AirPollutionMultivariate.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "AirPollutionMultivariate.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "YjRQU2tmYFGb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Importing libraries and loading dataset\n",
        "import numpy as np\n",
        "%matplotlib inline\n",
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "from datetime import datetime\n",
        "\n",
        "def parse(x):\n",
        "    return datetime.strptime(x, '%Y %m %d %H')\n",
        "                          \n",
        "org_col_names=[\"No\", \"year\",\"month\", \"day\", \"hour\", \"pm2.5\", \"DEWP\",\"TEMP\", \"PRES\", \"cbwd\", \"Iws\", \"Is\", \"Ir\"]\n",
        "col_names = ['pollution', 'dew', 'temp', 'pressure', 'w_dir', 'w_speed', 'snow', 'rain']   \n",
        "                            \n",
        "dataset = pd.read_csv('AirPollution.csv',index_col=0, date_parser=parse,parse_dates=[['year', 'month', 'day', 'hour']])                            "
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QWW63YcTnNTL",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "68e79825-5251-4c42-9ed2-f4603cbb2dc5"
      },
      "source": [
        "dataset.head()"
      ],
      "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>No</th>\n",
              "      <th>pm2.5</th>\n",
              "      <th>DEWP</th>\n",
              "      <th>TEMP</th>\n",
              "      <th>PRES</th>\n",
              "      <th>cbwd</th>\n",
              "      <th>Iws</th>\n",
              "      <th>Is</th>\n",
              "      <th>Ir</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>year_month_day_hour</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2010-01-01 00:00:00</th>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>-21</td>\n",
              "      <td>-11.0</td>\n",
              "      <td>1021.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>1.79</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2010-01-01 01:00:00</th>\n",
              "      <td>2</td>\n",
              "      <td>NaN</td>\n",
              "      <td>-21</td>\n",
              "      <td>-12.0</td>\n",
              "      <td>1020.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>4.92</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2010-01-01 02:00:00</th>\n",
              "      <td>3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>-21</td>\n",
              "      <td>-11.0</td>\n",
              "      <td>1019.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>6.71</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2010-01-01 03:00:00</th>\n",
              "      <td>4</td>\n",
              "      <td>NaN</td>\n",
              "      <td>-21</td>\n",
              "      <td>-14.0</td>\n",
              "      <td>1019.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>9.84</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2010-01-01 04:00:00</th>\n",
              "      <td>5</td>\n",
              "      <td>NaN</td>\n",
              "      <td>-20</td>\n",
              "      <td>-12.0</td>\n",
              "      <td>1018.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>12.97</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     No  pm2.5  DEWP  TEMP    PRES cbwd    Iws  Is  Ir\n",
              "year_month_day_hour                                                   \n",
              "2010-01-01 00:00:00   1    NaN   -21 -11.0  1021.0   NW   1.79   0   0\n",
              "2010-01-01 01:00:00   2    NaN   -21 -12.0  1020.0   NW   4.92   0   0\n",
              "2010-01-01 02:00:00   3    NaN   -21 -11.0  1019.0   NW   6.71   0   0\n",
              "2010-01-01 03:00:00   4    NaN   -21 -14.0  1019.0   NW   9.84   0   0\n",
              "2010-01-01 04:00:00   5    NaN   -20 -12.0  1018.0   NW  12.97   0   0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 2
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hLWvB7QcnuuZ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 136
        },
        "outputId": "90e41074-b8ec-4620-a46f-08a5f7ceeb61"
      },
      "source": [
        "# Data cleaning\n",
        "dataset.drop('No', axis=1, inplace=True)\n",
        "dataset.columns = col_names\n",
        "dataset['pollution'].fillna(0, inplace=True)\n",
        "dataset = dataset[24:] # drop the first day\n",
        "print(dataset.head(5))\n",
        "dataset.to_csv('pollution.csv') # save new CSV"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "                     pollution  dew  temp  pressure w_dir  w_speed  snow  rain\n",
            "year_month_day_hour                                                           \n",
            "2010-01-02 00:00:00      129.0  -16  -4.0    1020.0    SE     1.79     0     0\n",
            "2010-01-02 01:00:00      148.0  -15  -4.0    1020.0    SE     2.68     0     0\n",
            "2010-01-02 02:00:00      159.0  -11  -5.0    1021.0    SE     3.57     0     0\n",
            "2010-01-02 03:00:00      181.0   -7  -5.0    1022.0    SE     5.36     1     0\n",
            "2010-01-02 04:00:00      138.0   -7  -5.0    1022.0    SE     6.25     2     0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3OFd6lx-n5EP",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        },
        "outputId": "ece5ab3d-a45b-4101-f495-4c95b531ae34"
      },
      "source": [
        "# load dataset\n",
        "df = pd.read_csv('pollution.csv', header=0, index_col=0)\n",
        "df.describe()"
      ],
      "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>pollution</th>\n",
              "      <th>dew</th>\n",
              "      <th>temp</th>\n",
              "      <th>pressure</th>\n",
              "      <th>w_speed</th>\n",
              "      <th>snow</th>\n",
              "      <th>rain</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>43800.000000</td>\n",
              "      <td>43800.000000</td>\n",
              "      <td>43800.000000</td>\n",
              "      <td>43800.000000</td>\n",
              "      <td>43800.000000</td>\n",
              "      <td>43800.000000</td>\n",
              "      <td>43800.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>94.013516</td>\n",
              "      <td>1.828516</td>\n",
              "      <td>12.459041</td>\n",
              "      <td>1016.447306</td>\n",
              "      <td>23.894307</td>\n",
              "      <td>0.052763</td>\n",
              "      <td>0.195023</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>92.252276</td>\n",
              "      <td>14.429326</td>\n",
              "      <td>12.193384</td>\n",
              "      <td>10.271411</td>\n",
              "      <td>50.022729</td>\n",
              "      <td>0.760582</td>\n",
              "      <td>1.416247</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.000000</td>\n",
              "      <td>-40.000000</td>\n",
              "      <td>-19.000000</td>\n",
              "      <td>991.000000</td>\n",
              "      <td>0.450000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>24.000000</td>\n",
              "      <td>-10.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>1008.000000</td>\n",
              "      <td>1.790000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>68.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>1016.000000</td>\n",
              "      <td>5.370000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>132.250000</td>\n",
              "      <td>15.000000</td>\n",
              "      <td>23.000000</td>\n",
              "      <td>1025.000000</td>\n",
              "      <td>21.910000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>994.000000</td>\n",
              "      <td>28.000000</td>\n",
              "      <td>42.000000</td>\n",
              "      <td>1046.000000</td>\n",
              "      <td>585.600000</td>\n",
              "      <td>27.000000</td>\n",
              "      <td>36.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          pollution           dew  ...          snow          rain\n",
              "count  43800.000000  43800.000000  ...  43800.000000  43800.000000\n",
              "mean      94.013516      1.828516  ...      0.052763      0.195023\n",
              "std       92.252276     14.429326  ...      0.760582      1.416247\n",
              "min        0.000000    -40.000000  ...      0.000000      0.000000\n",
              "25%       24.000000    -10.000000  ...      0.000000      0.000000\n",
              "50%       68.000000      2.000000  ...      0.000000      0.000000\n",
              "75%      132.250000     15.000000  ...      0.000000      0.000000\n",
              "max      994.000000     28.000000  ...     27.000000     36.000000\n",
              "\n",
              "[8 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6bRhvqIcpYeV",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "fa0e3ca9-b0a8-4801-e9e6-12970046c684"
      },
      "source": [
        "dataset_columns = df.columns.tolist()\n",
        "dataset_columns"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['pollution', 'dew', 'temp', 'pressure', 'w_dir', 'w_speed', 'snow', 'rain']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OplB_zJ-pkqj",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "outputId": "e390c278-c79e-40d3-ae79-5febdd209119"
      },
      "source": [
        "#Box plot\n",
        "#pd.options.display.mpl_style = False\n",
        "df.boxplot()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fd65b63cf98>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 6
        },
        {
          "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": "QAO3Nx3kvify",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 450
        },
        "outputId": "35834c14-c397-44fd-d0fe-66e54e8bf412"
      },
      "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>pollution</th>\n",
              "      <th>dew</th>\n",
              "      <th>temp</th>\n",
              "      <th>pressure</th>\n",
              "      <th>w_dir</th>\n",
              "      <th>w_speed</th>\n",
              "      <th>snow</th>\n",
              "      <th>rain</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>year_month_day_hour</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2010-01-02 00:00:00</th>\n",
              "      <td>129.0</td>\n",
              "      <td>-16</td>\n",
              "      <td>-4.0</td>\n",
              "      <td>1020.0</td>\n",
              "      <td>SE</td>\n",
              "      <td>1.79</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2010-01-02 01:00:00</th>\n",
              "      <td>148.0</td>\n",
              "      <td>-15</td>\n",
              "      <td>-4.0</td>\n",
              "      <td>1020.0</td>\n",
              "      <td>SE</td>\n",
              "      <td>2.68</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2010-01-02 02:00:00</th>\n",
              "      <td>159.0</td>\n",
              "      <td>-11</td>\n",
              "      <td>-5.0</td>\n",
              "      <td>1021.0</td>\n",
              "      <td>SE</td>\n",
              "      <td>3.57</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2010-01-02 03:00:00</th>\n",
              "      <td>181.0</td>\n",
              "      <td>-7</td>\n",
              "      <td>-5.0</td>\n",
              "      <td>1022.0</td>\n",
              "      <td>SE</td>\n",
              "      <td>5.36</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2010-01-02 04:00:00</th>\n",
              "      <td>138.0</td>\n",
              "      <td>-7</td>\n",
              "      <td>-5.0</td>\n",
              "      <td>1022.0</td>\n",
              "      <td>SE</td>\n",
              "      <td>6.25</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2014-12-31 19:00:00</th>\n",
              "      <td>8.0</td>\n",
              "      <td>-23</td>\n",
              "      <td>-2.0</td>\n",
              "      <td>1034.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>231.97</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2014-12-31 20:00:00</th>\n",
              "      <td>10.0</td>\n",
              "      <td>-22</td>\n",
              "      <td>-3.0</td>\n",
              "      <td>1034.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>237.78</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2014-12-31 21:00:00</th>\n",
              "      <td>10.0</td>\n",
              "      <td>-22</td>\n",
              "      <td>-3.0</td>\n",
              "      <td>1034.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>242.70</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2014-12-31 22:00:00</th>\n",
              "      <td>8.0</td>\n",
              "      <td>-22</td>\n",
              "      <td>-4.0</td>\n",
              "      <td>1034.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>246.72</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2014-12-31 23:00:00</th>\n",
              "      <td>12.0</td>\n",
              "      <td>-21</td>\n",
              "      <td>-3.0</td>\n",
              "      <td>1034.0</td>\n",
              "      <td>NW</td>\n",
              "      <td>249.85</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>43800 rows × 8 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                     pollution  dew  temp  pressure w_dir  w_speed  snow  rain\n",
              "year_month_day_hour                                                           \n",
              "2010-01-02 00:00:00      129.0  -16  -4.0    1020.0    SE     1.79     0     0\n",
              "2010-01-02 01:00:00      148.0  -15  -4.0    1020.0    SE     2.68     0     0\n",
              "2010-01-02 02:00:00      159.0  -11  -5.0    1021.0    SE     3.57     0     0\n",
              "2010-01-02 03:00:00      181.0   -7  -5.0    1022.0    SE     5.36     1     0\n",
              "2010-01-02 04:00:00      138.0   -7  -5.0    1022.0    SE     6.25     2     0\n",
              "...                        ...  ...   ...       ...   ...      ...   ...   ...\n",
              "2014-12-31 19:00:00        8.0  -23  -2.0    1034.0    NW   231.97     0     0\n",
              "2014-12-31 20:00:00       10.0  -22  -3.0    1034.0    NW   237.78     0     0\n",
              "2014-12-31 21:00:00       10.0  -22  -3.0    1034.0    NW   242.70     0     0\n",
              "2014-12-31 22:00:00        8.0  -22  -4.0    1034.0    NW   246.72     0     0\n",
              "2014-12-31 23:00:00       12.0  -21  -3.0    1034.0    NW   249.85     0     0\n",
              "\n",
              "[43800 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QGNtHSYCppXH",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 258
        },
        "outputId": "e257b28c-6117-46d7-a333-76b380fa0d23"
      },
      "source": [
        "#Analysing relationship between different features\n",
        "# cor_cols = ['pollution', 'wnd_spd', 'rain', 'snow', 'temp']\n",
        "cor_cols = dataset_columns\n",
        "plt.matshow(df.corr())\n",
        "plt.xticks(range(len(cor_cols)), cor_cols)\n",
        "plt.yticks(range(len(cor_cols)), cor_cols)\n",
        "plt.colorbar()\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 288x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Zk5k12EiqKfP",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "92241b0c-fd17-4820-dbe6-ec0994372400"
      },
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "# Encode non categorical values\n",
        "values = df.values\n",
        "encoder = LabelEncoder()\n",
        "values[:,4] = encoder.fit_transform(values[:,4])\n",
        "values = values.astype('float32')\n",
        "values[:,4]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([2., 2., 2., ..., 1., 1., 1.], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mtVERgwftcNL",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Normalising data\n",
        "from sklearn.preprocessing import MinMaxScaler\n",
        "\n",
        "scaler = MinMaxScaler(feature_range=(0, 1))\n",
        "scaled_dataset = scaler.fit_transform(values)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fo5o7LHvtmam",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Creating a window for previous data\n",
        "def to_supervised(train):\n",
        "  window_size = 4\n",
        "  X = []\n",
        "  Y = []\n",
        "  for i in range(window_size, len(train)):\n",
        "    X.append(train[i-window_size:i,:])\n",
        "    Y.append(train[i,0:1])\n",
        "    \n",
        "  return X,Y"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9Japs75UuF-D",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "232fa0f4-c3e0-4981-a0ad-1cd21e580504"
      },
      "source": [
        "X, Y = to_supervised(scaled_dataset)\n",
        "X = np.array(X)\n",
        "Y = np.array(Y)\n",
        "print('Y' ,Y.shape)\n",
        "print('X' ,X.shape)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Y (43796, 1)\n",
            "X (43796, 4, 8)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "A--d92h9uQV_",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "06dc1c61-2e25-470b-aa5d-40d454ab7b84"
      },
      "source": [
        "#Splitting the dataset\n",
        "n_train = 24*365\n",
        "X_train, X_test = X[n_train:,] , X[:n_train,]\n",
        "print('X_train' ,X_train.shape)\n",
        "print('X_test' ,X_test.shape)\n",
        "\n",
        "Y_train, Y_test = Y[n_train:,] , Y[:n_train,]\n",
        "print('Y_train' ,Y_train.shape)\n",
        "print('Y_test' ,Y_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "X_train (35036, 4, 8)\n",
            "X_test (8760, 4, 8)\n",
            "Y_train (35036, 1)\n",
            "Y_test (8760, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MnzdpfwM7OUa",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "0a99ae98-78c9-4c79-f5df-e9fbbda12e73"
      },
      "source": [
        "#Importing LSTM model\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense, Dropout,LSTM\n",
        "model = Sequential()\n",
        "\n",
        "model.add(LSTM(units = 50, return_sequences = True, input_shape=(X_train.shape[1], X_train.shape[2])))\n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(LSTM(units = 50, return_sequences = True))\n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(LSTM(units = 50))\n",
        "model.add(Dropout(0.2))\n",
        "\n",
        "model.add(Dense(units = 1))\n",
        "\n",
        "model.compile(optimizer = 'adam', loss = 'mean_squared_error')         "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fI1ihWPs-_2w",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 714
        },
        "outputId": "dd7af362-3c73-469a-a6d4-b326af713e4f"
      },
      "source": [
        "#Train the model\n",
        "model.fit(X_train, Y_train, epochs = 20, batch_size = 32)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "35036/35036 [==============================] - 30s 871us/step - loss: 0.0021\n",
            "Epoch 2/20\n",
            "35036/35036 [==============================] - 29s 815us/step - loss: 0.0011\n",
            "Epoch 3/20\n",
            "35036/35036 [==============================] - 28s 788us/step - loss: 9.0931e-04\n",
            "Epoch 4/20\n",
            "35036/35036 [==============================] - 27s 778us/step - loss: 8.4520e-04\n",
            "Epoch 5/20\n",
            "35036/35036 [==============================] - 27s 783us/step - loss: 8.4552e-04\n",
            "Epoch 6/20\n",
            "35036/35036 [==============================] - 28s 788us/step - loss: 8.4213e-04\n",
            "Epoch 7/20\n",
            "35036/35036 [==============================] - 28s 795us/step - loss: 8.2123e-04\n",
            "Epoch 8/20\n",
            "35036/35036 [==============================] - 27s 776us/step - loss: 8.2782e-04\n",
            "Epoch 9/20\n",
            "35036/35036 [==============================] - 27s 773us/step - loss: 8.1744e-04\n",
            "Epoch 10/20\n",
            "35036/35036 [==============================] - 27s 769us/step - loss: 7.9839e-04\n",
            "Epoch 11/20\n",
            "35036/35036 [==============================] - 27s 761us/step - loss: 8.0556e-04\n",
            "Epoch 12/20\n",
            "35036/35036 [==============================] - 27s 769us/step - loss: 8.0480e-04\n",
            "Epoch 13/20\n",
            "35036/35036 [==============================] - 27s 770us/step - loss: 8.0340e-04\n",
            "Epoch 14/20\n",
            "35036/35036 [==============================] - 27s 769us/step - loss: 8.0428e-04\n",
            "Epoch 15/20\n",
            "35036/35036 [==============================] - 27s 763us/step - loss: 7.7593e-04\n",
            "Epoch 16/20\n",
            "35036/35036 [==============================] - 27s 774us/step - loss: 7.6892e-04\n",
            "Epoch 17/20\n",
            "35036/35036 [==============================] - 27s 778us/step - loss: 7.8611e-04\n",
            "Epoch 18/20\n",
            "35036/35036 [==============================] - 28s 785us/step - loss: 7.8265e-04\n",
            "Epoch 19/20\n",
            "35036/35036 [==============================] - 27s 775us/step - loss: 7.6438e-04\n",
            "Epoch 20/20\n",
            "35036/35036 [==============================] - 27s 780us/step - loss: 7.7393e-04\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.callbacks.History at 0x7fd5b6197f98>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "E0ORsAIKAwB2",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Taking predictions\n",
        "from keras.models import load_model\n",
        "model.save('AirPollutionMultivariate.h5')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "C6fEpNBnC7kz",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "b47dfbee-31f0-4e8f-d756-10af77354fc5"
      },
      "source": [
        "Y_pred = model.predict(X_test)\n",
        "\n",
        "from sklearn.metrics import mean_squared_error\n",
        "mse = mean_squared_error(Y_test, Y_pred)\n",
        "\n",
        "rmse = np.sqrt(mse)\n",
        "rmse"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.028314134"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mA0ofmp1DPSN",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "7407a957-df78-465f-d3be-83d508eb1956"
      },
      "source": [
        "#Y_predicted = scaler.inverse_transform(Y_pred)\n",
        "print('y_predicted',Y_pred.shape)\n",
        "#print('X_train' ,X_train.shape)\n",
        "print('X_test' ,X_test.shape)\n",
        "print('scaled Values shape', scaled_dataset.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "y_predicted (8760, 1)\n",
            "X_test (8760, 4, 8)\n",
            "scaled Values shape (43800, 8)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-4AniFeyDfD7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 85
        },
        "outputId": "cc64e769-44e1-4ffd-9bcd-35b483ed9e68"
      },
      "source": [
        "#Scaling back to the original scale\n",
        "d = scaled_dataset[:8760,:]\n",
        "print('dummy',d.shape)\n",
        "print('Y_pred',Y_pred.shape)\n",
        "Y_predicted = np.concatenate((Y_pred,d[:8760,1:]), axis =1)\n",
        "print('concat y_pred',Y_pred.shape)\n",
        "Y_tested = np.concatenate((Y_test, d[:8760,1:]), axis = 1)\n",
        "print('concat Y_test', Y_test.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "dummy (8760, 8)\n",
            "Y_pred (8760, 1)\n",
            "concat y_pred (8760, 1)\n",
            "concat Y_test (8760, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "13B5_7XIHPLh",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "40b6c243-090c-405c-875f-651eb40b9373"
      },
      "source": [
        "Y_predicted = scaler.inverse_transform(Y_predicted)\n",
        "Y_tested = scaler.inverse_transform(Y_tested)\n",
        "Y_predicted = Y_predicted[:,0:1]\n",
        "Y_tested = Y_tested[:,0:1]\n",
        "print('Y_tested', Y_tested.shape)\n",
        "print('Y_predicted', Y_predicted.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Y_tested (8760, 1)\n",
            "Y_predicted (8760, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-dtpEAfGQsYJ",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "ce0bbb49-d5ab-45a6-cd2b-d3fc74bb91f8"
      },
      "source": [
        "from sklearn.metrics import mean_squared_error\n",
        "mse = mean_squared_error(Y_tested, Y_predicted)\n",
        "\n",
        "rmse = np.sqrt(mse)\n",
        "rmse"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "28.14425"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "40LyAdCZfZet",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 421
        },
        "outputId": "72ed7906-b53b-48e6-bf4a-5673bcbdb2eb"
      },
      "source": [
        "#Plot the graph between actual vs predicted values\n",
        "plt.figure(figsize=(10,6))  \n",
        "plt.plot(Y_predicted[:100,:], color= 'green',label = 'Predicted Pollution level')\n",
        "plt.plot(Y_tested[:100,:] , color = 'red',label = 'Actual Pollution level')\n",
        "plt.title(\"Air Pollution Prediction (Multivariate)\")\n",
        "plt.xlabel(\"Date\")\n",
        "plt.ylabel(\"Pollution level\")\n",
        "plt.legend()\n",
        "plt.show()\n",
        "plt.savefig('graph.png')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 0 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ulPQvfoXmyC4",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 68
        },
        "outputId": "0de176eb-b4c3-47f5-b55b-bd220dff9176"
      },
      "source": [
        "def mean_absolute_percentage_error(y_true, y_pred): \n",
        "    y_true, y_pred = np.array(y_true), np.array(y_pred)\n",
        "    return np.mean(np.abs((y_true - y_pred) / y_true)) * 100\n",
        "\n",
        "mape = mean_absolute_percentage_error(Y_tested, Y_predicted)\n",
        "print('MAPE', mape)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "MAPE inf\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:3: RuntimeWarning: divide by zero encountered in true_divide\n",
            "  This is separate from the ipykernel package so we can avoid doing imports until\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BCODs7Kom7uK",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 51
        },
        "outputId": "267f304e-03d7-43b0-92ea-ebc300c39823"
      },
      "source": [
        "from sklearn.metrics import mean_squared_error\n",
        "mse = mean_squared_error(Y_tested, Y_predicted)\n",
        "#print('Mean squared error', mse)\n",
        "print('RMSE' , np.sqrt(mse))\n",
        "print(\"Mean of Test data \", np.mean(Y_tested))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "RMSE 28.14425\n",
            "Mean of Test data  96.041435\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r3vdque_zCoo",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}