iloc[criterion:]\n",
        "    test_y=y.iloc[criterion:]\n",
        "    \n",
        "    return train_X,test_X,train_y,test_y"
      ],
      "metadata": {
        "id": "64t01nkkOmf-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)"
      ],
      "metadata": {
        "id": "_m9jhWCvrzJ9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "R2 SCORE "
      ],
      "metadata": {
        "id": "AzhRiIgHUGnd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def r2_Score(Y_true, Y_pred):\n",
        "    Y_true = Y_true.values.tolist()\n",
        "    Y_pred = Y_pred.values.tolist()\n",
        "    Y_mean=np.mean(Y_true)\n",
        "\n",
        "    RSS=0\n",
        "    TSS=0\n",
        "\n",
        "    for i in range(len(Y_true)):\n",
        "        RSS=RSS+np.square(Y_true[i]-Y_pred[i])\n",
        "        TSS=TSS+np.square(Y_true[i]-Y_mean)\n",
        "        \n",
        "    return(1-(RSS/TSS))"
      ],
      "metadata": {
        "id": "KqdybFZHUIWZ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### ID3A"
      ],
      "metadata": {
        "id": "htQa0-MAx2ir"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "class DecisionTreeRegressor:\n",
        "  \n",
        "  def fit(self, X, y, min_leaf = 5):\n",
        "    self.dtree = Node(X, y, np.array(np.arange(len(y))), min_leaf)\n",
        "    return self\n",
        "  \n",
        "  def predict(self, X):\n",
        "    return self.dtree.predict(X.values)"
      ],
      "metadata": {
        "id": "v6RrA5Bh2Eh3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class Node:\n",
        "\n",
        "    def __init__(self, x, y, idxs, min_leaf=5):\n",
        "        self.x = x \n",
        "        self.y = y\n",
        "        self.idxs = idxs \n",
        "        self.min_leaf = min_leaf\n",
        "        self.row_count = len(idxs)\n",
        "        self.col_count = x.shape[1]\n",
        "        self.val = np.mean(y[idxs])\n",
        "        self.score = float('inf')\n",
        "        self.find_varsplit()\n",
        "        \n",
        "    def find_varsplit(self):\n",
        "        for c in range(self.col_count): self.find_better_split(c)\n",
        "        if self.is_leaf: return\n",
        "        x = self.split_col\n",
        "        lhs = np.nonzero(x <= self.split)[0]\n",
        "        rhs = np.nonzero(x > self.split)[0]\n",
        "        self.lhs = Node(self.x, self.y, self.idxs[lhs], self.min_leaf)\n",
        "        self.rhs = Node(self.x, self.y, self.idxs[rhs], self.min_leaf)\n",
        "        \n",
        "    def find_better_split(self, var_idx):\n",
        "      \n",
        "        x = self.x.values[self.idxs, var_idx]\n",
        "\n",
        "        for r in range(self.row_count):\n",
        "            lhs = x <= x[r]\n",
        "            rhs = x > x[r]\n",
        "            if rhs.sum() < self.min_leaf or lhs.sum() < self.min_leaf: continue\n",
        "\n",
        "            curr_score = self.find_score(lhs, rhs)\n",
        "            if curr_score < self.score: \n",
        "                self.var_idx = var_idx\n",
        "                self.score = curr_score\n",
        "                self.split = x[r]\n",
        "                \n",
        "    def find_score(self, lhs, rhs):\n",
        "        y = self.y[self.idxs]\n",
        "        lhs_std = y[lhs].std()\n",
        "        rhs_std = y[rhs].std()\n",
        "        return lhs_std * lhs.sum() + rhs_std * rhs.sum()\n",
        "                \n",
        "    @property\n",
        "    def split_col(self): return self.x.values[self.idxs,self.var_idx]\n",
        "                \n",
        "    @property\n",
        "    def is_leaf(self): return self.score == float('inf')                \n",
        "\n",
        "    def predict(self, x):\n",
        "        return np.array([self.predict_row(xi) for xi in x])\n",
        "\n",
        "    def predict_row(self, xi):\n",
        "        if self.is_leaf: return self.val\n",
        "        node = self.lhs if xi[self.var_idx] <= self.split else self.rhs\n",
        "        return node.predict_row(xi)"
      ],
      "metadata": {
        "id": "orHLD_xXq5Je"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "regressor = DecisionTreeRegressor().fit(X, y)\n",
        "preds = regressor.predict(X)\n",
        "# preds"
      ],
      "metadata": {
        "id": "HHgW6cwqrqVu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# X_test\n",
        "# y_test"
      ],
      "metadata": {
        "id": "r1QdjhMNPJjn"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "pred_test = regressor.predict(X_test)"
      ],
      "metadata": {
        "id": "ouxo-m0CvYmz"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# pred_test"
      ],
      "metadata": {
        "id": "dKllHkWivufO"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "plt.figure(figsize=(8, 8))\n",
        "\n",
        "\n",
        "ax = sns.distplot(y_test, hist=False, color=\"r\", label=\"Actual Value\")\n",
        "sns.distplot(pred_test, hist=False, color=\"b\", label=\"Fitted Values\" , ax=ax)\n",
        "\n",
        "\n",
        "plt.title('Actual vs Fitted Values for Price')\n",
        "\n",
        "\n",
        "plt.show()\n",
        "plt.close()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 602
        },
        "id": "_qaHr0sB7Qrk",
        "outputId": "1638e3ad-ebfe-407c-b7eb-2dfd0ba1eb55"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import mean_squared_error,r2_score\n",
        "\n",
        "rms = mean_squared_error(y_test, pred_test, squared=False)\n",
        "rms"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2Nl8vUKsvrVh",
        "outputId": "163e803f-dc74-4eef-ea4a-ecfd0ef40713"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "15215.940898382962"
            ]
          },
          "metadata": {},
          "execution_count": 540
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "r2 = r2_score(y_test, pred_test)\n",
        "r2"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ntz0UxbFweMH",
        "outputId": "31bc41e0-359e-403d-ac6e-6c0d28ff07e0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9787580583042875"
            ]
          },
          "metadata": {},
          "execution_count": 541
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "COUNT vs TARGET VARIABLE "
      ],
      "metadata": {
        "id": "8Q-8VQ6eScTf"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "plt.scatter(final_df.index,final_df['new_cases'])\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "0TTy3eDW0xh3",
        "outputId": "8ba88874-e0e6-40ef-cc46-0b4eb936afa8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAD4CAYAAAAZ1BptAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3df5DcdZ3n8ec7k07oYMkETFEwgQ3e5qCgUCJTECv7x8KWJKLILFICy53UHiV1tXonYmVN9igDt3rESpWgt553KJ54UpIg1BBQd44lVO0da5DJzkAMmCWIQFqUaDJ4mMFMJu/7oz89dHr629M/vr96+vWomkr3p789309/k3zf/fnx/nzM3REREalnQdYVEBGR/FKQEBGRSAoSIiISSUFCREQiKUiIiEikhVlXIG7vete7fMWKFVlXQ0Skq+zates37r6stnzeBYkVK1YwOjqadTVERLqKmb1cr1zdTSIiEklBQkREIilIiIhIJAUJERGJpCAhIiKR5t3sJhGRbjI8VmLLyF5+OTHJ6f1F1q89m6FVA1lXa4aChIhIRm4d3s19O1+hshZ3aWKSjQ/tBshNoFB3k4hIBobHSscFiIrJqWm2jOzNpE71KEiIiGRgy8jeWQGi4pcTk6nWpREFCRGRDDQKBKf3F1OsSWMKEiIiGWgUCC45Z9YSSplRkBARSdHwWIk1m3dQmpjEIo55cFeJ4bFSqvWK0nSQMLM+Mxszs0fD87PM7Ckz22dmW81sUShfHJ7vC6+vqPodG0P5XjNbW1W+LpTtM7MNVeV1zyEi0o2Gx0psfGg3pdDVFDUmkafB61ZaEp8Gnq96/iXgTnf/Y+AQcGMovxE4FMrvDMdhZucC1wLnAeuA/xYCTx/wNeCDwLnAdeHYRucQEek6W0b2Mjk13dSxeRm8bipImNly4EPAN8NzAy4Fvh8OuRcYCo+vDM8Jr/9ZOP5K4H53/4O7vwTsAy4KP/vc/efufgS4H7hyjnOIiHSdVm78eRm8brYlcRfw18Cx8PwUYMLdj4bn+4FK5scA8CpAeP2NcPxMec17osobneM4ZnaTmY2a2eiBAwea/EgiIulq5ca/fu3ZCdakeXMGCTP7MPC6u+9KoT5tcfe73X3Q3QeXLcvPrAARkWrr155NYUHUcPXb+ouF3GRcN7MsxxrgI2Z2OXAC8E7gK0C/mS0M3/SXA5Wh+BJwBrDfzBYCJwG/rSqvqH5PvfLfNjiHiEh3miNGFAt93PaR89KpSxPmbEm4+0Z3X+7uKygPPO9w9+uBJ4Crw2E3AA+Hx9vDc8LrO9zdQ/m1YfbTWcBK4CfA08DKMJNpUTjH9vCeqHOIiHSd2x/Zw9T07DlNfWYYMNBf5I6rzs9NKwI6W+Dvc8D9ZvYFYAy4J5TfA/wvM9sHHKR808fd95jZNuA54CjwSXefBjCzTwEjQB/wLXffM8c5RES6yvBYiUOHp+q+dsydlzZ/KOUaNcfKX9jnj8HBQR8dHc26GiIix6kk0NUz0F/kyQ2Xplyj45nZLncfrC1XxrWISAoaTX89fORobjKsaylIiIikoH9JIfK1Q4en2PjQ7lwGCgUJEZGEDY+VePOtow2PydNSHNUUJEREErZlZC9Tx+Ye/83LUhzVFCRERBLW7M0/L0txVFOQEBFJWDM3/2KhLzdLcVRTkBARSdj6tWdTLPRFvr50SSF3SXQVnSTTiYhIEyo3/y0je2c2G6oeoXhr6hijLx9ky8hefjkxyen9RdavPTsXQUPJdCIiKYpKqqsNHMVCX6qtCyXTiYjkQNQgdu3X9cmpaW7bvqfusWlSkBARSUFlb+tW+m4mJqcyT7BTkBARSVjt3tatyDrBTkFCRCRhrextXSvrBDsFCRGRBA2PldpqQVRknWCnICEikpBKN1O78pBgpzwJEZGEtNrN1F8scOLihbnKlVBLQkQkIY26mQp9x292XSz08eH3npZ0lVqmICEikoBGU1f7zNhy9XsZ6C/O7G390QsHeHBXefzCKQeYPOwxoe4mEZEENJq6Ou3O0KqB47qS1mzeMatrqpJQl2WXk1oSIiIJaDR1daDOjKWo47NOqFOQEBFJQNTUVYO6M5YaTXXNMqFOQUJEJAH1lgc34PrVZ9btPmo01TXLhDoFCRGRBAytGuCOq84/bnD6zmsu4AtD50cev3RJoe5rWSbUKUiIiOTEpivOm9X6yDqhTrObREQSUMm2rsxYqkxpBep2Nw2PlWaS7/rMmHZnIAcJdWpJiIgkoF629eTUdN1B6NpVYqfdZ1oQyrgWEZmHogab65VHBZSbt46zZvMOTYEVEZlvogab65U3mr2Udea1goSISALqTYEFOHzk6Kwb/lyzl6K6qdKgICEikoDKFNj+4vHTWg8dnprVMogKKNWyypVQkBARScjQqgFOXDx7Emlty6A6pyJKVrkSChIiIglqdgB7aNUAT264lLuuuSBXuRLKkxARSUAl78EjXo9qGVSmvG4Z2ZuLzYcUJEREYlabSFcr6yzqVihIiIjErNG2pXNlUbeaqZ00jUmIiMQsahzCgCc3XNrwZt9KpnYaFCRERGLWSiJdrVYytdOgICEiErN6eQ/NjkN0EmCSoCAhIhKzentJ3HHV+U2NKXQSYJIw58C1mZ0A/COwOBz/fXffZGZnAfcDpwC7gH/r7kfMbDHwHeBC4LfANe7+i/C7NgI3AtPAf3T3kVC+DvgK0Ad80903h/K654jps4uIJGZo1UBbA815mwJr7lGzeMMBZgac6O5vmlkB+L/Ap4FbgIfc/X4z++/AM+7+dTP7K+A97v7vzexa4M/d/RozOxf4HnARcDrwD8C/Dqf5F+ADwH7gaeA6d3/OzLbVO0ej+g4ODvro6GhbF0NEpFeZ2S53H6wtn7O7ycveDE8L4ceBS4Hvh/J7gaHw+MrwnPD6n4VAcyVwv7v/wd1fAvZRDhgXAfvc/eehlXA/cGV4T9Q5RERya3isxJrNOzhrww8yX+q7U02NSZhZn5mNA68DjwEvAhPufjQcsh+otIUGgFcBwutvUO4umimveU9U+SkNzlFbv5vMbNTMRg8cONDMRxIRSUT1BkJO9kt9d6qpIOHu0+5+AbCc8jf/cxKtVYvc/W53H3T3wWXLlmVdHRHpYXnLc+hUS7Ob3H0CeAJ4P9BvZpWB7+VAJUyWgDMAwusnUR7AnimveU9U+W8bnENEJJfylufQqTmDhJktM7P+8LhIeYD5ecrB4upw2A3Aw+Hx9vCc8PoOL4+ObweuNbPFYdbSSuAnlAeqV5rZWWa2CLgW2B7eE3UOEZFcylueQ6eaaUmcBjxhZs9SvqE/5u6PAp8DbjGzfZTHD+4Jx98DnBLKbwE2ALj7HmAb8Bzw98AnQzfWUeBTwAjl4LMtHEuDc4iI5FLe8hw6NecU2G6jKbAikrXKMuF5yHNoVtQUWK0CKyISs3YT6arlJdAoSIiI5EyelgvX2k0iIjkTNY32s9ueST3fQi0JEZEYxdFNFDVddto99RaFWhIiIjGJK9u60XTZtBPzFCRERGISV7Z1vWm01dJMzFN3k4hITOLKtq50JX1m2zj1shTSTMxTS0JEJCZxZ1svNJtVVuizVBPzFCRERGISZ7b1lpG9TB2b3Yw4cdHCVKfBKkiIiMRo8cK3b6tLlxSa3ra0Vimii2picqrturVDQUJEJAaVmU3VN/FDh6e4/ZE9beU29NXpaqo+V1oUJEREYlBvZhOUA0U702CnG6yrt/6B9JLqFCRERGLQaAZTO9NgBxoMdk8dc27bvify9TgpSIiIxGCuGUytToOdK1cirbEJBQkRkRjMdVNvdRrs0KoB7rjq/E6r1TEFCRGRGFRu6v3FwqzX2p0GO7RqgKVLZv8+ILI8bgoSIiIxGVo1wPimy7jrmgsY6C9ilMcW2p0GC7DpivMo9B0/06nQZ2y64rwYajw3LcshIhKzODYdqv5dQGYbEClIiIjkXJxBp1UKEiIiXSCr7UwVJEREci7L7Uw1cC0iknNx7VPRDrUkRERiklSXUFz7VLRDLQkRkRjEtXVpPXHvU9EKBQkRkRgk2SUU5z4VrVJ3k4hIDJLsEsoyV0JBQkQkBqf3F+tuFBRXl1BWuRLqbhIRiUGWXUJJUktCRCQGaXQJZZFQpyAhIhKTJLuEskqoU5AQEYlB0t/yG82eUpAQEcmxNL7lZ5VQp4FrEZEOpbFsRlYJdQoSIiIdSuNbftT2qIePHI0lqzuKgoSISIfS+JZf2R61WDj+tn3o8FRsy3/UoyAhItKhNHMk3po6NqssyRVhNXAtItKhtJbN2DKyF494LakBbAUJEZEYpLFsRqNAkNQA9pzdTWZ2hpk9YWbPmdkeM/t0KD/ZzB4zsxfCn0tDuZnZV81sn5k9a2bvq/pdN4TjXzCzG6rKLzSz3eE9XzUza3QOEZE8GR4rsWbzDs7a8APWbN6R2PhAVCAwSGz5j2bGJI4Cn3X3c4HVwCfN7FxgA/C4u68EHg/PAT4IrAw/NwFfh/INH9gEXAxcBGyquul/HfhE1fvWhfKoc4iI5EKS+0jUqjf2YcD1q89MrBUzZ5Bw99fc/Z/D4/8HPA8MAFcC94bD7gWGwuMrge942U6g38xOA9YCj7n7QXc/BDwGrAuvvdPdd7q7A9+p+V31ziEikgtpby26eOHbt+2lSwrcec0FfGHo/ETOBS3ObjKzFcAq4CngVHd/Lbz0K+DU8HgAeLXqbftDWaPy/XXKaXCO2nrdZGajZjZ64MCBVj6SiEhH0sqErrRYJianZsrqzXSKW9NBwszeATwI3Ozuv6t+LbQAogbdY9HoHO5+t7sPuvvgsmXLkqyGiMhx0sqETrvFUtFUkDCzAuUAcZ+7PxSKfx26igh/vh7KS8AZVW9fHsoalS+vU97oHCIiuZBWjkRu124KM43uAZ539y9XvbQdqMxQugF4uKr842GW02rgjdBlNAJcZmZLw4D1ZcBIeO13ZrY6nOvjNb+r3jlERHKhkgk90F/EgIH+IndcdX7sA8lZrd1k5V6cBgeY/Qnwf4DdQKUD7G8oj0tsA84EXgY+5u4Hw43+7yjPUDoM/KW7j4bf9e/CewG+6O7/M5QPAt8GisCPgP/g7m5mp9Q7R6P6Dg4O+ujoaNMXQESkG9SuNFuxdEmBTVec13FQMrNd7j44q3yuINFtFCREJG1p7Rg3PFbitu17jhu8hnL3Vqetl6ggobWbREQ6kGaexNCqAU5cPHuhjCQHsBUkREQ6kPaso7QHsBUkREQ6kPZNO+0BbAUJEZEOpH3TTnNZclCQEBHpSNo37cqU2/5iYabshEJyt3ItFS4i0oG09pKo9Yejby/JUdmdrro+cVGQEBFpU+3U1zuvuSDx4ACNB8sVJEREcqA2ua0y9RXi/zZfK83Bco1JiIi0IasF9yDdwXIFCRGRNmS14B7AJefUX+06qrwTChIiIm2I+tZ+UtWso6Q88bP6++ZElXdCQUJEpA3r155NYYHNKv/9kaOJ7XFdoTEJEZGcG1o1wDtOmD33Z2raEx+X0JiEiEgXmDg8Vbc86XGJNBP4FCRERNrUv6T++EPSGwGlmXWtICEi0obhsRJvvnV0VnmhzxJbkqNWvazruMdDFCRERNqwZWQvU8dmb9p24qKFmWddx0lBQkSkDVHjDm9M1h+nSOv8cY+HKEiIiLQhq/GIuc4T9/kVJEREWpSH8Yi0ZjhpgT8RkRZlPR4B6S1RriAhItKirMcjKoZWDSQelNTdJCLSorS3LM2SgoSISIvS3rI0S+puEhFpw+KFC2byFJYuKbDpivNSG4+oVrs7XtzjEgoSIiItqN2RDuCtqWMN3pFeXZLYHU/dTSIiLchyR7os6qIgISLSglLEzKao8iSlURcFCRGRFvTZ7I2GGpUnKY26KEiIiLRg2mcn0TUqT1IadVGQEBFpwUBELkRUeZLSqIuChIhIC/KUI5FGXTQFVkSkRXnJkUhj/SYFCRGRJuUpR6Ii6fWb1N0kItKkPOVIpEVBQkSkSWntBpcnChIiIk3qpdVfKxQkRESadMk5y6hNU5uvq79WKEiIiDRheKzEg7tKVKepGfDRC5Pf+CdLcwYJM/uWmb1uZj+tKjvZzB4zsxfCn0tDuZnZV81sn5k9a2bvq3rPDeH4F8zshqryC81sd3jPV83K+eRR5xARyUK9QWsHnvjZgWwqlJJmpsB+G/g74DtVZRuAx919s5ltCM8/B3wQWBl+Lga+DlxsZicDm4BBytd1l5ltd/dD4ZhPAE8BPwTWAT9qcA6RVFSv039SsYAZTByeSmwvYcm3Xhy0hiZaEu7+j8DBmuIrgXvD43uBoary73jZTqDfzE4D1gKPufvBEBgeA9aF197p7jvd3SkHoqE5ziGSuOu/8WNu3jpOaWISByYmpzh0eAqnvMLmZ7aOc+vw7qyrKSnqxUFraH9M4lR3fy08/hVwang8ALxaddz+UNaofH+d8kbnmMXMbjKzUTMbPXBgfjf9JHm3Du/myRdrvxcdz4Hv7nxFgaKHrDilfjC45JxlKdckXR0PXIcWQKLLH851Dne/290H3X1w2bL5/RcmyfveU6/OfVDw3Z2vMDxWSrA2kgfDYyX+KeKLw3wfk2g3SPw6dBUR/nw9lJeAM6qOWx7KGpUvr1Pe6BwiiRgeK3HB7f+75WWWb9u+J6EaSV5sGdkb+S11vo9JtLt203bgBmBz+PPhqvJPmdn9lAeu33D318xsBPgvVTOULgM2uvtBM/udma2mPHD9ceC/znEOkdgNj5VY/8AzTB1rvVE8MTnFms07KE1M0mfGtDsDGtyeVxrt9DbfxyTmDBJm9j3gT4F3mdl+yrOUNgPbzOxG4GXgY+HwHwKXA/uAw8BfAoRg8LfA0+G4/+zulbbbX1GeQVWkPKvpR6E86hwiHametVSZqbRlZG9bAaKichOptEKS2JBeslMJ/vXM50Q6APMMdlNK0uDgoI+OjmZdDcmpeqt4FhZYRwGikYH+Ik9uuDSR3y3pWbHhB5Gv/WLzh1KsSXLMbJe7D9aWK+Naekq9hKikAgTEuyG9ZGfpkkLd8ix2o0ub9pOQnpLFION5n/97Dh+ZVhJelxoeK/HmW0dnlRf6bN53NYFaEtJjshhk/P2R6ZkkvI0P7daU2S4TNV514qKFPRHwFSSkp7Sa+LSgdsnPDk1OTfPZbc8oUHSRqNbnG5NTKdckGwoS0lNaTXyqN1zR6X+aaXe1KLpIf8R4xHyf+lqhICE9pd0xiT4zjPJA5UkRN41WzPctL+eLXh+PAA1cyzxXnvL6LJMdblZ/zJ2XwlTHsxpMh2zFfM/U7XbDYyU+u+2ZuvkRvTIeAQoSMs/ULu89EVO/cXXXwun9xXimtlq5vr1ys+kmlXyaqAS6XhmPAAUJmQcqgaH2xh1XgKjdnnL92rNnJeS1wx3Wf/8ZQFnZeVMvn6Zar4xHgIKEdLlbh3dz385XYl+GeKC/eNyyHdU38crjSotlQYMlG+YyNe3cvHWc0ZcP8oWh82Opu3SuUUtxvu9pXUtBQrpKnN1JRv3155tZSmNo1dv7Gkct9XEMmG4ym/u7O18BUKDIgUazzhYY3HHV+T3V8tPsJukalZtx9W5xnXCgryYRop1viUOrBrjjqvMZ6C/OzIC65qIzWv7P1co+FpKc2x+JXvr9nScUeipAgFoS0kXm6idux/Qxb9i11KzqlgXAms07Wl4TatqdNZt3aOmODA2PlTh0OPrLR1zjXN1EQUK6RlJTRpO4Kbdb19LEJDdvHeeB0Ve47xPvj7VOMrdmclduHd7dU92C6m6SrjA8VioPIiQgiaS2Tme/PPniQe2fnYFmpjbf12Nb1qolIbkVNbU1bkm0UOKYJvvdna9w385XOL2/yCXnLOPRZ16b6e5YuqTApivOU7dUzBptLlThlL9Y9Mq1V5CQXOpkO9FWJTHnvXaabLuforJ6bGX2U8Whw1N8Zus4tz+yh4nDU1qGPCbNTmXupWx5BQnJpb956NlEAkTttNck57xXD2ZX9sCOk8PMIGtpYpJbto6z/oFxqlcgOXFRH1/8896astmuVrr3eimZTmMSkjvDYyUOd7jWUmGBUeibPb31+tVnHjdVNa057+vXnk2x0JfoOY4BtZft90emuXnrOCs2/IB3b/yBxjki3Dq8e1ZrLYqS6UQy1u5A8omL+o7bAa7yuzqd3hqH2u6n0/uLHD5ytOF0y7gd8/I4xwOj+/nSR9+j1kXQTICIY5p0t1KQkNwYHitx2/Y9Lc9FX2DwFxefWXdaYp7+M9fmUsS1mmyr/nD0GLdsG5+pUy8bHis1FSDmysCfzxQkJBfaGai+65oLuvomF9tqsm045nDz1nFu2TYeGWCbUb1MSh5bcHNplF1d0UtdS/UoSEgu3P7InpYCxEB/Mbc3nmbFtZpsJypdUC8deLPl5L3aNatKE5Osf+AZsPLChZWyjQ+Vx0Hy+PfVTHdfHuudJg1cS+bmWgqh1nwZOKxd86m/WEgqX3BO7STv1VsmZeqYzwSIirzuwtdMQtxAD81iiqKWhGSumSY/lKev5r37olW102SzXBuokrznlG+Oc13nVnIF8phXMFfgmi9fRjqlICGZaWVr0T4zXrzj8hRqlZ083EgrbYBmuolaGVPJY15Bo+utjPa3qbtJMnH9N37MzVvHm957+rqLz0i4RtmLupFm1QU1OfV2jsWazTtmdc/Uy/2Iyk/J2zfyRmuBLSksYOzzlylABGpJSCqSmN4639QbyC4W+vjohQM88bMDM7OFLjln2czzRkP9fWYcc49lFlVlddqbt47P+pbdbbOb5sqLWLQw2aTHbqMgIYkbHitxy7ZxWl1l49+s7o3gUBF10210g2203Mcxd17a/KE5j2vVocNTx225Wq9+eQoK1a7/xo958sWDDY95owf3jGhEQUISUT1/HupvE9pIf7HQUwGiojbhbi7r157NZ7aO172+1d1XSUy3/e7OV477Rt5fLGBGbhccvHV495wBAvI5fpIlBQmJ3a3Du2dmybSjb4Fx20fOi7VO89XQqgFGXz4463rXjgPUtlIWNLEkdququxJLE5N8pqq10Ui9hLy4gks73Zx5Gz/JmnnM/1CyNjg46KOjo1lXoye1MlspilYtbU+rN9raRLg01Pu7rVePYqGvrYUXO/1yAuXW0Pimyzr4Dd3LzHa5++CscgUJ6UQcgQFgzb86Wdt1pqw2sEwcPsLvj2SX/V2tmfWS2p0MEaXQZ2y5+r09+wVFQUJiEVdQqKYAkQ9ZtC7yQnkR0UFCYxJSV9zf0urpM+O6i8/oyQHqPKoet8hq4cG0FQsLeP5vP5h1NXJNQUKAePpzm6WWQ35Vz67qhZbFHVe9J+sq5J6CRA9KMyBUM+D6Hst96Gb18jZWnFLkn148mPq/nST0Fws93b3ULAWJeSSNLqJWabZSd6uXt1E74H3JOct49JnXZv7d1e4jnkfFQp+mWTdJA9c5UPlPV5qYpC/MX++G/2iNaCCwd+Wxm2oBcNKSQm4T/fKgaweuzWwd8BWgD/imu2+O+xxJzNhpVyXBqRsDxAKDL3+su3eLk+Y12pVucmp65gvPQGhtPLirlErgqLfelYJC+3IdJMysD/ga8AFgP/C0mW139+fiOsfwWIlbto6TfXjobmo59JZmdqWbdp/J/B5aNcDgH508E1ROKhb4/ZGjx21QVFhQDiqtrvFVWGC844SFaiUkJNdBArgI2OfuPwcws/uBK4HYgsSWkb0KEHNYuqTAh95zmr6ZyYyoXelqVXalq4xt1GZbz7WCbPWKt92yyux8k/cgMQC8WvV8P3Bx7UFmdhNwE8CZZ57Z0gnysNFLnvTS0tzSvjh2pYtazLCZG76CQnryHiSa4u53A3dDeeC6lffGsdZ+t9LMI2lXt+9KJ83Le5AoAdVbki0PZbFZv/bseTsm0cw+xSLtqLf0eGGBHTcmAfnclU5ak/cg8TSw0szOohwcrgX+Is4TVG6geZndVGuBwTHXDV/ypVt3pZPW5T5PwswuB+6iPAX2W+7+xUbHd2OehIhI1ro2T8Ldfwj8MOt6iIj0ogVZV0BERPJLQUJERCIpSIiISCQFCRERiZT72U2tMrMDwMttvv1dwG9irE430jXQNej1zw+9eQ3+yN2X1RbOuyDRCTMbrTcFrJfoGuga9PrnB12DaupuEhGRSAoSIiISSUHieHdnXYEc0DXQNej1zw+6BjM0JiEiIpHUkhARkUgKEiIiEklBIjCzdWa218z2mdmGrOuTBDP7lpm9bmY/rSo72cweM7MXwp9LQ7mZ2VfD9XjWzN6XXc3jY2ZnmNkTZvacme0xs0+H8p65DmZ2gpn9xMyeCdfg9lB+lpk9FT7rVjNbFMoXh+f7wusrsqx/XMysz8zGzOzR8LynPn+zFCQo/2MBvgZ8EDgXuM7Mzs22Von4NrCupmwD8Li7rwQeD8+hfC1Whp+bgK+nVMekHQU+6+7nAquBT4a/6166Dn8ALnX39wIXAOvMbDXwJeBOd/9j4BBwYzj+RuBQKL8zHDcffBp4vup5r33+5rh7z/8A7wdGqp5vBDZmXa+EPusK4KdVz/cCp4XHpwF7w+P/AVxX77j59AM8DHygV68DsAT4Z8p7x/8GWBjKZ/5PACPA+8PjheE4y7ruHX7u5ZS/DFwKPApYL33+Vn7UkigbAF6ter4/lPWCU939tfD4V8Cp4fG8vyah22AV8BQ9dh1CV8s48DrwGPAiMOHuR8Mh1Z9z5hqE198ATkm3xrG7C/hrmNm5+BR66/M3TUFCZnj5q1JPzIk2s3cADwI3u/vvql/rhevg7tPufgHlb9QXAedkXKXUmNmHgdfdfVfWdekGChJlJeCMqufLQ1kv+LWZnQYQ/nw9lM/ba2JmBcoB4j53fygU99x1AHD3CeAJyt0r/WZW2a2y+nPOXIPw+knAb1OuapzWAB8xs18A91PucvoKvfP5W6IgUfY0sDLMblgEXAtsz7hOadkO3BAe30C5j75S/vEwu2c18EZVd0zXMjMD7gGed/cvV73UM9fBzJaZWX94XKQ8JvM85WBxdTis9hpUrs3VwI7Q2upK7r7R3Ze7+wrK/9d3uPv19MjnbwCq4ooAAACnSURBVFnWgyJ5+QEuB/6Fct/sf8q6Pgl9xu8BrwFTlPtcb6Tct/o48ALwD8DJ4VijPOPrRWA3MJh1/WO6Bn9CuSvpWWA8/FzeS9cBeA8wFq7BT4HPh/J3Az8B9gEPAItD+Qnh+b7w+ruz/gwxXos/BR7t1c/fzI+W5RARkUjqbhIRkUgKEiIiEklBQkREIilIiIhIJAUJERGJpCAhIiKRFCRERCTS/wcUwScTVl81CAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### CART"
      ],
      "metadata": {
        "id": "O0cGnQk6DnTH"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class Node2:\n",
        "\n",
        "  def __init__(self,predicted_val,num_samples):\n",
        "          self.num_samples = num_samples\n",
        "          self.predicted_val = predicted_val\n",
        "          self.feature_index = 0\n",
        "          self.threshold = 0\n",
        "          self.left = None\n",
        "          self.right = None\n",
        "\n",
        "  # def is_leaf(self):\n",
        "  #         return ~(self.left or self.right)\n",
        "def standard_deviation(target):\n",
        "        avg=np.average(target)\n",
        "        sd=0;\n",
        "        n=len(target)\n",
        "        for i in target:\n",
        "          sd=sd+np.square((i-avg))\n",
        "        sd=np.sqrt(sd/n)\n",
        "        return sd\n",
        "\n",
        "\n",
        "\n",
        "class CART2:\n",
        "    \n",
        "\n",
        "    def __init__(self,max_depth = None):\n",
        "        self.max_depth = max_depth\n",
        "        \n",
        "\n",
        "    def fit(self, x, y):\n",
        "\n",
        "        self.n_features_=x.shape[1]\n",
        "        self.tree=self.grow_tree(x,y)\n",
        "\n",
        "    def grow_tree(self,X,y,depth=0):\n",
        "\n",
        "        predicted_val=np.mean(y)\n",
        "        node = Node2(num_samples=y.size,\n",
        "                    predicted_val=predicted_val)\n",
        "\n",
        "        if depth < self.max_depth:\n",
        "           \n",
        "            idx, thr = self._best_split(X, y)\n",
        "            \n",
        "            if idx is not None:\n",
        "                indices_left = X.iloc[:, idx] < thr\n",
        "                X_left, y_left = X[indices_left], y[indices_left]\n",
        "                X_right, y_right = X[~indices_left], y[~indices_left]\n",
        "                node.feature_index = int(idx)\n",
        "                node.threshold = float(thr)\n",
        "                node.left = self.grow_tree(X_left, y_left, depth + 1)\n",
        "                node.right = self.grow_tree(X_right, y_right, depth + 1)\n",
        "\n",
        "        return node\n",
        "\n",
        "    def predict(self, X,pred=[]):\n",
        "        \n",
        "        for inputs in X:\n",
        "                self._predict(inputs,pred=pred) \n",
        "        return pred\n",
        "\n",
        "   \n",
        "\n",
        "    def _predict(self, inputs,pred):\n",
        "\n",
        "        node = self.tree\n",
        "      \n",
        "        while node.left:\n",
        "            \n",
        "            if inputs[node.feature_index] < node.threshold:\n",
        "                node = node.left\n",
        "            else:\n",
        "                node = node.right\n",
        "        pred.append(node.predicted_val)\n",
        "        \n",
        "\n",
        "    def _best_split(self, X, y):\n",
        "        \n",
        "        m = y.size\n",
        "        \n",
        "        if m <= 1:\n",
        "            return None, None\n",
        "\n",
        "        sd_parent =standard_deviation(y)\n",
        "        \n",
        "        best_sd = sd_parent\n",
        "        best_idx, best_thr = None, None\n",
        "\n",
        "        for idx in range(self.n_features_):\n",
        "            thresholds, Val = zip(*sorted(zip(X.iloc[:, idx], y)))\n",
        "\n",
        "\n",
        "\n",
        "            sd_left=[0]\n",
        "            sd_right=Val\n",
        "            sd_right=sd_right[1:]\n",
        "\n",
        "            \n",
        "\n",
        "            for i in range(1,m):\n",
        "\n",
        "                split=(thresholds[i]+thresholds[i-1])/2\n",
        "\n",
        "                sd_l= standard_deviation(sd_left)\n",
        "                sd_r=standard_deviation(sd_right)\n",
        "              \n",
        "                sd_left.append(sd_right[0])\n",
        "                sd_right=sd_right[1:]\n",
        "\n",
        "                child_sd=(i*sd_l +(m-i)*sd_r)/m\n",
        "              \n",
        "                if thresholds[i] == thresholds[i - 1]:\n",
        "                    continue\n",
        "\n",
        "                if child_sd < best_sd:\n",
        "                   \n",
        "                    best_sd=child_sd\n",
        "                    best_idx=idx\n",
        "                    best_thr=split\n",
        "       \n",
        "        return best_idx, best_thr "
      ],
      "metadata": {
        "id": "P6CCFn1IDmlN"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "model=CART2(50)\n",
        "\n",
        "\n",
        "test =[]\n",
        "  \n",
        "for index, rows in X_test.iterrows():\n",
        "    my_list =[rows.retail_and_recreation, rows.grocery_and_pharmacy, rows.residential, rows.transit_stations, rows.parks, rows.workplaces]\n",
        "      \n",
        "    test.append(my_list)\n",
        "\n",
        "model.fit(X_train,y_train)\n",
        "\n",
        "Predictions=model.predict(test)"
      ],
      "metadata": {
        "id": "cy8dN8YGEDuf"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Predictions"
      ],
      "metadata": {
        "id": "gHzHerkyM1kG"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "ACCURACY AND PERFORMANCE"
      ],
      "metadata": {
        "id": "9Ui2J0AB6Mxm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "plt.figure(figsize=(8, 8))\n",
        "\n",
        "\n",
        "ax = sns.distplot(y_test, hist=False, color=\"r\", label=\"Actual Value\")\n",
        "sns.distplot(Predictions, hist=False, color=\"b\", label=\"Fitted Values\" , ax=ax)\n",
        "\n",
        "\n",
        "plt.title('Actual vs Fitted Values')\n",
        "\n",
        "\n",
        "plt.show()\n",
        "plt.close()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 602
        },
        "id": "RLFDf36p7b_L",
        "outputId": "dd8a23bf-5871-479e-d6b2-438c4d28eb4f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "RMSE = np.sqrt(np.sum(((y_test-Predictions)**2)/len(y_test)))\n",
        "RMSE"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_q07GFQXM8WG",
        "outputId": "b5ed9e6a-1979-4e94-ce66-8345930ae36d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "25603.981582423992"
            ]
          },
          "metadata": {},
          "execution_count": 547
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "FEATURE CORRELATION WITH TARGET VARIABLE"
      ],
      "metadata": {
        "id": "whD0d13dTqB-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "final_df.corr()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "id": "9QvIVe10Tu-e",
        "outputId": "77970f8a-3f45-473a-fc37-313ccc0c7f27"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                       retail_and_recreation  grocery_and_pharmacy  \\\n",
              "retail_and_recreation               1.000000              0.820786   \n",
              "grocery_and_pharmacy                0.820786              1.000000   \n",
              "residential                        -0.925891             -0.840430   \n",
              "transit_stations                    0.970580              0.913562   \n",
              "parks                               0.906370              0.664009   \n",
              "workplaces                          0.928552              0.867805   \n",
              "new_cases                          -0.134351             -0.052792   \n",
              "\n",
              "                       residential  transit_stations     parks  workplaces  \\\n",
              "retail_and_recreation    -0.925891          0.970580  0.906370    0.928552   \n",
              "grocery_and_pharmacy     -0.840430          0.913562  0.664009    0.867805   \n",
              "residential               1.000000         -0.928817 -0.768427   -0.969960   \n",
              "transit_stations         -0.928817          1.000000  0.880851    0.951951   \n",
              "parks                    -0.768427          0.880851  1.000000    0.794867   \n",
              "workplaces               -0.969960          0.951951  0.794867    1.000000   \n",
              "new_cases                 0.313037         -0.153770 -0.119766   -0.232405   \n",
              "\n",
              "                       new_cases  \n",
              "retail_and_recreation  -0.134351  \n",
              "grocery_and_pharmacy   -0.052792  \n",
              "residential             0.313037  \n",
              "transit_stations       -0.153770  \n",
              "parks                  -0.119766  \n",
              "workplaces             -0.232405  \n",
              "new_cases               1.000000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-05209980-74fe-4f57-b9c0-bf58be47af30\">\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>retail_and_recreation</th>\n",
              "      <th>grocery_and_pharmacy</th>\n",
              "      <th>residential</th>\n",
              "      <th>transit_stations</th>\n",
              "      <th>parks</th>\n",
              "      <th>workplaces</th>\n",
              "      <th>new_cases</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>retail_and_recreation</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.820786</td>\n",
              "      <td>-0.925891</td>\n",
              "      <td>0.970580</td>\n",
              "      <td>0.906370</td>\n",
              "      <td>0.928552</td>\n",
              "      <td>-0.134351</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>grocery_and_pharmacy</th>\n",
              "      <td>0.820786</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.840430</td>\n",
              "      <td>0.913562</td>\n",
              "      <td>0.664009</td>\n",
              "      <td>0.867805</td>\n",
              "      <td>-0.052792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>residential</th>\n",
              "      <td>-0.925891</td>\n",
              "      <td>-0.840430</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.928817</td>\n",
              "      <td>-0.768427</td>\n",
              "      <td>-0.969960</td>\n",
              "      <td>0.313037</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>transit_stations</th>\n",
              "      <td>0.970580</td>\n",
              "      <td>0.913562</td>\n",
              "      <td>-0.928817</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.880851</td>\n",
              "      <td>0.951951</td>\n",
              "      <td>-0.153770</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>parks</th>\n",
              "      <td>0.906370</td>\n",
              "      <td>0.664009</td>\n",
              "      <td>-0.768427</td>\n",
              "      <td>0.880851</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.794867</td>\n",
              "      <td>-0.119766</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>workplaces</th>\n",
              "      <td>0.928552</td>\n",
              "      <td>0.867805</td>\n",
              "      <td>-0.969960</td>\n",
              "      <td>0.951951</td>\n",
              "      <td>0.794867</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>-0.232405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>new_cases</th>\n",
              "      <td>-0.134351</td>\n",
              "      <td>-0.052792</td>\n",
              "      <td>0.313037</td>\n",
              "      <td>-0.153770</td>\n",
              "      <td>-0.119766</td>\n",
              "      <td>-0.232405</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-05209980-74fe-4f57-b9c0-bf58be47af30')\"\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-05209980-74fe-4f57-b9c0-bf58be47af30 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-05209980-74fe-4f57-b9c0-bf58be47af30');\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": 548
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Clearly, new_cases is correlated with residential only... the other features are showing negative correlation. Hence for clustering, we are taking only 'residential' in consideration"
      ],
      "metadata": {
        "id": "11qbbymOTxSo"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "# II\n",
        "Perform K-Means clustering on data examples. Using subsets in different\n",
        "clusters, design a “K-means Forest”. Predict results by taking the average (or\n",
        "weighted average by comparing test examples with cluster centroid) of results\n",
        "predicted by trees in the forest."
      ],
      "metadata": {
        "id": "ls9WVVLDB408"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "K MEANS CLUSTERING "
      ],
      "metadata": {
        "id": "xN_LvJ_bEgn2"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "data = final_df\n",
        "\n",
        "def initilize_centroids(data, k):\n",
        "    n_dims = data.shape[1]\n",
        "    centroids = [ [] for i in range(k) ]\n",
        "\n",
        "    for i in range(n_dims):\n",
        "        nd_column = data.iloc[:,i]\n",
        "        centroid_min = nd_column.min()\n",
        "        centroid_max = nd_column.max()\n",
        "        points = np.random.uniform(centroid_min, centroid_max, k)\n",
        "        for i in range(k):\n",
        "            centroids[i].append(points[i])\n",
        "        \n",
        "\n",
        "    centroids = pd.DataFrame(centroids, columns = data.columns)\n",
        "    return centroids\n",
        "\n",
        "def calculate_error(a, b):\n",
        "    return np.square(np.sum((a-b)**2))\n",
        "\n",
        "def assign_centroid(data, centroids):\n",
        "    '''\n",
        "        Assigns centroid to each of the data entries\n",
        "        to form cluster with the same. Using index for assignment\n",
        "    '''\n",
        "\n",
        "    n_observations = data.shape[0]\n",
        "    centroid_assign = []\n",
        "    centroid_errors = []\n",
        "    k = centroids.shape[0]\n",
        "\n",
        "    for entry in range(n_observations):\n",
        "        errors = []\n",
        "        for i in range(k):\n",
        "            errors.append(calculate_error(data.iloc[entry], centroids.iloc[i]))\n",
        "        centroid_index = errors.index(min(errors))\n",
        "        centroid_assign.append(centroid_index)\n",
        "        centroid_errors.append(min(errors))\n",
        "        #print(centroid_index, min(errors))\n",
        "    return (centroid_assign, centroid_errors)\n",
        "    \n",
        "\n",
        "\n",
        "\n",
        "def kmeans(data, k):\n",
        "    '''\n",
        "        Cretaing k clusters for the given dataset a\n",
        "    '''\n",
        "\n",
        "    centroids = initilize_centroids(data, k)\n",
        "    error = []\n",
        "    data_columns = data.columns\n",
        "    data['centroid'] = 0\n",
        "    data['error'] = 0\n",
        "    for i in range(10):\n",
        "        data['centroid'], data['error'] = assign_centroid(data.iloc[:,:2], centroids)\n",
        "        centroids = data.groupby('centroid').agg('mean').loc[:, data_columns].reset_index(drop = True)\n",
        "        # print(\"Total error = \", data['error'].sum())"
      ],
      "metadata": {
        "id": "NBgBKR3-wkrT"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# data\n",
        "kmeans(data,4)\n",
        "data"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "CbtBMuAHxOg4",
        "outputId": "039f0997-2688-4fa9-c8b3-02ee57d1518b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     retail_and_recreation  grocery_and_pharmacy  residential  \\\n",
              "0                    0.667                 1.667        0.000   \n",
              "1                    0.500                 1.750        0.000   \n",
              "2                    0.400                 1.800        0.200   \n",
              "3                    0.500                 2.000        0.000   \n",
              "4                   -0.143                 1.714        0.714   \n",
              "..                     ...                   ...          ...   \n",
              "466                -61.714               -25.000       24.143   \n",
              "467                -61.286               -24.429       23.714   \n",
              "468                -61.143               -24.714       23.714   \n",
              "469                -60.143               -23.429       23.286   \n",
              "470                -58.714               -21.143       22.714   \n",
              "\n",
              "     transit_stations   parks  workplaces  new_cases  centroid          error  \n",
              "0               2.000   3.000       3.000        0.0         3  420755.120750  \n",
              "1               2.000   3.250       3.000        0.0         3  408661.700720  \n",
              "2               1.800   2.800       3.200        0.0         3  401540.469492  \n",
              "3               2.333   3.167       3.333        0.0         3  403445.891827  \n",
              "4               1.429   3.571       0.143        0.0         3  371577.089154  \n",
              "..                ...     ...         ...        ...       ...            ...  \n",
              "466           -49.143 -41.000     -45.429   173790.0         0     964.550425  \n",
              "467           -48.714 -40.000     -44.571   165553.0         0    1057.910711  \n",
              "468           -49.000 -39.143     -44.286   152734.0         0    1252.471920  \n",
              "469           -48.286 -38.000     -43.429   127510.0         0    1800.518610  \n",
              "470           -47.000 -36.571     -42.429   132788.0         0    3901.816195  \n",
              "\n",
              "[471 rows x 9 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-185ee324-bcee-45c7-8973-48a34b33bd54\">\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>retail_and_recreation</th>\n",
              "      <th>grocery_and_pharmacy</th>\n",
              "      <th>residential</th>\n",
              "      <th>transit_stations</th>\n",
              "      <th>parks</th>\n",
              "      <th>workplaces</th>\n",
              "      <th>new_cases</th>\n",
              "      <th>centroid</th>\n",
              "      <th>error</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.667</td>\n",
              "      <td>1.667</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>3.000</td>\n",
              "      <td>3.000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>420755.120750</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.500</td>\n",
              "      <td>1.750</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.000</td>\n",
              "      <td>3.250</td>\n",
              "      <td>3.000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>408661.700720</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.400</td>\n",
              "      <td>1.800</td>\n",
              "      <td>0.200</td>\n",
              "      <td>1.800</td>\n",
              "      <td>2.800</td>\n",
              "      <td>3.200</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>401540.469492</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.500</td>\n",
              "      <td>2.000</td>\n",
              "      <td>0.000</td>\n",
              "      <td>2.333</td>\n",
              "      <td>3.167</td>\n",
              "      <td>3.333</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>403445.891827</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-0.143</td>\n",
              "      <td>1.714</td>\n",
              "      <td>0.714</td>\n",
              "      <td>1.429</td>\n",
              "      <td>3.571</td>\n",
              "      <td>0.143</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>371577.089154</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>466</th>\n",
              "      <td>-61.714</td>\n",
              "      <td>-25.000</td>\n",
              "      <td>24.143</td>\n",
              "      <td>-49.143</td>\n",
              "      <td>-41.000</td>\n",
              "      <td>-45.429</td>\n",
              "      <td>173790.0</td>\n",
              "      <td>0</td>\n",
              "      <td>964.550425</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>467</th>\n",
              "      <td>-61.286</td>\n",
              "      <td>-24.429</td>\n",
              "      <td>23.714</td>\n",
              "      <td>-48.714</td>\n",
              "      <td>-40.000</td>\n",
              "      <td>-44.571</td>\n",
              "      <td>165553.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1057.910711</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>468</th>\n",
              "      <td>-61.143</td>\n",
              "      <td>-24.714</td>\n",
              "      <td>23.714</td>\n",
              "      <td>-49.000</td>\n",
              "      <td>-39.143</td>\n",
              "      <td>-44.286</td>\n",
              "      <td>152734.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1252.471920</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>469</th>\n",
              "      <td>-60.143</td>\n",
              "      <td>-23.429</td>\n",
              "      <td>23.286</td>\n",
              "      <td>-48.286</td>\n",
              "      <td>-38.000</td>\n",
              "      <td>-43.429</td>\n",
              "      <td>127510.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1800.518610</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>470</th>\n",
              "      <td>-58.714</td>\n",
              "      <td>-21.143</td>\n",
              "      <td>22.714</td>\n",
              "      <td>-47.000</td>\n",
              "      <td>-36.571</td>\n",
              "      <td>-42.429</td>\n",
              "      <td>132788.0</td>\n",
              "      <td>0</td>\n",
              "      <td>3901.816195</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>471 rows × 9 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-185ee324-bcee-45c7-8973-48a34b33bd54')\"\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-185ee324-bcee-45c7-8973-48a34b33bd54 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-185ee324-bcee-45c7-8973-48a34b33bd54');\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": 550
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Use CART to create trees in the forest."
      ],
      "metadata": {
        "id": "lS5DA8lyCDDC"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Forest Construction on CART"
      ],
      "metadata": {
        "id": "YZZyTc_xMYIx"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Redefination"
      ],
      "metadata": {
        "id": "gLB_zrdrMX7a"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class Node:\n",
        "\n",
        "  def __init__(self,predicted_val,num_samples):\n",
        "          self.num_samples = num_samples\n",
        "          self.predicted_val = predicted_val\n",
        "          self.feature_index = 1\n",
        "          self.threshold = 0\n",
        "          self.left = None\n",
        "          self.right = None\n",
        "\n",
        "  def is_leaf_node(self):\n",
        "    \n",
        "        return self.predicted_val is not None"
      ],
      "metadata": {
        "id": "HkobjKC1OPrw"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class CART:\n",
        "    \n",
        "\n",
        "    def __init__(self,min_samples_split=2,n_feats=None,max_depth=None):\n",
        "        self.max_depth = max_depth\n",
        "        self.min_samples_split = min_samples_split\n",
        "        self.n_feats = n_feats\n",
        "        self.root = None\n",
        "\n",
        "    def fit(self, x, y):\n",
        "\n",
        "        self.n_feats=x.shape[1] if not self.n_feats else min(self.n_feats, X.shape[1])\n",
        "        self.root=self.grow_tree(x,y)\n",
        "\n",
        "\n",
        "    def grow_tree(self,X,y,depth=0):\n",
        "\n",
        "        n_samples, n_features = X.shape\n",
        "\n",
        "\n",
        "        feat_idxs = np.random.choice(n_features, self.n_feats, replace=False)\n",
        "\n",
        "        predicted_val=np.mean(y)\n",
        "        node = Node(num_samples=y.size,\n",
        "                    predicted_val=predicted_val)\n",
        "\n",
        "        if depth < self.max_depth and n_samples >= self.min_samples_split:\n",
        "           \n",
        "            idx, thr = self._best_split(X, y)\n",
        "            \n",
        "            if idx is not None:\n",
        "                indices_left = X.iloc[:, idx] < thr\n",
        "                X_left, y_left = X[indices_left], y[indices_left]\n",
        "                X_right, y_right = X[~indices_left], y[~indices_left]\n",
        "                node.feature_index = int(idx)\n",
        "                node.threshold = float(thr)\n",
        "                node.left = self.grow_tree(X_left, y_left, depth + 1)\n",
        "                node.right = self.grow_tree(X_right, y_right, depth + 1)\n",
        "\n",
        "        return node\n",
        "\n",
        "    def predicts(self, X):\n",
        "        return np.array([self._predict(inputs) for inputs in X])\n",
        "    \n",
        "\n",
        "    def _predict(self, inputs):\n",
        "\n",
        "        node = self.root\n",
        "      \n",
        "        while node.left:\n",
        "            \n",
        "            if inputs[node.feature_index] < node.threshold:\n",
        "                node = node.left\n",
        "            else:\n",
        "                node = node.right\n",
        "\n",
        "        return(node.predicted_val)\n",
        "        \n",
        "\n",
        "    def _best_split(self, X, y):\n",
        "        \n",
        "        m = y.size\n",
        "        \n",
        "        if m <= 1:\n",
        "            return None, None\n",
        "\n",
        "        sd_parent =standard_deviation(y)\n",
        "        \n",
        "        best_sd = sd_parent\n",
        "        best_idx, best_thr = None, None\n",
        "\n",
        "        for idx in range(self.n_feats):\n",
        "            thresholds, Val = zip(*sorted(zip(X.iloc[:, idx], y)))\n",
        "\n",
        "\n",
        "\n",
        "            sd_left=[0]\n",
        "            sd_right=Val\n",
        "            sd_right=sd_right[1:]\n",
        "\n",
        "            \n",
        "\n",
        "            for i in range(1,m):\n",
        "\n",
        "                split=(thresholds[i]+thresholds[i-1])/2\n",
        "\n",
        "                sd_l=standard_deviation(sd_left)\n",
        "                sd_r=standard_deviation(sd_right)\n",
        "              \n",
        "                sd_left.append(sd_right[0])\n",
        "                sd_right=sd_right[1:]\n",
        "\n",
        "                child_sd=(i*sd_l +(m-i)*sd_r)/m\n",
        "              \n",
        "                if thresholds[i] == thresholds[i - 1]:\n",
        "                    continue\n",
        "\n",
        "                if child_sd < best_sd:\n",
        "                   \n",
        "                    best_sd=child_sd\n",
        "                    best_idx=idx\n",
        "                    best_thr=split\n",
        "       \n",
        "        return best_idx, best_thr"
      ],
      "metadata": {
        "id": "9G8uYkp5N65h"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from collections import Counter\n",
        "\n",
        "def avg_val(y):\n",
        "    return(np.mean(y))\n",
        "\n",
        "class Forest:\n",
        "    \n",
        "    def __init__(self, n_trees=5, min_samples_split=2,\n",
        "                 max_depth=10, n_feats=None):\n",
        "        self.n_trees = n_trees\n",
        "        self.min_samples_split = min_samples_split\n",
        "        self.max_depth = max_depth\n",
        "        self.n_feats = n_feats\n",
        "        self.trees = []\n",
        "\n",
        "    def fit(self, X, y):\n",
        "        self.trees = []\n",
        "        for _ in range(self.n_trees):\n",
        "            tree = CART(min_samples_split=self.min_samples_split,\n",
        "                max_depth=self.max_depth, n_feats=self.n_feats)\n",
        "            X_samp, y_samp = X, y\n",
        "            tree.fit(X_samp, y_samp)\n",
        "            self.trees.append(tree)\n",
        "\n",
        "    def predict(self, X):\n",
        "        tree_preds = np.array([tree.predicts(X) for tree in self.trees])\n",
        "        tree_preds = np.swapaxes(tree_preds, 0, 1)\n",
        "        \n",
        "        y_pred = [avg_val(tree_pred) for tree_pred in tree_preds]\n",
        "        return np.array(y_pred)"
      ],
      "metadata": {
        "id": "Lz0yeU7UNCA0"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "X = data[['retail_and_recreation',\t'grocery_and_pharmacy',\t'residential',\t'transit_stations',\t'parks',\t'workplaces',\t'centroid',\t'error']]\n",
        "y = data['new_cases']\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)"
      ],
      "metadata": {
        "id": "q2U-sWNRy-L5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "clf = Forest(n_trees = 5)\n",
        "clf.fit(X_train,y_train)\n",
        "\n",
        "test=[]\n",
        "for index, rows in X_test.iterrows():\n",
        "    my_list =[rows.retail_and_recreation, rows.grocery_and_pharmacy, rows.residential, rows.transit_stations, rows.parks, rows.workplaces,rows.centroid,rows.error]\n",
        "\n",
        "    test.append(my_list)\n",
        "\n",
        "y_pred = clf.predict(test) "
      ],
      "metadata": {
        "id": "06Fqb3sxNEms"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qz9yjrFpNIkD",
        "outputId": "d374d624-2a00-444d-83be-2e017805dc4b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([3.84710182e+05, 1.69850000e+03, 2.37785556e+04, 1.01615714e+04,\n",
              "       5.18000000e+02, 3.70000000e+01, 5.62110000e+04, 1.43051579e+04,\n",
              "       1.61992000e+04, 7.32189286e+04, 1.64967500e+04, 7.32189286e+04,\n",
              "       6.30068750e+04, 5.45000000e+02, 3.41775000e+03, 1.09960000e+03,\n",
              "       1.52440000e+04, 0.00000000e+00, 2.55282500e+04, 7.21630000e+03,\n",
              "       4.15063590e+04, 4.15063590e+04, 4.86400000e+03, 7.32189286e+04,\n",
              "       5.38130000e+04, 1.38676000e+05, 9.32490000e+04, 2.37785556e+04,\n",
              "       3.62727000e+05, 3.41775000e+03, 1.64880000e+04, 1.61992000e+04,\n",
              "       2.37785556e+04, 2.37785556e+04, 6.00000000e+00, 1.64967500e+04,\n",
              "       6.30068750e+04, 5.72240000e+04, 5.27830000e+04, 3.76300000e+03,\n",
              "       1.52440000e+04, 4.15063590e+04, 4.67800000e+04, 7.32189286e+04,\n",
              "       3.41775000e+03, 1.01615714e+04, 1.64967500e+04, 1.84372000e+05,\n",
              "       2.22510000e+04, 2.35534125e+05, 1.64967500e+04, 6.30068750e+04,\n",
              "       1.09960000e+03, 5.27830000e+04, 1.28810000e+04, 7.64720000e+04,\n",
              "       7.21630000e+03, 8.82258333e+04, 1.59962500e+04, 1.61992000e+04,\n",
              "       3.52520000e+04, 5.38130000e+04, 2.00739000e+05, 1.09960000e+03,\n",
              "       0.00000000e+00, 7.32189286e+04, 5.43660000e+04, 1.43051579e+04,\n",
              "       1.03558000e+05, 9.32490000e+04, 1.64880000e+04, 4.15063590e+04,\n",
              "       3.84710182e+05, 1.64967500e+04, 1.01615714e+04, 2.35534125e+05,\n",
              "       9.11000000e+03, 1.43051579e+04, 9.32490000e+04, 9.01230000e+04,\n",
              "       2.87320000e+04, 7.23300000e+04, 1.28810000e+04, 2.32850000e+04,\n",
              "       0.00000000e+00, 1.61992000e+04, 5.38130000e+04, 5.20500000e+04,\n",
              "       7.32189286e+04, 2.76110000e+05, 2.00350000e+04, 5.18000000e+02,\n",
              "       1.28810000e+04, 7.32189286e+04, 1.01615714e+04])"
            ]
          },
          "metadata": {},
          "execution_count": 556
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "ACCURACY AND PERFORMANCE"
      ],
      "metadata": {
        "id": "rwfh7Lpk35Bq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "plt.figure(figsize=(8, 8))\n",
        "\n",
        "\n",
        "ax = sns.distplot(y_test, hist=False, color=\"r\", label=\"Actual Value\")\n",
        "sns.distplot(y_pred, hist=False, color=\"b\", label=\"Fitted Values\" , ax=ax)\n",
        "\n",
        "\n",
        "plt.title('Actual vs Fitted Values')\n",
        "\n",
        "\n",
        "plt.show()\n",
        "plt.close()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 602
        },
        "id": "CPbK1dNiATgN",
        "outputId": "3991f092-e9ba-4fd0-a211-bee5b5329030"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import mean_squared_error\n",
        "rmse=np.sqrt(mean_squared_error(y_test,y_pred))\n",
        "rmse\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0fEgfYnPNNzc",
        "outputId": "ab754acc-1bd3-4308-a6bc-54400b1b8da1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "24158.141365102838"
            ]
          },
          "metadata": {},
          "execution_count": 558
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "score=r2_score(y_test,y_pred)\n",
        "score"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "w5cRKYclOoaj",
        "outputId": "b03a08a3-e6fd-4530-bd0d-5c9b3e9d0a23"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9084702386203062"
            ]
          },
          "metadata": {},
          "execution_count": 559
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Use ID3-A to create trees in the forest."
      ],
      "metadata": {
        "id": "jpblYhd2CIrF"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "ID3A K MEANS FOREST"
      ],
      "metadata": {
        "id": "1mPFce5g3kiI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class Node_Id3:\n",
        "\n",
        "    def __init__(self, x, y, idxs, min_leaf=5):\n",
        "        self.x = x \n",
        "        self.y = y\n",
        "        self.idxs = idxs \n",
        "        self.min_leaf = min_leaf\n",
        "        self.row_count = len(idxs)\n",
        "        self.col_count = x.shape[1]\n",
        "        self.val = np.mean(y[idxs])\n",
        "        self.score = float('inf')\n",
        "        self.find_varsplit()\n",
        "      \n",
        "    def find_varsplit(self):\n",
        "        for c in range(self.col_count): self.find_better_split(c)\n",
        "        if self.is_leaf: return\n",
        "        x = self.split_col\n",
        "        lhs = np.nonzero(x <= self.split)[0]\n",
        "        rhs = np.nonzero(x > self.split)[0]\n",
        "        self.lhs = Node_Id3(self.x, self.y, self.idxs[lhs], self.min_leaf)\n",
        "        self.rhs = Node_Id3(self.x, self.y, self.idxs[rhs], self.min_leaf)\n",
        "        \n",
        "    def find_better_split(self, var_idx):\n",
        "      \n",
        "        x = self.x.values[self.idxs, var_idx]\n",
        "\n",
        "        for r in range(self.row_count):\n",
        "            lhs = x <= x[r]\n",
        "            rhs = x > x[r]\n",
        "            if rhs.sum() < self.min_leaf or lhs.sum() < self.min_leaf: continue\n",
        "\n",
        "            curr_score = self.find_score(lhs, rhs)\n",
        "            if curr_score < self.score: \n",
        "                self.var_idx = var_idx\n",
        "                self.score = curr_score\n",
        "                self.split = x[r]\n",
        "                \n",
        "    def find_score(self, lhs, rhs):\n",
        "        y = self.y[self.idxs]\n",
        "        lhs_std = y[lhs].std()\n",
        "        rhs_std = y[rhs].std()\n",
        "        return lhs_std * lhs.sum() + rhs_std * rhs.sum()\n",
        "                \n",
        "    @property\n",
        "    def split_col(self): return self.x.values[self.idxs,self.var_idx]\n",
        "                \n",
        "    @property\n",
        "    def is_leaf(self): return self.score == float('inf')                \n",
        "\n",
        "    def predict_arr(self, x):\n",
        "        return np.array([self.predict_row(xi) for xi in x])\n",
        "\n",
        "    def predict_row(self, xi):\n",
        "        if self.is_leaf: return self.val\n",
        "        node = self.lhs if xi[self.var_idx] <= self.split else self.rhs\n",
        "        return node.predict_row(xi)"
      ],
      "metadata": {
        "id": "UHd5Zp0NOux-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class ID3Af:\n",
        "\n",
        "  def __init__(self,min_leaf):\n",
        "    self.min_leaf=min_leaf\n",
        "  \n",
        "  def fit(self, X, y):\n",
        "    min_leaf = self.min_leaf\n",
        "    self.dtree = Node_Id3(X, y, np.array(np.arange(len(y))), min_leaf)\n",
        "    return self\n",
        "  \n",
        "  def predict_Id3(self, X):\n",
        "    return self.dtree.predict_arr(X)\n",
        "\n",
        "y_train= y_train.reset_index()\n",
        "y_train =y_train['new_cases']\n"
      ],
      "metadata": {
        "id": "n46g7v7n8WAH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from collections import Counter\n",
        "\n",
        "\n",
        "def avg_val(y):\n",
        "    return(np.mean(y))\n",
        "\n",
        "class ID3Forest:\n",
        "    \n",
        "    def __init__(self, n_trees=3,min_leaf=5):\n",
        "        self.n_trees = n_trees\n",
        "        self.min_leaf=min_leaf\n",
        "        self.trees = []\n",
        "\n",
        "    def fit(self, X, y):\n",
        "        self.trees = []\n",
        "        for _ in range(self.n_trees):\n",
        "            tree = ID3Af(min_leaf=self.min_leaf)\n",
        "            X_samp, y_samp = X, y\n",
        "            tree.fit(X_samp, y_samp)\n",
        "            self.trees.append(tree)\n",
        "\n",
        "    def predict(self, X):\n",
        "        tree_preds = np.array([tree.predict_Id3(X) for tree in self.trees])\n",
        "        tree_preds = np.swapaxes(tree_preds, 0, 1)\n",
        "        \n",
        "        y_pred = [avg_val(tree_pred) for tree_pred in tree_preds]\n",
        "        return np.array(y_pred)"
      ],
      "metadata": {
        "id": "gDpKI25fJE85"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "clf2 = ID3Forest(n_trees = 5)\n",
        "clf2.fit(X_train,y_train)\n",
        "\n",
        "test=[]\n",
        "\n",
        "for index, rows in X_test.iterrows():\n",
        "    my_list =[rows.retail_and_recreation, rows.grocery_and_pharmacy, rows.residential, rows.transit_stations, rows.parks, rows.workplaces,rows.centroid,rows.error]\n",
        "    test.append(my_list)\n",
        "\n",
        "y_predictions = clf2.predict(test) \n"
      ],
      "metadata": {
        "id": "H2VnmPEXJJk5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# y_pred"
      ],
      "metadata": {
        "id": "fVESBY4RJKqk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import mean_squared_error\n",
        "\n",
        "rmse=np.sqrt(mean_squared_error(y_test,y_predictions))"
      ],
      "metadata": {
        "id": "2nXqaVVaJQ5E"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "rmse"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FxY2Mt9aJSEI",
        "outputId": "778aede9-2c09-44bc-ceff-aaa52aaba161"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "33714.19424102068"
            ]
          },
          "metadata": {},
          "execution_count": 566
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "score=r2_score(y_test,y_predictions)\n",
        "score"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "haLIVIbNKnrY",
        "outputId": "1462d946-78c7-4850-e659-1c6799fa0107"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.8217371429120209"
            ]
          },
          "metadata": {},
          "execution_count": 567
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "plt.figure(figsize=(8, 8))\n",
        "\n",
        "\n",
        "ax = sns.distplot(y_test, hist=False, color=\"r\", label=\"Actual Value\")\n",
        "sns.distplot(y_predictions, hist=False, color=\"b\", label=\"Fitted Values\" , ax=ax)\n",
        "\n",
        "\n",
        "plt.title('Actual vs Fitted Values')\n",
        "\n",
        "\n",
        "plt.show()\n",
        "plt.close()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 602
        },
        "id": "800XOlP2LS8c",
        "outputId": "22bcceea-58c9-4ab2-ae31-6024d95e008a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Compare the accuracy and performance of K-means Forests created\n",
        "using CART and ID3-A for parts A. and B."
      ],
      "metadata": {
        "id": "fmFCpQL1CLQg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "plt.figure(figsize=(8, 8))\n",
        "\n",
        "\n",
        "ax = sns.distplot(y_test, hist=False, color=\"r\", label=\"Actual Value\")\n",
        "sns.distplot(y_predictions, hist=False, color=\"b\", label=\"Fitted Values\" , ax=ax)\n",
        "sns.distplot(y_pred, hist=False, color=\"g\", label=\"Fitted Values\" , ax=ax)\n",
        "\n",
        "plt.title('Actual vs Fitted Values ')\n",
        "\n",
        "\n",
        "plt.show()\n",
        "plt.close()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 637
        },
        "id": "bHBZ326lCOnj",
        "outputId": "227f33f9-8a2d-4d15-eb63-772b0198f62a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n",
            "/usr/local/lib/python3.7/dist-packages/seaborn/distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `kdeplot` (an axes-level function for kernel density plots).\n",
            "  warnings.warn(msg, FutureWarning)\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "RMSE AND R2 SCORE IS CALCULATED ABOVE FOR EACH ALGORITHM USED \n",
        "IN MY CASE ID3A IS GIVING BETTER RESULT THAN CART "
      ],
      "metadata": {
        "id": "wEz9jY46TeTe"
      }
    }
  ]
}