Clustering_No_Twitter.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.3"
    },
    "colab": {
      "name": "Clustering No Twitter.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/christophersingh/Mental-Health-Classification-Research-/blob/main/Clustering_No_Twitter.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KrhJmSD9tvnk"
      },
      "source": [
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "plt.rcParams[\"figure.figsize\"] = (20,10)\n",
        "from sklearn import model_selection\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.naive_bayes import GaussianNB\n",
        "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "6ixI6sgLtvny",
        "outputId": "d200bb65-7680-479d-eb9b-844630c546aa"
      },
      "source": [
        "#read in the data\n",
        "ny = pd.read_csv('./new_york.csv')\n",
        "ny"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>indicator</th>\n",
              "      <th>group</th>\n",
              "      <th>state</th>\n",
              "      <th>subgroup</th>\n",
              "      <th>phase</th>\n",
              "      <th>time_period</th>\n",
              "      <th>time_period_label</th>\n",
              "      <th>time_period_start_date</th>\n",
              "      <th>time_period_end_date</th>\n",
              "      <th>value</th>\n",
              "      <th>lowci</th>\n",
              "      <th>highci</th>\n",
              "      <th>confidence_interval</th>\n",
              "      <th>quartile_range</th>\n",
              "      <th>age</th>\n",
              "      <th>CASE_RATE</th>\n",
              "      <th>HOSPITALIZED_RATE</th>\n",
              "      <th>DEATH_RATE</th>\n",
              "      <th>date</th>\n",
              "      <th>CASE_COUNT</th>\n",
              "      <th>HOSPITALIZED_COUNT</th>\n",
              "      <th>DEATH_COUNT</th>\n",
              "      <th>counts</th>\n",
              "      <th>sentiment__negative</th>\n",
              "      <th>sentiment__positive</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>171.46</td>\n",
              "      <td>13.55</td>\n",
              "      <td>0.23</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>1554.45</td>\n",
              "      <td>153.69</td>\n",
              "      <td>12.88</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>2529.03</td>\n",
              "      <td>630.34</td>\n",
              "      <td>113.37</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>2552.64</td>\n",
              "      <td>1192.50</td>\n",
              "      <td>364.17</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>2976.74</td>\n",
              "      <td>1830.07</td>\n",
              "      <td>909.18</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</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",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35779</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>20207.72</td>\n",
              "      <td>1160.60</td>\n",
              "      <td>67.68</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>224684.00000</td>\n",
              "      <td>13326.000000</td>\n",
              "      <td>794.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>15941.19</td>\n",
              "      <td>2540.10</td>\n",
              "      <td>476.51</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>161559.00000</td>\n",
              "      <td>25652.000000</td>\n",
              "      <td>4800.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>7067.11</td>\n",
              "      <td>2418.35</td>\n",
              "      <td>756.82</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>50798.00000</td>\n",
              "      <td>17383.000000</td>\n",
              "      <td>5440.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>7477.43</td>\n",
              "      <td>3995.12</td>\n",
              "      <td>1989.81</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>41994.00000</td>\n",
              "      <td>22437.000000</td>\n",
              "      <td>11175.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>6302.86</td>\n",
              "      <td>958.30</td>\n",
              "      <td>266.66</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>525458.00000</td>\n",
              "      <td>79892.000000</td>\n",
              "      <td>22231.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 25 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                               indicator  ... sentiment__positive\n",
              "0                        Symptoms of Depressive Disorder  ...                50.0\n",
              "1                        Symptoms of Depressive Disorder  ...                50.0\n",
              "2                        Symptoms of Depressive Disorder  ...                50.0\n",
              "3                        Symptoms of Depressive Disorder  ...                50.0\n",
              "4                        Symptoms of Depressive Disorder  ...                50.0\n",
              "...                                                  ...  ...                 ...\n",
              "35779  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "35780  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "35781  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "35782  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "35783  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "\n",
              "[35784 rows x 25 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FGbwgPNWtvn0",
        "outputId": "67423f0f-f3a3-4a81-817a-8302bcc47f8b"
      },
      "source": [
        "ny[\"indicator\"].replace({\"Symptoms of Anxiety Disorder\": \"0\", \"Symptoms of Depressive Disorder\": \"1\", \"Symptoms of Anxiety Disorder or Depressive Disorder\": \"2\"}, inplace=True)\n",
        "ny[\"age\"].replace({\"18 - 29 years\": \"29\", \"30 - 39 years\": \"39\", \"40 - 49 years\": \"49\", \"50 - 59 years\":\"59\",\"60 - 69 years\":\"69\",\"70 - 79 years\":\"79\",\"80 years and above\":\"80\"}, inplace=True)\n",
        "\n",
        "data_to_use = ny\n",
        "del data_to_use['group']\n",
        "del data_to_use['state']\n",
        "del data_to_use['subgroup']\n",
        "del data_to_use['time_period_label']\n",
        "del data_to_use['confidence_interval']\n",
        "del data_to_use['quartile_range']\n",
        "del data_to_use['counts']\n",
        "\n",
        "ny = data_to_use\n",
        "ny.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 35784 entries, 0 to 35783\n",
            "Data columns (total 18 columns):\n",
            " #   Column                  Non-Null Count  Dtype  \n",
            "---  ------                  --------------  -----  \n",
            " 0   indicator               35784 non-null  object \n",
            " 1   phase                   35784 non-null  object \n",
            " 2   time_period             35784 non-null  float64\n",
            " 3   time_period_start_date  35784 non-null  object \n",
            " 4   time_period_end_date    35784 non-null  object \n",
            " 5   value                   35784 non-null  float64\n",
            " 6   lowci                   35784 non-null  float64\n",
            " 7   highci                  35784 non-null  float64\n",
            " 8   age                     35784 non-null  object \n",
            " 9   CASE_RATE               35784 non-null  float64\n",
            " 10  HOSPITALIZED_RATE       35784 non-null  float64\n",
            " 11  DEATH_RATE              33957 non-null  float64\n",
            " 12  date                    35784 non-null  object \n",
            " 13  CASE_COUNT              35784 non-null  float64\n",
            " 14  HOSPITALIZED_COUNT      35784 non-null  float64\n",
            " 15  DEATH_COUNT             35784 non-null  float64\n",
            " 16  sentiment__negative     35784 non-null  float64\n",
            " 17  sentiment__positive     35784 non-null  float64\n",
            "dtypes: float64(12), object(6)\n",
            "memory usage: 4.9+ MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5EFoAULztvn2",
        "outputId": "f086bdd2-6926-4b10-9787-fbd9d58895ea"
      },
      "source": [
        "ny['DEATH_RATE'].fillna((ny['DEATH_RATE'].mean()), inplace=True)\n",
        "ny.isna().sum()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "indicator                 0\n",
              "phase                     0\n",
              "time_period               0\n",
              "time_period_start_date    0\n",
              "time_period_end_date      0\n",
              "value                     0\n",
              "lowci                     0\n",
              "highci                    0\n",
              "age                       0\n",
              "CASE_RATE                 0\n",
              "HOSPITALIZED_RATE         0\n",
              "DEATH_RATE                0\n",
              "date                      0\n",
              "CASE_COUNT                0\n",
              "HOSPITALIZED_COUNT        0\n",
              "DEATH_COUNT               0\n",
              "sentiment__negative       0\n",
              "sentiment__positive       0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bRzqS3xDtvn3"
      },
      "source": [
        "#Add this cell after API change\n",
        "ny.loc[ny['phase'] == ('Oct 28 � Dec 21'), 'phase'] = 3\n",
        "ny.loc[ny['phase'] == ('3 (Oct 28 � Dec 21)'), 'phase'] = 4\n",
        "ny.loc[ny['phase'] == ('3 (Jan 6 � Feb 1)'), 'phase'] = 5\n",
        "ny.rename(columns = {'counts':'twitter_count'}, inplace = True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 444
        },
        "id": "6ZFTU8tRtvn4",
        "outputId": "0c7cd2ad-974b-4e2e-8018-4c5c18eb727c"
      },
      "source": [
        "from sklearn import preprocessing\n",
        "\n",
        "ny[\"age\"] = ny['age'].astype(float)\n",
        "ny[\"phase\"] = ny['phase'].astype(float)\n",
        "ny[\"indicator\"] = ny['indicator'].astype(float)\n",
        "ny[\"time_period\"] = ny['time_period'].astype(float)\n",
        "\n",
        "\"\"\"def min_max_normalization(df):\n",
        "    # copy the dataframe\n",
        "    df_norm = df.copy()\n",
        "    # apply min-max scaling\n",
        "    for column in df_norm.columns:\n",
        "        df_norm[column] = (df_norm[column] - df_norm[column].min()) / (df_norm[column].max() - df_norm[column].min())\n",
        "        \n",
        "    return df_norm\n",
        "\n",
        "ny = min_max_normalization(data_to_use)\"\"\"\n",
        "ny"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>indicator</th>\n",
              "      <th>phase</th>\n",
              "      <th>time_period</th>\n",
              "      <th>time_period_start_date</th>\n",
              "      <th>time_period_end_date</th>\n",
              "      <th>value</th>\n",
              "      <th>lowci</th>\n",
              "      <th>highci</th>\n",
              "      <th>age</th>\n",
              "      <th>CASE_RATE</th>\n",
              "      <th>HOSPITALIZED_RATE</th>\n",
              "      <th>DEATH_RATE</th>\n",
              "      <th>date</th>\n",
              "      <th>CASE_COUNT</th>\n",
              "      <th>HOSPITALIZED_COUNT</th>\n",
              "      <th>DEATH_COUNT</th>\n",
              "      <th>sentiment__negative</th>\n",
              "      <th>sentiment__positive</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>171.46</td>\n",
              "      <td>13.55</td>\n",
              "      <td>0.23</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>1554.45</td>\n",
              "      <td>153.69</td>\n",
              "      <td>12.88</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2529.03</td>\n",
              "      <td>630.34</td>\n",
              "      <td>113.37</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2552.64</td>\n",
              "      <td>1192.50</td>\n",
              "      <td>364.17</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2976.74</td>\n",
              "      <td>1830.07</td>\n",
              "      <td>909.18</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</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",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35779</th>\n",
              "      <td>2.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>20207.72</td>\n",
              "      <td>1160.60</td>\n",
              "      <td>67.68</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>224684.00000</td>\n",
              "      <td>13326.000000</td>\n",
              "      <td>794.000000</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>2.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>15941.19</td>\n",
              "      <td>2540.10</td>\n",
              "      <td>476.51</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>161559.00000</td>\n",
              "      <td>25652.000000</td>\n",
              "      <td>4800.000000</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>2.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>7067.11</td>\n",
              "      <td>2418.35</td>\n",
              "      <td>756.82</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>50798.00000</td>\n",
              "      <td>17383.000000</td>\n",
              "      <td>5440.000000</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>2.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>7477.43</td>\n",
              "      <td>3995.12</td>\n",
              "      <td>1989.81</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>41994.00000</td>\n",
              "      <td>22437.000000</td>\n",
              "      <td>11175.000000</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>2.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>6302.86</td>\n",
              "      <td>958.30</td>\n",
              "      <td>266.66</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>525458.00000</td>\n",
              "      <td>79892.000000</td>\n",
              "      <td>22231.000000</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 18 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       indicator  phase  ...  sentiment__negative sentiment__positive\n",
              "0            1.0    1.0  ...                500.0                50.0\n",
              "1            1.0    1.0  ...                500.0                50.0\n",
              "2            1.0    1.0  ...                500.0                50.0\n",
              "3            1.0    1.0  ...                500.0                50.0\n",
              "4            1.0    1.0  ...                500.0                50.0\n",
              "...          ...    ...  ...                  ...                 ...\n",
              "35779        2.0    5.0  ...                501.0                49.0\n",
              "35780        2.0    5.0  ...                501.0                49.0\n",
              "35781        2.0    5.0  ...                501.0                49.0\n",
              "35782        2.0    5.0  ...                501.0                49.0\n",
              "35783        2.0    5.0  ...                501.0                49.0\n",
              "\n",
              "[35784 rows x 18 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 444
        },
        "id": "uhBHB_wBtvn6",
        "outputId": "500c40dd-a16b-422e-fe22-fec186c2d02a"
      },
      "source": [
        "ny = ny[['phase', 'time_period', 'value', 'lowci', 'highci', 'age', 'CASE_RATE', 'HOSPITALIZED_RATE','DEATH_RATE','CASE_COUNT','HOSPITALIZED_COUNT','DEATH_COUNT','indicator']]\n",
        "ny"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>phase</th>\n",
              "      <th>time_period</th>\n",
              "      <th>value</th>\n",
              "      <th>lowci</th>\n",
              "      <th>highci</th>\n",
              "      <th>age</th>\n",
              "      <th>CASE_RATE</th>\n",
              "      <th>HOSPITALIZED_RATE</th>\n",
              "      <th>DEATH_RATE</th>\n",
              "      <th>CASE_COUNT</th>\n",
              "      <th>HOSPITALIZED_COUNT</th>\n",
              "      <th>DEATH_COUNT</th>\n",
              "      <th>indicator</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>171.46</td>\n",
              "      <td>13.55</td>\n",
              "      <td>0.23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>1554.45</td>\n",
              "      <td>153.69</td>\n",
              "      <td>12.88</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2529.03</td>\n",
              "      <td>630.34</td>\n",
              "      <td>113.37</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2552.64</td>\n",
              "      <td>1192.50</td>\n",
              "      <td>364.17</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2976.74</td>\n",
              "      <td>1830.07</td>\n",
              "      <td>909.18</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</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",
              "      <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>35779</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>20207.72</td>\n",
              "      <td>1160.60</td>\n",
              "      <td>67.68</td>\n",
              "      <td>224684.00000</td>\n",
              "      <td>13326.000000</td>\n",
              "      <td>794.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>15941.19</td>\n",
              "      <td>2540.10</td>\n",
              "      <td>476.51</td>\n",
              "      <td>161559.00000</td>\n",
              "      <td>25652.000000</td>\n",
              "      <td>4800.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>7067.11</td>\n",
              "      <td>2418.35</td>\n",
              "      <td>756.82</td>\n",
              "      <td>50798.00000</td>\n",
              "      <td>17383.000000</td>\n",
              "      <td>5440.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>7477.43</td>\n",
              "      <td>3995.12</td>\n",
              "      <td>1989.81</td>\n",
              "      <td>41994.00000</td>\n",
              "      <td>22437.000000</td>\n",
              "      <td>11175.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>6302.86</td>\n",
              "      <td>958.30</td>\n",
              "      <td>266.66</td>\n",
              "      <td>525458.00000</td>\n",
              "      <td>79892.000000</td>\n",
              "      <td>22231.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 13 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       phase  time_period  value  ...  HOSPITALIZED_COUNT   DEATH_COUNT  indicator\n",
              "0        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "1        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "2        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "3        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "4        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "...      ...          ...    ...  ...                 ...           ...        ...\n",
              "35779    5.0         23.0   35.8  ...        13326.000000    794.000000        2.0\n",
              "35780    5.0         23.0   35.8  ...        25652.000000   4800.000000        2.0\n",
              "35781    5.0         23.0   35.8  ...        17383.000000   5440.000000        2.0\n",
              "35782    5.0         23.0   35.8  ...        22437.000000  11175.000000        2.0\n",
              "35783    5.0         23.0   35.8  ...        79892.000000  22231.000000        2.0\n",
              "\n",
              "[35784 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tpuetoYZtvn-"
      },
      "source": [
        "names = ['Symptoms of Anxiety Disorder', 'Symptoms of Depression Disorder', 'Symptoms of Anxiety or Depression Disorder' ]\n",
        "feature_names = ['phase', 'time_period', 'value', 'lowci', 'highci', 'age', 'CASE_RATE', 'HOSPITALIZED_RATE', 'DEATH_RATE', 'CASE_COUNT', 'HOSPITALIZED_COUNT', 'DEATH_COUNT']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xkZ7ifpXtvn-",
        "outputId": "3ca446ce-2cc1-4ca5-a0e8-bb8bbada7250"
      },
      "source": [
        "dataset = ny.values\n",
        "dataset"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1.00000000e+00, 1.00000000e+00, 2.87000000e+01, ...,\n",
              "        1.99421750e+04, 6.75480913e+03, 1.00000000e+00],\n",
              "       [1.00000000e+00, 1.00000000e+00, 2.87000000e+01, ...,\n",
              "        1.99421750e+04, 6.75480913e+03, 1.00000000e+00],\n",
              "       [1.00000000e+00, 1.00000000e+00, 2.87000000e+01, ...,\n",
              "        1.99421750e+04, 6.75480913e+03, 1.00000000e+00],\n",
              "       ...,\n",
              "       [5.00000000e+00, 2.30000000e+01, 3.58000000e+01, ...,\n",
              "        1.73830000e+04, 5.44000000e+03, 2.00000000e+00],\n",
              "       [5.00000000e+00, 2.30000000e+01, 3.58000000e+01, ...,\n",
              "        2.24370000e+04, 1.11750000e+04, 2.00000000e+00],\n",
              "       [5.00000000e+00, 2.30000000e+01, 3.58000000e+01, ...,\n",
              "        7.98920000e+04, 2.22310000e+04, 2.00000000e+00]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "17n-xrZctvn_"
      },
      "source": [
        "X = dataset[:, :-1]\n",
        "y = dataset[:,-1]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ljPcv-WAtvoA"
      },
      "source": [
        "import pandas as pd\n",
        "from sklearn import preprocessing\n",
        "\n",
        "min_max_scaler = preprocessing.MinMaxScaler()\n",
        "x_scaled = min_max_scaler.fit_transform(X)\n",
        "y_scaled = min_max_scaler.fit_transform(y.reshape(-1, 1))\n",
        "X = pd.DataFrame(x_scaled)\n",
        "y = pd.DataFrame(y_scaled)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "3tEVcn7JtvoD",
        "outputId": "7cbf37bd-34fc-47d1-b653-ccfe1e1c5bd0"
      },
      "source": [
        "X"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "      <th>11</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.340102</td>\n",
              "      <td>0.294737</td>\n",
              "      <td>0.372642</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.164543</td>\n",
              "      <td>0.245592</td>\n",
              "      <td>0.303533</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.340102</td>\n",
              "      <td>0.294737</td>\n",
              "      <td>0.372642</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.069024</td>\n",
              "      <td>0.035197</td>\n",
              "      <td>0.006358</td>\n",
              "      <td>0.164543</td>\n",
              "      <td>0.245592</td>\n",
              "      <td>0.303533</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.340102</td>\n",
              "      <td>0.294737</td>\n",
              "      <td>0.372642</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.117665</td>\n",
              "      <td>0.154911</td>\n",
              "      <td>0.056866</td>\n",
              "      <td>0.164543</td>\n",
              "      <td>0.245592</td>\n",
              "      <td>0.303533</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.340102</td>\n",
              "      <td>0.294737</td>\n",
              "      <td>0.372642</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.118844</td>\n",
              "      <td>0.296102</td>\n",
              "      <td>0.182923</td>\n",
              "      <td>0.164543</td>\n",
              "      <td>0.245592</td>\n",
              "      <td>0.303533</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.340102</td>\n",
              "      <td>0.294737</td>\n",
              "      <td>0.372642</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.140010</td>\n",
              "      <td>0.456232</td>\n",
              "      <td>0.456855</td>\n",
              "      <td>0.164543</td>\n",
              "      <td>0.245592</td>\n",
              "      <td>0.303533</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35779</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.700508</td>\n",
              "      <td>0.684211</td>\n",
              "      <td>0.683962</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.288090</td>\n",
              "      <td>0.033902</td>\n",
              "      <td>0.422109</td>\n",
              "      <td>0.162334</td>\n",
              "      <td>0.035282</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.700508</td>\n",
              "      <td>0.684211</td>\n",
              "      <td>0.683962</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.787060</td>\n",
              "      <td>0.634561</td>\n",
              "      <td>0.239387</td>\n",
              "      <td>0.300823</td>\n",
              "      <td>0.317444</td>\n",
              "      <td>0.215562</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.700508</td>\n",
              "      <td>0.684211</td>\n",
              "      <td>0.683962</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.344159</td>\n",
              "      <td>0.603983</td>\n",
              "      <td>0.380276</td>\n",
              "      <td>0.088013</td>\n",
              "      <td>0.213387</td>\n",
              "      <td>0.244363</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.700508</td>\n",
              "      <td>0.684211</td>\n",
              "      <td>0.683962</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.364637</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.071098</td>\n",
              "      <td>0.276986</td>\n",
              "      <td>0.502453</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.700508</td>\n",
              "      <td>0.684211</td>\n",
              "      <td>0.683962</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.306015</td>\n",
              "      <td>0.237281</td>\n",
              "      <td>0.133913</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 12 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "        0    1         2         3   ...        8         9         10        11\n",
              "0      0.0  0.0  0.340102  0.294737  ...  0.000000  0.164543  0.245592  0.303533\n",
              "1      0.0  0.0  0.340102  0.294737  ...  0.006358  0.164543  0.245592  0.303533\n",
              "2      0.0  0.0  0.340102  0.294737  ...  0.056866  0.164543  0.245592  0.303533\n",
              "3      0.0  0.0  0.340102  0.294737  ...  0.182923  0.164543  0.245592  0.303533\n",
              "4      0.0  0.0  0.340102  0.294737  ...  0.456855  0.164543  0.245592  0.303533\n",
              "...    ...  ...       ...       ...  ...       ...       ...       ...       ...\n",
              "35779  1.0  1.0  0.700508  0.684211  ...  0.033902  0.422109  0.162334  0.035282\n",
              "35780  1.0  1.0  0.700508  0.684211  ...  0.239387  0.300823  0.317444  0.215562\n",
              "35781  1.0  1.0  0.700508  0.684211  ...  0.380276  0.088013  0.213387  0.244363\n",
              "35782  1.0  1.0  0.700508  0.684211  ...  1.000000  0.071098  0.276986  0.502453\n",
              "35783  1.0  1.0  0.700508  0.684211  ...  0.133913  1.000000  1.000000  1.000000\n",
              "\n",
              "[35784 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "H7AZFbTQtvoD",
        "outputId": "c0d4678d-dd43-4f00-d7fe-b7536056d2c9"
      },
      "source": [
        "y"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35779</th>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 1 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "         0\n",
              "0      0.5\n",
              "1      0.5\n",
              "2      0.5\n",
              "3      0.5\n",
              "4      0.5\n",
              "...    ...\n",
              "35779  1.0\n",
              "35780  1.0\n",
              "35781  1.0\n",
              "35782  1.0\n",
              "35783  1.0\n",
              "\n",
              "[35784 rows x 1 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ax5RitlRtvoE"
      },
      "source": [
        "# Helper\n",
        "\n",
        "def load_dataset(filename):\n",
        "    # load the dataset as a pandas DataFrame\n",
        "    data = pd.read_csv(filename, header=None)\n",
        "    # retrieve numpy array\n",
        "    dataset = data.values\n",
        "    # split into input (X) and output (y) variables\n",
        "    X = dataset[:, :-1]\n",
        "    y = dataset[:,-1]\n",
        "    return X, y"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "cUhooIrjtvoE"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import OneHotEncoder, StandardScaler\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5, random_state=42)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "q2hb_F83tvoE"
      },
      "source": [
        "from pandas import read_csv\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.feature_selection import SelectKBest\n",
        "from sklearn.feature_selection import f_classif\n",
        "from matplotlib import pyplot"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 485
        },
        "id": "Ia1UvrdAtvoE",
        "outputId": "0ed44689-774a-4c76-bf42-93fedf57211a"
      },
      "source": [
        "def select_features(X_train, y_train, X_test):\n",
        "    # configure to select all features\n",
        "    fs = SelectKBest(score_func=f_classif, k='all')\n",
        "    # learn relationship from training data\n",
        "    fs.fit(X_train, y_train)\n",
        "    # transform train input data\n",
        "    X_train_fs = fs.transform(X_train)\n",
        "    # transform test input data\n",
        "    X_test_fs = fs.transform(X_test)\n",
        "    return X_train_fs, X_test_fs, fs\n",
        " \n",
        "\"\"\"# load the dataset\n",
        "X, y = load_dataset('pima-indians-diabetes.csv')\n",
        "# split into train and test sets\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=1)\n",
        "\"\"\"\n",
        "# feature selection\n",
        "X_train_fs, X_test_fs, fs = select_features(X_train, y_train, X_test)\n",
        "# what are scores for the features\n",
        "for i in range(len(fs.scores_)):\n",
        "    print('Feature %d: %f' % (i, fs.scores_[i]))\n",
        "# plot the scores\n",
        "pyplot.bar([i for i in range(len(fs.scores_))], fs.scores_)\n",
        "pyplot.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Feature 0: 0.286601\n",
            "Feature 1: 0.089414\n",
            "Feature 2: 36075.900252\n",
            "Feature 3: 41365.241769\n",
            "Feature 4: 29433.792617\n",
            "Feature 5: 0.397847\n",
            "Feature 6: 0.004857\n",
            "Feature 7: 0.823220\n",
            "Feature 8: 0.631747\n",
            "Feature 9: 0.409344\n",
            "Feature 10: 0.906795\n",
            "Feature 11: 0.953940\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 485
        },
        "id": "MpIN4_netvoG",
        "outputId": "315f9797-ac4c-4ed1-9dfb-752a7ccadc3d"
      },
      "source": [
        "#feature selection \n",
        "\n",
        "def select_features(X_train, y_train, X_test):\n",
        "    # configure to select all features\n",
        "    fs = SelectKBest(score_func=f_classif, k='all')\n",
        "    # learn relationship from training data\n",
        "    fs.fit(X_train, y_train)\n",
        "    # transform train input data\n",
        "    X_train_fs = fs.transform(X_train)\n",
        "    # transform test input data\n",
        "    X_test_fs = fs.transform(X_test)\n",
        "    return X_train_fs, X_test_fs, fs\n",
        " \n",
        "\n",
        "X_train_fs, X_test_fs, fs = select_features(X_train, y_train, X_test)\n",
        "# what are scores for the features\n",
        "for i in range(len(fs.scores_)):\n",
        "    print('Feature %d: %f' % (i, fs.scores_[i]))\n",
        "# plot the scores\n",
        "pyplot.bar([i for i in range(len(fs.scores_))], fs.scores_)\n",
        "pyplot.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Feature 0: 0.286601\n",
            "Feature 1: 0.089414\n",
            "Feature 2: 36075.900252\n",
            "Feature 3: 41365.241769\n",
            "Feature 4: 29433.792617\n",
            "Feature 5: 0.397847\n",
            "Feature 6: 0.004857\n",
            "Feature 7: 0.823220\n",
            "Feature 8: 0.631747\n",
            "Feature 9: 0.409344\n",
            "Feature 10: 0.906795\n",
            "Feature 11: 0.953940\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "o9cefrNttvoG",
        "outputId": "d2c8f5bf-34b6-4d31-d12b-b982b06465ac"
      },
      "source": [
        "#https://www.dezyre.com/recipes/select-features-using-best-anova-f-values-in-python\n",
        "\n",
        "def ProjectPro_Ex_56():\n",
        "    print()\n",
        "    print('**How we can select features using best ANOVA F-values in Python**')\n",
        "\n",
        "    # loading libraries\n",
        "    from sklearn.datasets import load_iris\n",
        "    from sklearn.feature_selection import SelectKBest\n",
        "    from sklearn.feature_selection import f_classif\n",
        "\n",
        "    # Selecting Features With Best ANOVA F-Values\n",
        "    # Creating an SelectKBest object to select features with two best ANOVA F-Values\n",
        "    fvalue_Best = SelectKBest(f_classif, k='all')\n",
        "    # Applying the SelectKBest object to the features and target\n",
        "    X_kbest = fvalue_Best.fit_transform(X, y)\n",
        "    print(X_kbest)\n",
        "\n",
        "    # Showing results\n",
        "    print('Original number of features:', X.shape)\n",
        "    print('Reduced number of features:', X_kbest.shape)\n",
        "    return X_kbest\n",
        "\n",
        "result = ProjectPro_Ex_56()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\n",
            "**How we can select features using best ANOVA F-values in Python**\n",
            "[[0.         0.         0.34010152 ... 0.16454271 0.24559151 0.30353311]\n",
            " [0.         0.         0.34010152 ... 0.16454271 0.24559151 0.30353311]\n",
            " [0.         0.         0.34010152 ... 0.16454271 0.24559151 0.30353311]\n",
            " ...\n",
            " [1.         1.         0.70050761 ... 0.0880131  0.21338686 0.24436344]\n",
            " [1.         1.         0.70050761 ... 0.07109755 0.27698638 0.50245263]\n",
            " [1.         1.         0.70050761 ... 1.         1.         1.        ]]\n",
            "Original number of features: (35784, 12)\n",
            "Reduced number of features: (35784, 12)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ac3b5gxStvoH"
      },
      "source": [
        "#https://kgptalkie.com/feature-selection-based-on-univariate-anova-test-for-classification-machine-learning-kgp-talkie/\n",
        "\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Om2hK4IntvoI"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.feature_selection import VarianceThreshold\n",
        "from sklearn.feature_selection import f_classif, f_regression\n",
        "from sklearn.feature_selection import SelectKBest, SelectPercentile"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gJvj5a8KtvoI",
        "outputId": "dd1c023b-fac7-4369-cdee-ea17d1314f80"
      },
      "source": [
        "X.shape, y.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((35784, 12), (35784, 1))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JXkgAHb8tvoJ"
      },
      "source": [
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.5, random_state = 1, stratify = y)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aUqeVXjCtvoJ"
      },
      "source": [
        "constant_filter = VarianceThreshold(threshold=0.01)\n",
        "constant_filter.fit(X_train)\n",
        "X_train_filter = constant_filter.transform(X_train)\n",
        "X_test_filter = constant_filter.transform(X_test)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "j6UALVUvtvoJ",
        "outputId": "beca8930-2698-4910-9b03-a48c76054933"
      },
      "source": [
        "X_train_filter.shape, X_test_filter.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((17892, 12), (17892, 12))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nFvvpN23tvoK"
      },
      "source": [
        "X_train_T = X_train_filter.T\n",
        "X_test_T = X_test_filter.T"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FyR9-kX0tvoL"
      },
      "source": [
        "X_train_T = pd.DataFrame(X_train_T)\n",
        "X_test_T = pd.DataFrame(X_test_T)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YEAhALCitvoN",
        "outputId": "a46372d2-f99b-440c-db5d-2e688f3915fd"
      },
      "source": [
        "X_train_T.duplicated().sum()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mgRLU36CtvoO"
      },
      "source": [
        "duplicated_features = X_train_T.duplicated()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0eA2WHVVtvoO"
      },
      "source": [
        "features_to_keep = [not index for index in duplicated_features]\n",
        "X_train_unique = X_train_T[features_to_keep].T\n",
        "X_test_unique = X_test_T[features_to_keep].T"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VFQ63j6itvoO",
        "outputId": "70eb64d4-fedd-4625-9134-b5ee4a21c78e"
      },
      "source": [
        "X_train_unique.shape, X_train.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((17892, 12), (17892, 12))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zVIqGccKtvoP",
        "outputId": "1dc49b05-f591-40f1-e8bb-7861b2c33328"
      },
      "source": [
        "sel = f_classif(X_train_unique, y_train)\n",
        "sel"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([1.56579620e-01, 3.57054103e-01, 3.62278773e+04, 4.16040035e+04,\n",
              "        2.95416375e+04, 9.41215062e-01, 5.21075844e-01, 3.27348209e-01,\n",
              "        2.37387347e-01, 4.75595467e-02, 1.95450533e-01, 2.57441409e-01]),\n",
              " array([0.85506461, 0.69973963, 0.        , 0.        , 0.        ,\n",
              "        0.39017281, 0.59389029, 0.72083702, 0.78868822, 0.95355381,\n",
              "        0.82246578, 0.77302978]))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Mgkh7bO_tvoP"
      },
      "source": [
        "p_values = pd.Series(sel[1])\n",
        "p_values.index = X_train_unique.columns\n",
        "p_values.sort_values(ascending = True, inplace = True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 502
        },
        "id": "2z9pazKLtvoQ",
        "outputId": "a1430cc9-2382-4245-d34c-1cc2088f8f8e"
      },
      "source": [
        "p_values.plot.bar(figsize = (8, 8))\n",
        "plt.title('pvalues with respect to features')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "opaP-OxWtvoS"
      },
      "source": [
        "p_values = p_values[p_values<0.05]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vwImSRY4tvoT",
        "outputId": "1f593efb-0a36-4e5a-f354-3a0b9efba6aa"
      },
      "source": [
        "p_values.index"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Int64Index([2, 3, 4], dtype='int64')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jeSQbyBOtvoT"
      },
      "source": [
        "X_train_p = X_train_unique[p_values.index]\n",
        "X_test_p = X_test_unique[p_values.index]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lp2iOt4AtvoU",
        "outputId": "72ca48a2-bdd6-4527-eaf0-2d564916e577"
      },
      "source": [
        "#https://chrisalbon.com/machine_learning/feature_selection/anova_f-value_for_feature_selection/\n",
        "\n",
        "# Create an SelectKBest object to select features with six best ANOVA F-Values\n",
        "fvalue_selector = SelectKBest(f_classif, k=6)\n",
        "\n",
        "# Apply the SelectKBest object to the features and target\n",
        "X_kbest = fvalue_selector.fit_transform(X, y)\n",
        "\n",
        "# Show results\n",
        "print('Original number of features:', X.shape[1])\n",
        "print('Reduced number of features:', X_kbest.shape[1])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Original number of features: 12\n",
            "Reduced number of features: 6\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "px7tlRaPtvoU"
      },
      "source": [
        "# PCA Analysis Below"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ij2BWHp6tvoV"
      },
      "source": [
        "from numpy import mean\n",
        "from numpy import std\n",
        "import pandas as pd\n",
        "from sklearn.datasets import make_classification\n",
        "from sklearn.model_selection import cross_val_score\n",
        "from sklearn.model_selection import RepeatedStratifiedKFold\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.decomposition import PCA\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from matplotlib import pyplot"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "lSUDlopXtvoV",
        "outputId": "f5fe917a-c2aa-4ce5-b573-37135205dccf"
      },
      "source": [
        "ny = pd.read_csv('./new_york.csv')\n",
        "ny"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>indicator</th>\n",
              "      <th>group</th>\n",
              "      <th>state</th>\n",
              "      <th>subgroup</th>\n",
              "      <th>phase</th>\n",
              "      <th>time_period</th>\n",
              "      <th>time_period_label</th>\n",
              "      <th>time_period_start_date</th>\n",
              "      <th>time_period_end_date</th>\n",
              "      <th>value</th>\n",
              "      <th>lowci</th>\n",
              "      <th>highci</th>\n",
              "      <th>confidence_interval</th>\n",
              "      <th>quartile_range</th>\n",
              "      <th>age</th>\n",
              "      <th>CASE_RATE</th>\n",
              "      <th>HOSPITALIZED_RATE</th>\n",
              "      <th>DEATH_RATE</th>\n",
              "      <th>date</th>\n",
              "      <th>CASE_COUNT</th>\n",
              "      <th>HOSPITALIZED_COUNT</th>\n",
              "      <th>DEATH_COUNT</th>\n",
              "      <th>counts</th>\n",
              "      <th>sentiment__negative</th>\n",
              "      <th>sentiment__positive</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>171.46</td>\n",
              "      <td>13.55</td>\n",
              "      <td>0.23</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>1554.45</td>\n",
              "      <td>153.69</td>\n",
              "      <td>12.88</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>2529.03</td>\n",
              "      <td>630.34</td>\n",
              "      <td>113.37</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>2552.64</td>\n",
              "      <td>1192.50</td>\n",
              "      <td>364.17</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Symptoms of Depressive Disorder</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>Apr 23 - May 5</td>\n",
              "      <td>2020-04-23T00:00:00.000</td>\n",
              "      <td>2020-05-05T00:00:00.000</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>24.6 - 33.1</td>\n",
              "      <td>24.1 - 28.7</td>\n",
              "      <td>18 - 29 years</td>\n",
              "      <td>2976.74</td>\n",
              "      <td>1830.07</td>\n",
              "      <td>909.18</td>\n",
              "      <td>2020-04-23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>295066.0</td>\n",
              "      <td>500.0</td>\n",
              "      <td>50.0</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",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35779</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>20207.72</td>\n",
              "      <td>1160.60</td>\n",
              "      <td>67.68</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>224684.00000</td>\n",
              "      <td>13326.000000</td>\n",
              "      <td>794.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>15941.19</td>\n",
              "      <td>2540.10</td>\n",
              "      <td>476.51</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>161559.00000</td>\n",
              "      <td>25652.000000</td>\n",
              "      <td>4800.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>7067.11</td>\n",
              "      <td>2418.35</td>\n",
              "      <td>756.82</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>50798.00000</td>\n",
              "      <td>17383.000000</td>\n",
              "      <td>5440.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>7477.43</td>\n",
              "      <td>3995.12</td>\n",
              "      <td>1989.81</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>41994.00000</td>\n",
              "      <td>22437.000000</td>\n",
              "      <td>11175.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>Symptoms of Anxiety Disorder or Depressive Dis...</td>\n",
              "      <td>By State</td>\n",
              "      <td>New York</td>\n",
              "      <td>New York</td>\n",
              "      <td>3 (Jan 6 � Feb 1)</td>\n",
              "      <td>23.0</td>\n",
              "      <td>Jan 20 - Feb 1</td>\n",
              "      <td>2021-01-20T00:00:00.000</td>\n",
              "      <td>2021-02-01T00:00:00.000</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>32.0 - 39.7</td>\n",
              "      <td>28.8-37.6</td>\n",
              "      <td>80 years and above</td>\n",
              "      <td>6302.86</td>\n",
              "      <td>958.30</td>\n",
              "      <td>266.66</td>\n",
              "      <td>2021-02-01</td>\n",
              "      <td>525458.00000</td>\n",
              "      <td>79892.000000</td>\n",
              "      <td>22231.000000</td>\n",
              "      <td>419169.0</td>\n",
              "      <td>501.0</td>\n",
              "      <td>49.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 25 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                               indicator  ... sentiment__positive\n",
              "0                        Symptoms of Depressive Disorder  ...                50.0\n",
              "1                        Symptoms of Depressive Disorder  ...                50.0\n",
              "2                        Symptoms of Depressive Disorder  ...                50.0\n",
              "3                        Symptoms of Depressive Disorder  ...                50.0\n",
              "4                        Symptoms of Depressive Disorder  ...                50.0\n",
              "...                                                  ...  ...                 ...\n",
              "35779  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "35780  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "35781  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "35782  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "35783  Symptoms of Anxiety Disorder or Depressive Dis...  ...                49.0\n",
              "\n",
              "[35784 rows x 25 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DenM70wutvoW",
        "outputId": "dfbcd7b8-feb5-449b-ba6b-5203a15369c7"
      },
      "source": [
        "ny[\"indicator\"].replace({\"Symptoms of Anxiety Disorder\": \"0\", \"Symptoms of Depressive Disorder\": \"1\", \"Symptoms of Anxiety Disorder or Depressive Disorder\": \"2\"}, inplace=True)\n",
        "ny[\"age\"].replace({\"18 - 29 years\": \"29\", \"30 - 39 years\": \"39\", \"40 - 49 years\": \"49\", \"50 - 59 years\":\"59\",\"60 - 69 years\":\"69\",\"70 - 79 years\":\"79\",\"80 years and above\":\"80\"}, inplace=True)\n",
        "\n",
        "data_to_use = ny\n",
        "del data_to_use['group']\n",
        "del data_to_use['state']\n",
        "del data_to_use['subgroup']\n",
        "del data_to_use['time_period_label']\n",
        "del data_to_use['confidence_interval']\n",
        "del data_to_use['quartile_range']\n",
        "del data_to_use['counts']\n",
        "del data_to_use['sentiment__negative']\n",
        "del data_to_use['sentiment__positive']\n",
        "\n",
        "\n",
        "ny = data_to_use\n",
        "ny.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 35784 entries, 0 to 35783\n",
            "Data columns (total 16 columns):\n",
            " #   Column                  Non-Null Count  Dtype  \n",
            "---  ------                  --------------  -----  \n",
            " 0   indicator               35784 non-null  object \n",
            " 1   phase                   35784 non-null  object \n",
            " 2   time_period             35784 non-null  float64\n",
            " 3   time_period_start_date  35784 non-null  object \n",
            " 4   time_period_end_date    35784 non-null  object \n",
            " 5   value                   35784 non-null  float64\n",
            " 6   lowci                   35784 non-null  float64\n",
            " 7   highci                  35784 non-null  float64\n",
            " 8   age                     35784 non-null  object \n",
            " 9   CASE_RATE               35784 non-null  float64\n",
            " 10  HOSPITALIZED_RATE       35784 non-null  float64\n",
            " 11  DEATH_RATE              33957 non-null  float64\n",
            " 12  date                    35784 non-null  object \n",
            " 13  CASE_COUNT              35784 non-null  float64\n",
            " 14  HOSPITALIZED_COUNT      35784 non-null  float64\n",
            " 15  DEATH_COUNT             35784 non-null  float64\n",
            "dtypes: float64(10), object(6)\n",
            "memory usage: 4.4+ MB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QTYJcvlKtvoX"
      },
      "source": [
        "ny['DEATH_RATE'].fillna((ny['DEATH_RATE'].mean()), inplace=True)\n",
        "#Add this cell after API change\n",
        "ny.loc[ny['phase'] == ('Oct 28 � Dec 21'), 'phase'] = 3\n",
        "ny.loc[ny['phase'] == ('3 (Oct 28 � Dec 21)'), 'phase'] = 4\n",
        "ny.loc[ny['phase'] == ('3 (Jan 6 � Feb 1)'), 'phase'] = 5\n",
        "ny.rename(columns = {'counts':'twitter_count'}, inplace = True)\n",
        "\n",
        "from sklearn import preprocessing\n",
        "\n",
        "ny[\"age\"] = ny['age'].astype(float)\n",
        "ny[\"phase\"] = ny['phase'].astype(float)\n",
        "ny[\"indicator\"] = ny['indicator'].astype(float)\n",
        "ny[\"time_period\"] = ny['time_period'].astype(float)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 444
        },
        "id": "xWFTzb2_tvoX",
        "outputId": "7a6a7844-f63b-442c-ee02-10ba3d30bdc1"
      },
      "source": [
        "ny = ny[['phase', 'time_period', 'value', 'lowci', 'highci', 'age', 'CASE_RATE', 'HOSPITALIZED_RATE','DEATH_RATE','CASE_COUNT','HOSPITALIZED_COUNT','DEATH_COUNT','indicator']]\n",
        "ny"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>phase</th>\n",
              "      <th>time_period</th>\n",
              "      <th>value</th>\n",
              "      <th>lowci</th>\n",
              "      <th>highci</th>\n",
              "      <th>age</th>\n",
              "      <th>CASE_RATE</th>\n",
              "      <th>HOSPITALIZED_RATE</th>\n",
              "      <th>DEATH_RATE</th>\n",
              "      <th>CASE_COUNT</th>\n",
              "      <th>HOSPITALIZED_COUNT</th>\n",
              "      <th>DEATH_COUNT</th>\n",
              "      <th>indicator</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>171.46</td>\n",
              "      <td>13.55</td>\n",
              "      <td>0.23</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>1554.45</td>\n",
              "      <td>153.69</td>\n",
              "      <td>12.88</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2529.03</td>\n",
              "      <td>630.34</td>\n",
              "      <td>113.37</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2552.64</td>\n",
              "      <td>1192.50</td>\n",
              "      <td>364.17</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>28.7</td>\n",
              "      <td>24.6</td>\n",
              "      <td>33.1</td>\n",
              "      <td>29.0</td>\n",
              "      <td>2976.74</td>\n",
              "      <td>1830.07</td>\n",
              "      <td>909.18</td>\n",
              "      <td>90629.21686</td>\n",
              "      <td>19942.175032</td>\n",
              "      <td>6754.809134</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",
              "      <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>35779</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>20207.72</td>\n",
              "      <td>1160.60</td>\n",
              "      <td>67.68</td>\n",
              "      <td>224684.00000</td>\n",
              "      <td>13326.000000</td>\n",
              "      <td>794.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>15941.19</td>\n",
              "      <td>2540.10</td>\n",
              "      <td>476.51</td>\n",
              "      <td>161559.00000</td>\n",
              "      <td>25652.000000</td>\n",
              "      <td>4800.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>7067.11</td>\n",
              "      <td>2418.35</td>\n",
              "      <td>756.82</td>\n",
              "      <td>50798.00000</td>\n",
              "      <td>17383.000000</td>\n",
              "      <td>5440.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>7477.43</td>\n",
              "      <td>3995.12</td>\n",
              "      <td>1989.81</td>\n",
              "      <td>41994.00000</td>\n",
              "      <td>22437.000000</td>\n",
              "      <td>11175.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>5.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>35.8</td>\n",
              "      <td>32.0</td>\n",
              "      <td>39.7</td>\n",
              "      <td>80.0</td>\n",
              "      <td>6302.86</td>\n",
              "      <td>958.30</td>\n",
              "      <td>266.66</td>\n",
              "      <td>525458.00000</td>\n",
              "      <td>79892.000000</td>\n",
              "      <td>22231.000000</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 13 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "       phase  time_period  value  ...  HOSPITALIZED_COUNT   DEATH_COUNT  indicator\n",
              "0        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "1        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "2        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "3        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "4        1.0          1.0   28.7  ...        19942.175032   6754.809134        1.0\n",
              "...      ...          ...    ...  ...                 ...           ...        ...\n",
              "35779    5.0         23.0   35.8  ...        13326.000000    794.000000        2.0\n",
              "35780    5.0         23.0   35.8  ...        25652.000000   4800.000000        2.0\n",
              "35781    5.0         23.0   35.8  ...        17383.000000   5440.000000        2.0\n",
              "35782    5.0         23.0   35.8  ...        22437.000000  11175.000000        2.0\n",
              "35783    5.0         23.0   35.8  ...        79892.000000  22231.000000        2.0\n",
              "\n",
              "[35784 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vrN2RUzotvoY"
      },
      "source": [
        "values = ny.values\n",
        " \n",
        "Y = values[:,12]\n",
        "X = values[:,0:11]\n",
        "\n",
        "import pandas as pd\n",
        "from sklearn import preprocessing\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "min_max_scaler = preprocessing.MinMaxScaler()\n",
        "#x = StandardScaler().fit_transform(x)\n",
        "x_scaled = StandardScaler().fit_transform(X)\n",
        "X = pd.DataFrame(x_scaled)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "fxfF8C1StvoY",
        "outputId": "016f1212-0f4e-4bbe-f02d-3e4a5e90f476"
      },
      "source": [
        "X"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>0</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "      <th>4</th>\n",
              "      <th>5</th>\n",
              "      <th>6</th>\n",
              "      <th>7</th>\n",
              "      <th>8</th>\n",
              "      <th>9</th>\n",
              "      <th>10</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>-1.046623</td>\n",
              "      <td>-2.028335</td>\n",
              "      <td>-0.519765</td>\n",
              "      <td>-0.617653</td>\n",
              "      <td>-0.412523</td>\n",
              "      <td>-1.574064</td>\n",
              "      <td>-1.255775</td>\n",
              "      <td>-1.196508</td>\n",
              "      <td>-0.878629</td>\n",
              "      <td>-3.168163e-16</td>\n",
              "      <td>4.000775e-16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-1.046623</td>\n",
              "      <td>-2.028335</td>\n",
              "      <td>-0.519765</td>\n",
              "      <td>-0.617653</td>\n",
              "      <td>-0.412523</td>\n",
              "      <td>-1.574064</td>\n",
              "      <td>-0.837174</td>\n",
              "      <td>-1.056190</td>\n",
              "      <td>-0.856327</td>\n",
              "      <td>-3.168163e-16</td>\n",
              "      <td>4.000775e-16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>-1.046623</td>\n",
              "      <td>-2.028335</td>\n",
              "      <td>-0.519765</td>\n",
              "      <td>-0.617653</td>\n",
              "      <td>-0.412523</td>\n",
              "      <td>-1.574064</td>\n",
              "      <td>-0.542190</td>\n",
              "      <td>-0.578936</td>\n",
              "      <td>-0.679167</td>\n",
              "      <td>-3.168163e-16</td>\n",
              "      <td>4.000775e-16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-1.046623</td>\n",
              "      <td>-2.028335</td>\n",
              "      <td>-0.519765</td>\n",
              "      <td>-0.617653</td>\n",
              "      <td>-0.412523</td>\n",
              "      <td>-1.574064</td>\n",
              "      <td>-0.535043</td>\n",
              "      <td>-0.016063</td>\n",
              "      <td>-0.237016</td>\n",
              "      <td>-3.168163e-16</td>\n",
              "      <td>4.000775e-16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-1.046623</td>\n",
              "      <td>-2.028335</td>\n",
              "      <td>-0.519765</td>\n",
              "      <td>-0.617653</td>\n",
              "      <td>-0.412523</td>\n",
              "      <td>-1.574064</td>\n",
              "      <td>-0.406678</td>\n",
              "      <td>0.622315</td>\n",
              "      <td>0.723818</td>\n",
              "      <td>-3.168163e-16</td>\n",
              "      <td>4.000775e-16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35779</th>\n",
              "      <td>1.655567</td>\n",
              "      <td>1.392188</td>\n",
              "      <td>0.725645</td>\n",
              "      <td>0.752476</td>\n",
              "      <td>0.685975</td>\n",
              "      <td>1.221661</td>\n",
              "      <td>4.808764</td>\n",
              "      <td>-0.048004</td>\n",
              "      <td>-0.759717</td>\n",
              "      <td>1.459284e+00</td>\n",
              "      <td>-3.637985e-01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>1.655567</td>\n",
              "      <td>1.392188</td>\n",
              "      <td>0.725645</td>\n",
              "      <td>0.752476</td>\n",
              "      <td>0.685975</td>\n",
              "      <td>1.221661</td>\n",
              "      <td>3.517378</td>\n",
              "      <td>1.333245</td>\n",
              "      <td>-0.038964</td>\n",
              "      <td>7.721221e-01</td>\n",
              "      <td>3.139617e-01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>1.655567</td>\n",
              "      <td>1.392188</td>\n",
              "      <td>0.725645</td>\n",
              "      <td>0.752476</td>\n",
              "      <td>0.685975</td>\n",
              "      <td>1.221661</td>\n",
              "      <td>0.831388</td>\n",
              "      <td>1.211341</td>\n",
              "      <td>0.455212</td>\n",
              "      <td>-4.335917e-01</td>\n",
              "      <td>-1.407194e-01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>1.655567</td>\n",
              "      <td>1.392188</td>\n",
              "      <td>0.725645</td>\n",
              "      <td>0.752476</td>\n",
              "      <td>0.685975</td>\n",
              "      <td>1.221661</td>\n",
              "      <td>0.955583</td>\n",
              "      <td>2.790110</td>\n",
              "      <td>2.628930</td>\n",
              "      <td>-5.294296e-01</td>\n",
              "      <td>1.371810e-01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>1.655567</td>\n",
              "      <td>1.392188</td>\n",
              "      <td>0.725645</td>\n",
              "      <td>0.752476</td>\n",
              "      <td>0.685975</td>\n",
              "      <td>1.221661</td>\n",
              "      <td>0.600066</td>\n",
              "      <td>-0.250560</td>\n",
              "      <td>-0.408922</td>\n",
              "      <td>4.733427e+00</td>\n",
              "      <td>3.296415e+00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 11 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "             0         1         2   ...        8             9             10\n",
              "0     -1.046623 -2.028335 -0.519765  ... -0.878629 -3.168163e-16  4.000775e-16\n",
              "1     -1.046623 -2.028335 -0.519765  ... -0.856327 -3.168163e-16  4.000775e-16\n",
              "2     -1.046623 -2.028335 -0.519765  ... -0.679167 -3.168163e-16  4.000775e-16\n",
              "3     -1.046623 -2.028335 -0.519765  ... -0.237016 -3.168163e-16  4.000775e-16\n",
              "4     -1.046623 -2.028335 -0.519765  ...  0.723818 -3.168163e-16  4.000775e-16\n",
              "...         ...       ...       ...  ...       ...           ...           ...\n",
              "35779  1.655567  1.392188  0.725645  ... -0.759717  1.459284e+00 -3.637985e-01\n",
              "35780  1.655567  1.392188  0.725645  ... -0.038964  7.721221e-01  3.139617e-01\n",
              "35781  1.655567  1.392188  0.725645  ...  0.455212 -4.335917e-01 -1.407194e-01\n",
              "35782  1.655567  1.392188  0.725645  ...  2.628930 -5.294296e-01  1.371810e-01\n",
              "35783  1.655567  1.392188  0.725645  ... -0.408922  4.733427e+00  3.296415e+00\n",
              "\n",
              "[35784 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "U70oban6l1VX"
      },
      "source": [
        "cluster_df = X"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WnHNYX8etvoZ"
      },
      "source": [
        "from sklearn.decomposition import PCA\n",
        "\n",
        "pca = PCA(n_components=2)\n",
        "principalComponents = pca.fit_transform(X)\n",
        "\n",
        "principalDf = pd.DataFrame(data = principalComponents\n",
        "             , columns = ['PC1', 'PC2'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VAZNjj7AtvoZ"
      },
      "source": [
        "finalDf = pd.concat([principalDf, ny[['indicator']]], axis = 1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "3pr7Ghpetvoa",
        "outputId": "022ea412-cdd3-49b4-8516-142090205791"
      },
      "source": [
        "finalDf"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>PC1</th>\n",
              "      <th>PC2</th>\n",
              "      <th>indicator</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>-2.280133</td>\n",
              "      <td>-1.959285</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-2.141880</td>\n",
              "      <td>-1.739945</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>-1.956870</td>\n",
              "      <td>-1.399967</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-1.798810</td>\n",
              "      <td>-1.061538</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-1.528919</td>\n",
              "      <td>-0.495138</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35779</th>\n",
              "      <td>3.202578</td>\n",
              "      <td>2.257716</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>3.181432</td>\n",
              "      <td>2.476547</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>2.369157</td>\n",
              "      <td>1.441021</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>2.970185</td>\n",
              "      <td>2.718918</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>2.662399</td>\n",
              "      <td>1.086144</td>\n",
              "      <td>2.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 3 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "            PC1       PC2  indicator\n",
              "0     -2.280133 -1.959285        1.0\n",
              "1     -2.141880 -1.739945        1.0\n",
              "2     -1.956870 -1.399967        1.0\n",
              "3     -1.798810 -1.061538        1.0\n",
              "4     -1.528919 -0.495138        1.0\n",
              "...         ...       ...        ...\n",
              "35779  3.202578  2.257716        2.0\n",
              "35780  3.181432  2.476547        2.0\n",
              "35781  2.369157  1.441021        2.0\n",
              "35782  2.970185  2.718918        2.0\n",
              "35783  2.662399  1.086144        2.0\n",
              "\n",
              "[35784 rows x 3 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 246
        },
        "id": "Cyceu0Mktvoa",
        "outputId": "7a9112d4-6c93-4d22-942d-a0d0c9021878"
      },
      "source": [
        "plt.figure(figsize=(8,6))\n",
        "targets = ['Anxiety', 'Depression', 'Anxiety or Depression']\n",
        "plt.scatter(finalDf.iloc[:,0],finalDf.iloc[:,1],\n",
        "            c=finalDf['indicator'],\n",
        "            cmap='rainbow'\n",
        "           )\n",
        "plt.xlabel('First principal component')\n",
        "plt.ylabel('Second Principal Component')\n",
        "pass;"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-12-345d69eb148d>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfigure\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfigsize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m8\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m6\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mtargets\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m'Anxiety'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'Depression'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'Anxiety or Depression'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m plt.scatter(finalDf.iloc[:,0],finalDf.iloc[:,1],\n\u001b[1;32m      4\u001b[0m             \u001b[0mc\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mfinalDf\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'indicator'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m             \u001b[0mcmap\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'rainbow'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'plt' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 246
        },
        "id": "TxWWzcDWtvoc",
        "outputId": "65d2da94-2700-462f-edff-e379a91d162d"
      },
      "source": [
        "import seaborn as sns\n",
        "plt.figure(figsize=(8,6))\n",
        "sns.scatterplot(\n",
        "    x=\"PC1\", y=\"PC2\",\n",
        "    hue=\"indicator\",\n",
        "    palette=sns.color_palette(\"hls\", 3),\n",
        "    data=finalDf,\n",
        "    legend=\"full\",\n",
        "    alpha=0.3\n",
        ")"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-13-1f9fd2d93c95>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mseaborn\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0msns\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfigure\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfigsize\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m8\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m6\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      3\u001b[0m sns.scatterplot(\n\u001b[1;32m      4\u001b[0m     \u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"PC1\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"PC2\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m     \u001b[0mhue\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"indicator\"\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'plt' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LYTAWb_otvod"
      },
      "source": [
        "pca.explained_variance_ratio_"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 246
        },
        "id": "ymUCklCRtvoe",
        "outputId": "e5a2fd15-d4f9-4001-a3bc-8c66b3134235"
      },
      "source": [
        "import plotly.express as px\n",
        "from sklearn.decomposition import PCA\n",
        "\n",
        "\n",
        "pca = PCA()\n",
        "components = pca.fit_transform(ny[feature_names])\n",
        "labels = {\n",
        "    str(i): f\"PC {i+1} ({var:.1f}%)\"\n",
        "    for i, var in enumerate(pca.explained_variance_ratio_ * 100)\n",
        "}\n",
        "\n",
        "fig = px.scatter_matrix(\n",
        "    components,\n",
        "    labels=labels,\n",
        "    dimensions=range(2),\n",
        "    color=ny[\"indicator\"]\n",
        ")\n",
        "fig.update_traces(diagonal_visible=False)\n",
        "fig.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-14-58f3a3838a7b>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mpca\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mPCA\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m \u001b[0mcomponents\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpca\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_transform\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mny\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mfeature_names\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      7\u001b[0m labels = {\n\u001b[1;32m      8\u001b[0m     \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;34mf\"PC {i+1} ({var:.1f}%)\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'feature_names' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "a3bbeFGdtvog",
        "outputId": "0718ff55-734d-4cfc-9a0a-ba9693e43035"
      },
      "source": [
        "\n",
        "values = finalDf.values\n",
        " \n",
        "Y = values[:,-1]\n",
        "X = values[:,0:2]\n",
        "Y"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1., 1., 1., ..., 2., 2., 2.])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jRdWvDOOl1M1",
        "outputId": "cbd27a35-b6d8-4b40-91d9-d0476bf20915"
      },
      "source": [
        "X"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[-2.28013281, -1.95928524],\n",
              "       [-2.14187953, -1.73994476],\n",
              "       [-1.95686961, -1.39996711],\n",
              "       ...,\n",
              "       [ 2.36915657,  1.44102143],\n",
              "       [ 2.97018547,  2.71891842],\n",
              "       [ 2.66239893,  1.08614425]])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t8XFT8w-tvoh"
      },
      "source": [
        "import pandas as pd\n",
        "from sklearn import metrics\n",
        "from sklearn.cluster import KMeans\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nZIaq-5Qtvoh"
      },
      "source": [
        "finalDf2 = finalDf\n",
        "finalDf3 = finalDf "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AvLSOfzqtvoi"
      },
      "source": [
        "from sklearn.cluster import KMeans\n",
        "kmeans = KMeans(n_clusters=3)\n",
        "kmeans.fit(X,Y)\n",
        "y_kmeans = kmeans.predict(X)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "jI0D2Lpqtvoi",
        "outputId": "edd8d13a-c85a-4897-9624-374398490f0c"
      },
      "source": [
        "plt.scatter(finalDf.iloc[:,0], finalDf.iloc[:,1], c=y_kmeans, s=50, cmap='gist_rainbow')\n",
        "\n",
        "centers = kmeans.cluster_centers_\n",
        "plt.scatter(centers[:, 0], centers[:, -1], c='blue', s=200, alpha=0.5);"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "s00qpsfHmWf8",
        "outputId": "725b7a4d-0d3f-40b9-fa50-993b6e0605be"
      },
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.metrics import classification_report\n",
        "from sklearn.metrics import confusion_matrix\n",
        "\n",
        "acc = accuracy_score(finalDf.iloc[:,2] , y_kmeans)\n",
        "print(\"Accuracy:\",accuracy_score(finalDf.iloc[:,2] , y_kmeans))\n",
        "print(classification_report(finalDf.iloc[:,2] , y_kmeans))\n",
        "print(confusion_matrix(finalDf.iloc[:,2] , y_kmeans))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy: 0.18916275430359938\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.43      0.53      0.47     11928\n",
            "         1.0       0.05      0.04      0.05     11928\n",
            "         2.0       0.00      0.00      0.00     11928\n",
            "\n",
            "    accuracy                           0.19     35784\n",
            "   macro avg       0.16      0.19      0.17     35784\n",
            "weighted avg       0.16      0.19      0.17     35784\n",
            "\n",
            "[[ 6279  4200  1449]\n",
            " [ 1288   490 10150]\n",
            " [ 6993  4935     0]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "26ylgXqYlrD1",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2e9d4f7f-b2d4-4c82-bdc3-d8c80e0c2a54"
      },
      "source": [
        "# K meansfrom sklearn.cluster import KMeans\n",
        "from sklearn.metrics import silhouette_score \n",
        "from sklearn.metrics import calinski_harabasz_score\n",
        "from sklearn.metrics import davies_bouldin_score\n",
        "# Fit K-Means\n",
        "kmeans_1 = KMeans(n_clusters=3,random_state= 42)\n",
        "# Use fit_predict to cluster the dataset\n",
        "predictions = kmeans_1.fit_predict(cluster_df)\n",
        "# Calculate cluster validation metrics\n",
        "score_kemans_s = silhouette_score(cluster_df, kmeans_1.labels_, metric='euclidean')\n",
        "score_kemans_c = calinski_harabasz_score(cluster_df, kmeans_1.labels_)\n",
        "score_kemans_d = davies_bouldin_score(cluster_df, predictions)\n",
        "print('Silhouette Score: %.4f' % score_kemans_s)\n",
        "print('Calinski Harabasz Score: %.4f' % score_kemans_c)\n",
        "print('Davies Bouldin Score: %.4f' % score_kemans_d)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Silhouette Score: 0.2023\n",
            "Calinski Harabasz Score: 8916.7504\n",
            "Davies Bouldin Score: 1.6260\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EPggK_-A0oHB"
      },
      "source": [
        "finalDf['cluster'] = y_kmeans\n",
        "finalDf['indicator'] = finalDf['indicator'] + 1\n",
        "finalDf['cluster'] = finalDf['cluster'] + 1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "t6IbK9ym0vJ0",
        "outputId": "5a81d2d8-ed8c-4299-c221-5caeca673fe3"
      },
      "source": [
        "import math\n",
        "def get_entropys(df_clustered, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3,1)):\n",
        "    p = [[round(df_clustered[(df_clustered[label_col_name]==label) & (df_clustered[cluster_col_name]==cluster)].shape[0]/\n",
        "                df_clustered[df_clustered[cluster_col_name]==cluster].shape[0],3) for label in cluster_values] for cluster in cluster_values]\n",
        "    p_log_p = [[p_i*math.log(p_i) for p_i in p_i_s if p_i!=0] for p_i_s in p]\n",
        "    e = [-round(sum(p_log_p_i),3) for p_log_p_i in p_log_p]\n",
        "    return e\n",
        "  \n",
        "entropys_5 = get_entropys(finalDf, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3,1))\n",
        "print(\"Clusters' Entropy\")\n",
        "print(\"Entropy: {0:.3f}\".format(min(entropys_5)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Clusters' Entropy\n",
            "Entropy: 0.377\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kLeIkAKqDPSo",
        "outputId": "c6dffb1f-5807-49ca-f6dd-0eee90c73c36"
      },
      "source": [
        "import numpy as np\n",
        "from sklearn import metrics\n",
        "\n",
        "def purity_score(y_true, y_pred):\n",
        "    # compute contingency matrix (also called confusion matrix)\n",
        "    contingency_matrix = metrics.cluster.contingency_matrix(y_true, y_pred)\n",
        "    # return purity\n",
        "    return np.sum(np.amax(contingency_matrix, axis=0)) / np.sum(contingency_matrix) \n",
        "\n",
        "purity = purity_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The purity is:\",purity)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The purity is: 0.61697965571205\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ovYHN6hLDUx7",
        "outputId": "332a603e-0c17-4772-dc53-d228182f30fd"
      },
      "source": [
        "from sklearn.metrics.cluster import v_measure_score\n",
        "v_measure = v_measure_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The v_measure is:\",v_measure)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The v_measure is: 0.33727263075389413\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pL2vZHVWDX_d",
        "outputId": "71ec5cf3-ac76-4d8b-a8d5-f75017d4961a"
      },
      "source": [
        "from sklearn.metrics.cluster import adjusted_rand_score\n",
        "randScore = adjusted_rand_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The rand_score is:\",randScore)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The rand_score is: 0.31730196619316403\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "epe_NuKRtvol"
      },
      "source": [
        "from sklearn.datasets import load_iris\n",
        "from sklearn.cluster import AgglomerativeClustering\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OEa1iB_Ntvom"
      },
      "source": [
        "df = finalDf\n",
        "labels_final = df['indicator'].values\n",
        "#df = df[:,0:1]\n",
        "#del df['indicator']\n",
        "df = df.drop(['indicator'], axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "jKNShkDjtvom",
        "outputId": "f1c36045-d997-4385-ead0-dc6dd17d8d04"
      },
      "source": [
        "df"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>PC1</th>\n",
              "      <th>PC2</th>\n",
              "      <th>cluster</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>-2.280133</td>\n",
              "      <td>-1.959285</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>-2.141880</td>\n",
              "      <td>-1.739945</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>-1.956870</td>\n",
              "      <td>-1.399967</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-1.798810</td>\n",
              "      <td>-1.061538</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-1.528919</td>\n",
              "      <td>-0.495138</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35779</th>\n",
              "      <td>3.202578</td>\n",
              "      <td>2.257716</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35780</th>\n",
              "      <td>3.181432</td>\n",
              "      <td>2.476547</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35781</th>\n",
              "      <td>2.369157</td>\n",
              "      <td>1.441021</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35782</th>\n",
              "      <td>2.970185</td>\n",
              "      <td>2.718918</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35783</th>\n",
              "      <td>2.662399</td>\n",
              "      <td>1.086144</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>35784 rows × 3 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "            PC1       PC2  cluster\n",
              "0     -2.280133 -1.959285        3\n",
              "1     -2.141880 -1.739945        3\n",
              "2     -1.956870 -1.399967        3\n",
              "3     -1.798810 -1.061538        3\n",
              "4     -1.528919 -0.495138        2\n",
              "...         ...       ...      ...\n",
              "35779  3.202578  2.257716        1\n",
              "35780  3.181432  2.476547        1\n",
              "35781  2.369157  1.441021        1\n",
              "35782  2.970185  2.718918        1\n",
              "35783  2.662399  1.086144        1\n",
              "\n",
              "[35784 rows x 3 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j6f8KVBGtvom"
      },
      "source": [
        "#Creating the model\n",
        " \n",
        "agg_clustering = AgglomerativeClustering(n_clusters = 3, affinity = 'euclidean', linkage = 'complete')\n",
        " \n",
        "#predicting the labels\n",
        " \n",
        "labels = agg_clustering.fit_predict(X, Y)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qye8SSTvtvom",
        "outputId": "0a12351f-17e1-42dd-bee1-892c48aa3d0f"
      },
      "source": [
        "labels"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 0, 0, ..., 1, 1, 1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XuDskvc8ZdeC",
        "outputId": "38e72666-a7ce-4ae7-a408-ba700a3ec7b9"
      },
      "source": [
        "def countX(lst, x):\n",
        "    count = 0\n",
        "    for ele in lst:\n",
        "        if (ele == x):\n",
        "            count = count + 1\n",
        "    return count\n",
        "\n",
        "print('{} has occurred {} times'.format(0, countX(labels, 0)))\n",
        "print('{} has occurred {} times'.format(1, countX(labels, 1)))\n",
        "print('{} has occurred {} times'.format(2, countX(labels, 2)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0 has occurred 18984 times\n",
            "1 has occurred 10402 times\n",
            "2 has occurred 6398 times\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "y-m5VRZVnV9-",
        "outputId": "58b6e4f3-a1de-4389-8807-766e04715d6d"
      },
      "source": [
        "print(\"Accuracy:\",accuracy_score(finalDf.iloc[:,2] , labels))\n",
        "print(classification_report(finalDf.iloc[:,2] , labels))\n",
        "print(confusion_matrix(finalDf.iloc[:,2] , labels))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy: 0.2104851330203443\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.37      0.59      0.46     11928\n",
            "         1.0       0.05      0.04      0.04     11928\n",
            "         2.0       0.00      0.00      0.00     11928\n",
            "\n",
            "    accuracy                           0.21     35784\n",
            "   macro avg       0.14      0.21      0.17     35784\n",
            "weighted avg       0.14      0.21      0.17     35784\n",
            "\n",
            "[[7042 4837   49]\n",
            " [5089  490 6349]\n",
            " [6853 5075    0]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Uu6UVKsJ06JW"
      },
      "source": [
        "finalDf.drop(['cluster'], axis=1, inplace=True)\n",
        "finalDf['cluster'] = labels\n",
        "finalDf['cluster'] = finalDf['cluster'] + 1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RYtPEHlc1OG4",
        "outputId": "590d6f6e-c359-41d9-c0ed-2f46ad9c1f1b"
      },
      "source": [
        "import math\n",
        "def get_entropys(df_clustered, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3,1)):\n",
        "    p = [[round(df_clustered[(df_clustered[label_col_name]==label) & (df_clustered[cluster_col_name]==cluster)].shape[0]/\n",
        "                df_clustered[df_clustered[cluster_col_name]==cluster].shape[0],3) for label in cluster_values] for cluster in cluster_values]\n",
        "    p_log_p = [[p_i*math.log(p_i) for p_i in p_i_s if p_i!=0] for p_i_s in p]\n",
        "    e = [-round(sum(p_log_p_i),3) for p_log_p_i in p_log_p]\n",
        "    return e\n",
        "  \n",
        "entropys_5 = get_entropys(finalDf, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3,1))\n",
        "print(\"Clusters' Entropy\")\n",
        "print(\"Entropy: {0:.3f}\".format(min(entropys_5)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Clusters' Entropy\n",
            "Entropy: 0.494\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e8JwSc34FL7k",
        "outputId": "59a04062-a9cf-4f9d-f0dd-c2a9a45df176"
      },
      "source": [
        "import numpy as np\n",
        "from sklearn import metrics\n",
        "\n",
        "def purity_score(y_true, y_pred):\n",
        "    # compute contingency matrix (also called confusion matrix)\n",
        "    contingency_matrix = metrics.cluster.contingency_matrix(y_true, y_pred)\n",
        "    # return purity\n",
        "    return np.sum(np.amax(contingency_matrix, axis=0)) / np.sum(contingency_matrix) \n",
        "\n",
        "purity = purity_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The purity is:\",purity)\n",
        "\n",
        "from sklearn.metrics.cluster import v_measure_score\n",
        "v_measure = v_measure_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The v_measure is:\",v_measure)\n",
        "\n",
        "from sklearn.metrics.cluster import adjusted_rand_score\n",
        "randScore = adjusted_rand_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The rand_score is:\",randScore)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The purity is: 0.5160406885758998\n",
            "The v_measure is: 0.2530496216487975\n",
            "The rand_score is: 0.14178826784813234\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mDBYTPqqvxK6",
        "outputId": "eb9b6dc5-05f4-4c33-8b4a-f722123e027b"
      },
      "source": [
        "# Agglomerative clustering\n",
        "from numpy import unique\n",
        "from numpy import where\n",
        "from sklearn.cluster import AgglomerativeClustering\n",
        "from matplotlib import pyplot# define the model\n",
        "model = AgglomerativeClustering(n_clusters=3)\n",
        "# fit model and predict clusters\n",
        "yhat = model.fit(cluster_df)\n",
        "yhat_2 = model.fit_predict(cluster_df)\n",
        "# retrieve unique clusters\n",
        "clusters = unique(yhat)\n",
        "# Calculate cluster validation metrics\n",
        "score_AGclustering_s = silhouette_score(cluster_df, yhat.labels_, metric='euclidean')\n",
        "score_AGclustering_c = calinski_harabasz_score(cluster_df, yhat.labels_)\n",
        "score_AGclustering_d = davies_bouldin_score(cluster_df, yhat_2)\n",
        "print('Silhouette Score: %.4f' % score_AGclustering_s)\n",
        "print('Calinski Harabasz Score: %.4f' % score_AGclustering_c)\n",
        "print('Davies Bouldin Score: %.4f' % score_AGclustering_d)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Silhouette Score: 0.2307\n",
            "Calinski Harabasz Score: 8757.3304\n",
            "Davies Bouldin Score: 1.5863\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "35Y1d9QBtvon",
        "outputId": "4f28c53d-0299-49e0-c438-c91aca81f3f7"
      },
      "source": [
        "#filter rows of original data\n",
        "filtered_label0 = df[labels == 0]\n",
        "filtered_label1 = df[labels == 1]\n",
        "filtered_label2 = df[labels == 2]\n",
        " \n",
        "#Plotting the results\n",
        "plt.scatter(filtered_label0.iloc[:,0] , filtered_label0.iloc[:,1] , color = 'red')\n",
        "plt.scatter(filtered_label1.iloc[:,0] , filtered_label1.iloc[:,1] , color = 'black')\n",
        "plt.scatter(filtered_label2.iloc[:,0] , filtered_label2.iloc[:,1] , color = 'green')\n",
        "plt.title(\"Hierarchical clustering - Agglomerative clustering\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Zg3EDjGuD9y0"
      },
      "source": [
        "\"\"\"import math\n",
        "def get_entropys(df_clustered, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3,1)):\n",
        "    p = [[round(df_clustered[(df_clustered[label_col_name]==label) & (df_clustered[cluster_col_name]==cluster)].shape[0]/\n",
        "                df_clustered[df_clustered[cluster_col_name]==cluster].shape[0],3) for label in cluster_values] for cluster in cluster_values]\n",
        "    p_log_p = [[p_i*math.log(p_i) for p_i in p_i_s if p_i!=0] for p_i_s in p]\n",
        "    e = [-round(sum(p_log_p_i),3) for p_log_p_i in p_log_p]\n",
        "    return e\n",
        "  \n",
        "entropys_5 = get_entropys(cluster_df, label_col_name=finalDf['indicator'], cluster_col_name=yhat.labels_, cluster_values=range(1,3,1))\n",
        "print(\"Clusters' Entropy\")\n",
        "print(\"Entropy: {0:.3f}\".format(min(entropys_5)))\"\"\""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NNEpLyCptvon"
      },
      "source": [
        "from sklearn.datasets import load_iris\n",
        "from sklearn.cluster import AgglomerativeClustering\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from scipy.cluster.hierarchy import dendrogram , linkage"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AdxtWiBctvoo"
      },
      "source": [
        "def plot_dendrogram(model, **kwargs):\n",
        "    # Create linkage matrix and then plot the dendrogram\n",
        "\n",
        "    # create the counts of samples under each node\n",
        "    counts = np.zeros(model.children_.shape[0])\n",
        "    n_samples = len(model.labels_)\n",
        "    for i, merge in enumerate(model.children_):\n",
        "        current_count = 0\n",
        "        for child_idx in merge:\n",
        "            if child_idx < n_samples:\n",
        "                current_count += 1  # leaf node\n",
        "            else:\n",
        "                current_count += counts[child_idx - n_samples]\n",
        "        counts[i] = current_count\n",
        "\n",
        "    linkage_matrix = np.column_stack([model.children_, model.distances_,\n",
        "                                      counts]).astype(float)\n",
        "\n",
        "    # Plot the corresponding dendrogram\n",
        "    dendrogram(linkage_matrix, **kwargs)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 246
        },
        "id": "VlkjmFQNtvop",
        "outputId": "1628aaa6-10a6-457c-a683-045091e87d85"
      },
      "source": [
        "Z = linkage(df, method = 'ward')\n",
        " \n",
        "\"\"\"#plotting dendrogram\n",
        "dendro = dendrogram(Z)\n",
        "plt.title('Dendrogram')\n",
        "plt.ylabel('Euclidean distance')\n",
        "plt.show()\"\"\""
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-35-9dabf2923506>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mZ\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlinkage\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmethod\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'ward'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \"\"\"#plotting dendrogram\n\u001b[1;32m      4\u001b[0m \u001b[0mdendro\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mdendrogram\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mZ\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtitle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'Dendrogram'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'df' is not defined"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        },
        "id": "Mt9Te75u1YF8",
        "outputId": "838bcf98-535a-4b68-c94b-785fe6058d6e"
      },
      "source": [
        "from matplotlib.pyplot import figure\n",
        "figure(figsize=(20, 20))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Figure size 1440x1440 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1440x1440 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oUNlqDy4mEBl"
      },
      "source": [
        "**TRY FCLUSTER() TO EXTRACT LABELS**\n",
        "\n",
        "**Also try method='complete',and method='single' for linkage**\n",
        "\n",
        "\n",
        "```\n",
        "from scipy.cluster.hierarchy import fcluster\n",
        "\n",
        "# Use fcluster to extract labels: labels\n",
        "labels = fcluster(mergings, 6, criterion='distance')\n",
        "```\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FXR58SX3p-03"
      },
      "source": [
        "**t-SNE Clustering**\n",
        "\n",
        "\n",
        "\n",
        "```\n",
        "# Import TSNE\n",
        "from sklearn.manifold import TSNE\n",
        "\n",
        "# Create a TSNE instance: model\n",
        "model = TSNE(learning_rate=200)\n",
        "\n",
        "# Apply fit_transform to samples: tsne_features\n",
        "tsne_features = model.fit_transform(samples)\n",
        "\n",
        "# Select the 0th feature: xs\n",
        "xs = tsne_features[:,0]\n",
        "\n",
        "# Select the 1st feature: ys\n",
        "ys = tsne_features[:,1]\n",
        "\n",
        "# Scatter plot, coloring by variety_numbers\n",
        "plt.scatter(xs, ys, c=variety_numbers)\n",
        "plt.show()\n",
        "\n",
        "```\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 507
        },
        "id": "15Mwq5c21pdP",
        "outputId": "69250530-92c9-4b91-8778-b3023d6518a1"
      },
      "source": [
        "plt.title('Hierarchical Clustering Dendrogram')\n",
        "plt.rcParams[\"figure.figsize\"] = (30,20)\n",
        "# Plot axis labels\n",
        "plt.xlabel('data index')\n",
        "plt.xticks([])\n",
        "plt.ylabel('distance (Ward)')\n",
        "\n",
        "# Make the dendrogram\n",
        "dendrogram(Z, leaf_rotation=90)\n",
        "plt.xticks([])\n",
        "# Show the graph\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "NameError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-37-9159dac07ec2>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m \u001b[0;31m# Make the dendrogram\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 9\u001b[0;31m \u001b[0mdendrogram\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mZ\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mleaf_rotation\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m90\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     10\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxticks\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     11\u001b[0m \u001b[0;31m# Show the graph\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mNameError\u001b[0m: name 'Z' is not defined"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 441
        },
        "id": "Rs7H7gwmv6nS",
        "outputId": "824732a3-af0d-483d-db80-509c5097ee96"
      },
      "source": [
        "from sklearn.neighbors import NearestNeighbors\n",
        "nearest_neighbors = NearestNeighbors(n_neighbors=11)\n",
        "neighbors = nearest_neighbors.fit(cluster_df)\n",
        "distances, indices = neighbors.kneighbors(cluster_df)\n",
        "distances = np.sort(distances[:,10], axis=0)\n",
        "!pip install -q kneed\n",
        "from kneed import KneeLocator\n",
        "i = np.arange(len(distances))\n",
        "knee = KneeLocator(i, distances, S=1, curve='convex', direction='increasing', interp_method='polynomial')\n",
        "fig = plt.figure(figsize=(5, 5))\n",
        "knee.plot_knee()\n",
        "plt.xlabel(\"Points\")\n",
        "plt.ylabel(\"Distance\")\n",
        "print(distances[knee.knee])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.1342385303409489\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x360 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 858
        },
        "id": "8Ve4TDG8v98L",
        "outputId": "a122fdf8-1d77-4e45-8049-f905d1bbe6d9"
      },
      "source": [
        "from sklearn.metrics import silhouette_score \n",
        "from sklearn.cluster import DBSCAN\n",
        "\n",
        "def get_dbscan_score(data, center):\n",
        "    '''\n",
        "    \n",
        "    INPUT:\n",
        "        data - the dataset you want to fit kmeans to\n",
        "        center - the number of centers you want (the k value)\n",
        "    OUTPUT:\n",
        "        score - the Silhouette Score for DBSCAN\n",
        "    '''\n",
        "    #instantiate kmeans\n",
        "    dbscan = DBSCAN(eps= 0.1342385303409489, min_samples=center)\n",
        "\n",
        "    # Then fit the model to your data using the fit method\n",
        "    model = dbscan.fit(cluster_df)\n",
        "    \n",
        "    # Calculate Silhoutte Score\n",
        "\n",
        "    score = silhouette_score(cluster_df, model.labels_, metric='euclidean')\n",
        "    \n",
        "    return score\n",
        "\n",
        "scores = []\n",
        "centers = list(range(2,30))\n",
        "\n",
        "for center in centers:\n",
        "    scores.append(get_dbscan_score(cluster_df, center))\n",
        "    \n",
        "plt.plot(centers, scores, linestyle='--', marker='o', color='b');\n",
        "plt.xlabel('min_samples');\n",
        "plt.ylabel('Silhouette Score');\n",
        "plt.title('Silhouette Score vs. min_samples');\n",
        "\n",
        "df3 = pd.DataFrame(centers,columns=['min_samples'])\n",
        "df3['scores'] = scores\n",
        "df4 = df3[df3.scores == df3.scores.max()]\n",
        "print('Optimal number of min_samples based on silhouette score:', df4['min_samples'].tolist())"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Optimal number of min_samples based on silhouette score: [2]\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 2160x1440 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F44MdolYwwOx",
        "outputId": "af7f5f1b-65b9-4f73-dc97-29b1d290ebaf"
      },
      "source": [
        "from numpy import unique\n",
        "from numpy import where\n",
        "from sklearn.cluster import DBSCAN\n",
        "from sklearn.metrics import silhouette_score \n",
        "from sklearn.metrics import calinski_harabasz_score\n",
        "from sklearn.metrics import davies_bouldin_score\n",
        "from matplotlib import pyplot\n",
        "# define dataset\n",
        "# define the model\n",
        "model = DBSCAN(eps=0.1342385303409489, min_samples= 7)\n",
        "# rule of thumb for min_samples: 2*len(cluster_df.columns)\n",
        "# fit model and predict clusters\n",
        "yhat = model.fit_predict(cluster_df)\n",
        "# retrieve unique clusters\n",
        "clusters = unique(yhat)\n",
        "# Calculate cluster validation metrics\n",
        "score_dbsacn_s = silhouette_score(cluster_df, yhat, metric='euclidean')\n",
        "score_dbsacn_c = calinski_harabasz_score(cluster_df, yhat)\n",
        "score_dbsacn_d = davies_bouldin_score(cluster_df, yhat)\n",
        "print('Silhouette Score: %.4f' % score_dbsacn_s)\n",
        "print('Calinski Harabasz Score: %.4f' % score_dbsacn_c)\n",
        "print('Davies Bouldin Score: %.4f' % score_dbsacn_d)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Silhouette Score: 0.8212\n",
            "Calinski Harabasz Score: 546.0145\n",
            "Davies Bouldin Score: 0.7945\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S5MaKvqu1bB-"
      },
      "source": [
        "cluster_df['cluster'] = yhat\n",
        "#cluster_df['cluster'] = cluster_df['cluster'] + 1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AmuD1wW21jYA",
        "outputId": "0a6f1b64-3651-4d51-d54a-4acfb12a92bd"
      },
      "source": [
        "import math\n",
        "def get_entropys(df_clustered, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3,1)):\n",
        "    p = [[round(df_clustered[(df_clustered[label_col_name]==label) & (df_clustered[cluster_col_name]==cluster)].shape[0]/\n",
        "                df_clustered[df_clustered[cluster_col_name]==cluster].shape[0],3) for label in cluster_values] for cluster in cluster_values]\n",
        "    p_log_p = [[p_i*math.log(p_i) for p_i in p_i_s if p_i!=0] for p_i_s in p]\n",
        "    e = [-round(sum(p_log_p_i),3) for p_log_p_i in p_log_p]\n",
        "    return e\n",
        "  \n",
        "entropys_5 = get_entropys(finalDf, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3,1))\n",
        "print(\"Clusters' Entropy\")\n",
        "print(entropys_5)\n",
        "print(\"Entropy: {0:.3f}\".format(min(entropys_5)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Clusters' Entropy\n",
            "[0.721, 0.494]\n",
            "Entropy: 0.494\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zTt9fztvF0Ba",
        "outputId": "969f605f-e89c-43da-d397-1a8aa28a36df"
      },
      "source": [
        "import numpy as np\n",
        "from sklearn import metrics\n",
        "\n",
        "def purity_score(y_true, y_pred):\n",
        "    # compute contingency matrix (also called confusion matrix)\n",
        "    contingency_matrix = metrics.cluster.contingency_matrix(y_true, y_pred)\n",
        "    # return purity\n",
        "    return np.sum(np.amax(contingency_matrix, axis=0)) / np.sum(contingency_matrix) \n",
        "\n",
        "purity = purity_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The purity is:\",purity)\n",
        "\n",
        "from sklearn.metrics.cluster import v_measure_score\n",
        "v_measure = v_measure_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The v_measure is:\",v_measure)\n",
        "\n",
        "from sklearn.metrics.cluster import adjusted_rand_score\n",
        "randScore = adjusted_rand_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The rand_score is:\",randScore)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The purity is: 0.5160406885758998\n",
            "The v_measure is: 0.2530496216487975\n",
            "The rand_score is: 0.14178826784813234\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 81
        },
        "id": "XVR_V0RwzA8N",
        "outputId": "2264b1ff-4498-4a95-d406-42a66e708e08"
      },
      "source": [
        "from sklearn.mixture import GaussianMixture\n",
        "n_components = range(1, 30)\n",
        "covariance_type = ['spherical', 'tied', 'diag', 'full']\n",
        "score=[]\n",
        "for cov in covariance_type:\n",
        "    for n_comp in n_components:\n",
        "        gmm=GaussianMixture(n_components=n_comp,covariance_type=cov, random_state = 10)\n",
        "        gmm.fit(cluster_df)\n",
        "        score.append((cov,n_comp,gmm.bic(cluster_df)))\n",
        "score_1 = pd.DataFrame(score)\n",
        "score_1.columns = ['Covariance_Type', 'N_Components','BIC_Score']\n",
        "score_2 = score_1[score_1.BIC_Score == score_1.BIC_Score.min()]\n",
        "\n",
        "score_2.head(n=2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Covariance_Type</th>\n",
              "      <th>N_Components</th>\n",
              "      <th>BIC_Score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>115</th>\n",
              "      <td>full</td>\n",
              "      <td>29</td>\n",
              "      <td>-1.490699e+06</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    Covariance_Type  N_Components     BIC_Score\n",
              "115            full            29 -1.490699e+06"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 79
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 81
        },
        "id": "f7En9TFgzDwV",
        "outputId": "528867ac-86a8-41c8-fdd0-5641214c04db"
      },
      "source": [
        "# Silhouette Score for GMM\n",
        "\n",
        "\n",
        "n_components = range(2, 30)\n",
        "covariance_type = ['spherical', 'tied', 'diag', 'full']\n",
        "score=[]\n",
        "for cov in covariance_type:\n",
        "    for n_comp in n_components:\n",
        "        gmm=GaussianMixture(n_components=n_comp,covariance_type=cov,random_state = 10)\n",
        "        model = gmm.fit(cluster_df)\n",
        "        model_2 = model.predict(cluster_df)\n",
        "        score_s = silhouette_score(cluster_df, model_2, metric='euclidean')\n",
        "        score.append((cov,n_comp,score_s))\n",
        "score_1 = pd.DataFrame(score)\n",
        "score_1.columns = ['Covariance_Type', 'N_Components','Silhouette_Score']\n",
        "score_2 = score_1[score_1.Silhouette_Score == score_1.Silhouette_Score.max()]\n",
        "score_2.head(n=2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Covariance_Type</th>\n",
              "      <th>N_Components</th>\n",
              "      <th>Silhouette_Score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>spherical</td>\n",
              "      <td>7</td>\n",
              "      <td>0.246607</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  Covariance_Type  N_Components  Silhouette_Score\n",
              "5       spherical             7          0.246607"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 80
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 81
        },
        "id": "iMyMbtLqzGsM",
        "outputId": "8f715a9e-a8fe-4874-ef1a-047614fdd503"
      },
      "source": [
        "# Calinski Harabasz Score for GMM\n",
        "\n",
        "\n",
        "n_components = range(2, 30)\n",
        "covariance_type = ['spherical', 'tied', 'diag', 'full']\n",
        "score=[]\n",
        "for cov in covariance_type:\n",
        "    for n_comp in n_components:\n",
        "        gmm=GaussianMixture(n_components=n_comp,covariance_type=cov, random_state = 10)\n",
        "        model = gmm.fit(cluster_df)\n",
        "        model_2 = model.predict(cluster_df)\n",
        "        score_c = calinski_harabasz_score(cluster_df, model_2)\n",
        "        score.append((cov,n_comp,score_c))\n",
        "score_1 = pd.DataFrame(score)\n",
        "score_1.columns = ['Covariance_Type', 'N_Components','Calinski_Harabasz_Score']\n",
        "score_2 = score_1[score_1.Calinski_Harabasz_Score == score_1.Calinski_Harabasz_Score.max()]\n",
        "score_2.head(n=5)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Covariance_Type</th>\n",
              "      <th>N_Components</th>\n",
              "      <th>Calinski_Harabasz_Score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>56</th>\n",
              "      <td>diag</td>\n",
              "      <td>2</td>\n",
              "      <td>9451.982699</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Covariance_Type  N_Components  Calinski_Harabasz_Score\n",
              "56            diag             2              9451.982699"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 81
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 81
        },
        "id": "14xwV00nzJjH",
        "outputId": "740c8b4c-e2f1-44e0-c83d-8d9ee123a364"
      },
      "source": [
        "# Davies Bouldin Score for GMM\n",
        "\n",
        "\n",
        "n_components = range(2, 30)\n",
        "covariance_type = ['spherical', 'tied', 'diag', 'full']\n",
        "score=[]\n",
        "for cov in covariance_type:\n",
        "    for n_comp in n_components:\n",
        "        gmm=GaussianMixture(n_components=n_comp,covariance_type=cov, random_state = 10)\n",
        "        model = gmm.fit(cluster_df)\n",
        "        model_2 = model.predict(cluster_df)\n",
        "        score_d = davies_bouldin_score(cluster_df, model_2)\n",
        "        score.append((cov,n_comp,score_d))\n",
        "score_1 = pd.DataFrame(score)\n",
        "score_1.columns = ['Covariance_Type', 'N_Components','Davies_Bouldin_Score']\n",
        "score_2 = score_1[score_1.Davies_Bouldin_Score == score_1.Davies_Bouldin_Score.max()]\n",
        "score_2.head(n=2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Covariance_Type</th>\n",
              "      <th>N_Components</th>\n",
              "      <th>Davies_Bouldin_Score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>tied</td>\n",
              "      <td>6</td>\n",
              "      <td>3.306541</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Covariance_Type  N_Components  Davies_Bouldin_Score\n",
              "32            tied             6              3.306541"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-gRBYtAPzMSt",
        "outputId": "904c8b86-bf9d-4c69-8f2f-b1c53861b9fa"
      },
      "source": [
        "from numpy import unique\n",
        "from numpy import where\n",
        "from sklearn.datasets import make_classification\n",
        "from sklearn.mixture import GaussianMixture\n",
        "from matplotlib import pyplot\n",
        "# define the model\n",
        "model = GaussianMixture(n_components= 2,covariance_type= \"diag\", random_state = 10)\n",
        "# fit the model\n",
        "model.fit(cluster_df)\n",
        "# assign a cluster to each example\n",
        "yhat = model.predict(cluster_df)\n",
        "# retrieve unique clusters\n",
        "clusters = unique(yhat)\n",
        "\n",
        "\n",
        "\n",
        "# Calculate cluster validation score\n",
        "\n",
        "score_dbsacn_s = silhouette_score(cluster_df, yhat, metric='euclidean')\n",
        "\n",
        "score_dbsacn_c = calinski_harabasz_score(cluster_df, yhat)\n",
        "\n",
        "score_dbsacn_d = davies_bouldin_score(cluster_df, yhat)\n",
        "\n",
        "\n",
        "\n",
        "print('Silhouette Score: %.4f' % score_dbsacn_s)\n",
        "print('Calinski Harabasz Score: %.4f' % score_dbsacn_c)\n",
        "print('Davies Bouldin Score: %.4f' % score_dbsacn_d)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Silhouette Score: 0.4443\n",
            "Calinski Harabasz Score: 21754.9420\n",
            "Davies Bouldin Score: 0.9142\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wr3ozwUe1vGd"
      },
      "source": [
        "cluster_df['cluster'] = yhat\n",
        "cluster_df['cluster'] = cluster_df['cluster'] + 1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IGB-FiRP1xv7",
        "outputId": "aecc6ad3-a7bd-4d13-af87-c77cbadadd21"
      },
      "source": [
        "import math\n",
        "def get_entropys(df_clustered, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3,1)):\n",
        "    p = [[round(df_clustered[(df_clustered[label_col_name]==label) & (df_clustered[cluster_col_name]==cluster)].shape[0]/\n",
        "                df_clustered[df_clustered[cluster_col_name]==cluster].shape[0],3) for label in cluster_values] for cluster in cluster_values]\n",
        "    p_log_p = [[p_i*math.log(p_i) for p_i in p_i_s if p_i!=0] for p_i_s in p]\n",
        "    e = [-round(sum(p_log_p_i),3) for p_log_p_i in p_log_p]\n",
        "    return e\n",
        "  \n",
        "entropys_5 = get_entropys(finalDf, label_col_name='indicator', cluster_col_name='cluster', cluster_values=range(1,3, 1))\n",
        "print(\"Clusters' Entropy\")\n",
        "print(\"Entropy: {0:.3f}\".format(min(entropys_5)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Clusters' Entropy\n",
            "Entropy: 0.494\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kEEXxjNk10wD",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b25ef002-82c0-45d6-bcb8-01603754f4e8"
      },
      "source": [
        "import numpy as np\n",
        "from sklearn import metrics\n",
        "\n",
        "def purity_score(y_true, y_pred):\n",
        "    # compute contingency matrix (also called confusion matrix)\n",
        "    contingency_matrix = metrics.cluster.contingency_matrix(y_true, y_pred)\n",
        "    # return purity\n",
        "    return np.sum(np.amax(contingency_matrix, axis=0)) / np.sum(contingency_matrix) \n",
        "\n",
        "purity = purity_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The purity is:\",purity)\n",
        "\n",
        "from sklearn.metrics.cluster import v_measure_score\n",
        "v_measure = v_measure_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The v_measure is:\",v_measure)\n",
        "\n",
        "from sklearn.metrics.cluster import adjusted_rand_score\n",
        "randScore = adjusted_rand_score(finalDf['indicator'], finalDf['cluster'])\n",
        "print(\"The rand_score is:\",randScore)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The purity is: 0.5160406885758998\n",
            "The v_measure is: 0.2530496216487975\n",
            "The rand_score is: 0.14178826784813234\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}