02_example_of_preparation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Datasci project (example of preparation).ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "machine_shape": "hm"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "TPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QyArwhIQDhNG"
      },
      "source": [
        "If you want to run this colab please follow the instructions below\n",
        "\n",
        "1. Download and upload the data from https://drive.google.com/drive/u/1/folders/1TcE9NNShm0irVqZM7y4Xtz7ibjHF9tAo to the colab (Data science project 2020/Data science project1/PM2.5_kaggle_2021)\n",
        "\n",
        "2. Unzip the uploaded folder by typing the zip file name in cell 1 and rename foler after unzip to 'PM2.5_kaggle_2021'\n",
        "\n",
        "3. Edit path in cell 3 to '/content/PM2.5_kaggle_2021'"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tV1yF6l91xzI"
      },
      "source": [
        "import os\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import datetime as dt\n",
        "import matplotlib.pyplot as plt\n",
        "import math\n",
        "import datetime\n",
        "from math import pi"
      ],
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pQ_crx6z2LzA"
      },
      "source": [
        "path = '/content/drive/Shareddrives/Data science project/PM2.5_kaggle_2021'"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hMvcqHYh3SGU"
      },
      "source": [
        "# Prepare data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "M50tJb7t2No4"
      },
      "source": [
        "def combineDateTime(df):\n",
        "    df['datetime'] = pd.to_datetime({'year': df['year'], 'month': df['month'], 'day':df['day'],'hour': df['UTC hour']})\n",
        "    drop_columns = ['year', 'month', 'day', 'UTC hour']\n",
        "    df = df.drop(drop_columns, axis=1)\n",
        "    return df\n",
        "\n",
        "def preparePM25(df):\n",
        "    df = combineDateTime(df)\n",
        "    df = df.drop(['PM10_mask', 'Retrospective'], axis=1)\n",
        "    df = resampling(df, 1)\n",
        "    return df\n",
        "\n",
        "def prepareTempAndWind(file):\n",
        "    df = pd.read_csv(os.path.join(path, file))\n",
        "    df['datetime'] = pd.to_datetime(df['datetime'])\n",
        "    df = df.rename(columns={'Temp(C)': 'Temp'})\n",
        "    df = resampling(df, 1)\n",
        "    return df\n",
        "\n",
        "def prepareFire():\n",
        "    for country in ['Thailand', 'Cambodia', 'Lao_PDR', 'Myanmar']:\n",
        "        dfs = []\n",
        "        for year in ['2016', '2017', '2018', '2019']:\n",
        "            df = pd.read_csv(os.path.join(path, 'Fire hotspot', 'viirs-snpp_' + year + '_' + country + '.csv'))\n",
        "            dfs.append(df)\n",
        "        df = pd.concat(dfs, axis=0, ignore_index=True)\n",
        "        df['acq_time'] = roundup(df['acq_time'])\n",
        "        df['acq_time'] = df['acq_time'].astype(str)\n",
        "        df['datetime'] = pd.to_datetime(df['acq_date'] + df['acq_time'], format='%Y-%m-%d%H%M')\n",
        "        df = changeTimezone(df)\n",
        "        df['hotspot_count_' + country] = 1\n",
        "        df = df[['hotspot_count_' + country]]\n",
        "        df = df.groupby('datetime').agg({'hotspot_count_' + country: 'sum'})\n",
        "        if country == 'Thailand':\n",
        "            df_fire = df.copy()\n",
        "        else:\n",
        "            df_fire = df_fire.merge(df, left_index=True, right_index=True, how='outer')\n",
        "    df_fire = df_fire.fillna(0)\n",
        "    return df_fire\n",
        "\n",
        "def prepareData(pm25, temp, wind, fire):\n",
        "    df = pm25.merge(temp, left_index=True, right_index=True)\n",
        "    df = df.merge(wind.drop(['lat', 'long'], axis=1), left_index=True, right_index=True)\n",
        "    df = df.merge(fire, how='left', left_index=True, right_index=True)\n",
        "    # df['latitude'].fillna(df['lat'], inplace=True)\n",
        "    # df['longitude'].fillna(df['long'], inplace=True)\n",
        "    # df['distance'] = distance(df['lat'], df['long'], df['latitude'], df['longitude'])\n",
        "    # df['direction'] = direction(df['lat'], df['long'], df['latitude'], df['longitude'])\n",
        "    # df['direction_diff'] = abs(df['direction'] - df['WindDir'])\n",
        "    # df['bright_ti4'].fillna(0, inplace=True)\n",
        "    # df['bright_ti5'].fillna(0, inplace=True)\n",
        "    # df['frp'].fillna(0, inplace=True)\n",
        "    for col in fire.columns:\n",
        "        df[col] = df[col].fillna(0)\n",
        "    df = df[~df.index.duplicated(keep='first')]\n",
        "    df.index = pd.DatetimeIndex(df.index)\n",
        "    return df\n",
        "\n",
        "def changeTimezone(df):\n",
        "    if type(df.index) is pd.core.indexes.datetimes.DatetimeIndex:\n",
        "        df = df.reset_index('datetime')\n",
        "    df['datetime'] = df['datetime'].dt.tz_localize('UTC').dt.tz_convert('Asia/Bangkok')\n",
        "    df['datetime'] = df['datetime'].dt.tz_localize(None)\n",
        "    df = addDay(df)\n",
        "\n",
        "    df = df.set_index('datetime')\n",
        "    return df\n",
        "\n",
        "def addDay(df):\n",
        "    if type(df.index) is pd.core.indexes.datetimes.DatetimeIndex:\n",
        "        df = df.reset_index('datetime')\n",
        "    df['day_of_year'] = df['datetime'].dt.dayofyear\n",
        "    df['day_of_week'] = df['datetime'].dt.dayofweek\n",
        "    return df\n",
        "\n",
        "def resampling(df, hour):\n",
        "    if not type(df.index) is pd.core.indexes.datetimes.DatetimeIndex:\n",
        "        df = df.set_index('datetime')\n",
        "    df = df.loc[~df.index.duplicated(keep='first')]\n",
        "    df = df.resample(str(hour)+'H').interpolate(method='linear')\n",
        "    return df\n",
        "\n",
        "def distance(lat1, long1, lat2, long2):\n",
        "    R = 6373.0\n",
        "    lat1, long1, lat2, long2 = np.radians([lat1, long1, lat2, long2])\n",
        "    a = np.sin((lat2-lat1)/2.0)**2 + np.cos(lat1) * np.cos(lat2) * np.sin((long2-long1)/2.0)**2\n",
        "    return R * 2 * np.arcsin(np.sqrt(a))\n",
        "  \n",
        "def direction(lat1, long1, lat2, long2):\n",
        "    dir = np.arctan2(lat1-lat2, long1-long2)/pi * 180\n",
        "    return dir + 360*(dir < 0)\n",
        "\n",
        "def roundup(x):\n",
        "    return ((np.ceil(x / 100.0)) * 100).astype(int)\n",
        "\n",
        "def addPM25Lastyear(df):\n",
        "    df_shift = df.shift(periods=24*365).rename(columns={'PM2.5': 'PM2.5_lastyear'})\n",
        "    df = df.merge(df_shift['PM2.5_lastyear'], left_index=True, right_index=True)\n",
        "    df['PM2.5_lastyear'].fillna(df['PM2.5'], inplace=True)\n",
        "    return df\n",
        "\n",
        "def addPM25LastyearForTest(df, df_train):\n",
        "    df_all = df_train.drop(columns=['PM2.5_lastyear'], axis=1).append(df)\n",
        "    df = addPM25Lastyear(df_all)\n",
        "    df = df.iloc[df_train.shape[0]:]\n",
        "    return df"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "608xO6RhQrVb"
      },
      "source": [
        "## PM2.5 data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sSGzlt0augin",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "53ac2da5-af0f-481d-e72e-b45b9ffa83f9"
      },
      "source": [
        "file = 'Bangkok/Train/Bangkok.txt'\n",
        "bkk_pm25_train = pd.read_csv(os.path.join(path, file), delimiter = '\\t', skiprows=10, header=None,\n",
        "                names=['year', 'month', 'day', 'UTC hour', 'PM2.5', 'PM10_mask', 'Retrospective'])\n",
        "bkk_pm25_train = preparePM25(bkk_pm25_train)\n",
        "bkk_pm25_train = changeTimezone(bkk_pm25_train)\n",
        "bkk_pm25_train = addDay(bkk_pm25_train)\n",
        "bkk_pm25_train = bkk_pm25_train.set_index('datetime')\n",
        "bkk_pm25_train.head(5)"
      ],
      "execution_count": 32,
      "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>PM2.5</th>\n",
              "      <th>day_of_year</th>\n",
              "      <th>day_of_week</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2016-03-03 15:00:00</th>\n",
              "      <td>62.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 16:00:00</th>\n",
              "      <td>62.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 17:00:00</th>\n",
              "      <td>55.5</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 18:00:00</th>\n",
              "      <td>55.5</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 19:00:00</th>\n",
              "      <td>47.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     PM2.5  day_of_year  day_of_week\n",
              "datetime                                            \n",
              "2016-03-03 15:00:00   62.9           63            3\n",
              "2016-03-03 16:00:00   62.9           63            3\n",
              "2016-03-03 17:00:00   55.5           63            3\n",
              "2016-03-03 18:00:00   55.5           63            3\n",
              "2016-03-03 19:00:00   47.9           63            3"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oWFYeBouu6Uh",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "d66fd400-0b47-4a72-ae15-617792d998ab"
      },
      "source": [
        "file = 'Bangkok/Test/Bangkok.csv'\n",
        "bkk_pm25_test = pd.read_csv(os.path.join(path, file), skiprows=1,\n",
        "                            names=['datetime', 'PM2.5'])\n",
        "bkk_pm25_test['datetime'] = pd.to_datetime(bkk_pm25_test['datetime'])\n",
        "bkk_pm25_test = addDay(bkk_pm25_test)\n",
        "bkk_pm25_test = bkk_pm25_test.set_index('datetime')\n",
        "bkk_pm25_test.head(5)"
      ],
      "execution_count": 33,
      "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>PM2.5</th>\n",
              "      <th>day_of_year</th>\n",
              "      <th>day_of_week</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2019-03-18 08:00:00</th>\n",
              "      <td>45.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 09:00:00</th>\n",
              "      <td>46.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 10:00:00</th>\n",
              "      <td>47.7</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 11:00:00</th>\n",
              "      <td>43.7</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 12:00:00</th>\n",
              "      <td>41.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     PM2.5  day_of_year  day_of_week\n",
              "datetime                                            \n",
              "2019-03-18 08:00:00   45.6           77            0\n",
              "2019-03-18 09:00:00   46.6           77            0\n",
              "2019-03-18 10:00:00   47.7           77            0\n",
              "2019-03-18 11:00:00   43.7           77            0\n",
              "2019-03-18 12:00:00   41.6           77            0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qdrOp7kP2kj9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 375
        },
        "outputId": "a08a3b72-2fcf-47f4-bdf8-0ca7b18e0119"
      },
      "source": [
        "plt.figure(figsize=(20,6))\n",
        "plt.plot(bkk_pm25_train['PM2.5'])\n",
        "plt.show()"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kkiFLozJQvXm"
      },
      "source": [
        "## Temperature data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bUM2QCUgvn2E",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "0530a372-ac72-49fc-d5a1-b1ae1f5eb551"
      },
      "source": [
        "bkk_temp_train = prepareTempAndWind('Bangkok/Train/3H_temperature_Bangkok.csv')\n",
        "bkk_temp_train.head(5)"
      ],
      "execution_count": 35,
      "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>Temp</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2016-03-03 01:00:00</th>\n",
              "      <td>25.500000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 02:00:00</th>\n",
              "      <td>25.333333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 03:00:00</th>\n",
              "      <td>25.166667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 04:00:00</th>\n",
              "      <td>25.000000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 05:00:00</th>\n",
              "      <td>25.466667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                          Temp     lat      long\n",
              "datetime                                        \n",
              "2016-03-03 01:00:00  25.500000  13.754  100.5014\n",
              "2016-03-03 02:00:00  25.333333  13.754  100.5014\n",
              "2016-03-03 03:00:00  25.166667  13.754  100.5014\n",
              "2016-03-03 04:00:00  25.000000  13.754  100.5014\n",
              "2016-03-03 05:00:00  25.466667  13.754  100.5014"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p8aAfkDtvtwI",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "40d757c2-d7a9-45b4-cf99-c14db9051881"
      },
      "source": [
        "bkk_temp_test = prepareTempAndWind('Bangkok/Test/3H_temperature_Bangkok.csv')\n",
        "bkk_temp_test.head(5)"
      ],
      "execution_count": 36,
      "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>Temp</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2019-03-18 01:00:00</th>\n",
              "      <td>29.200000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 02:00:00</th>\n",
              "      <td>29.033333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 03:00:00</th>\n",
              "      <td>28.866667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 04:00:00</th>\n",
              "      <td>28.700000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 05:00:00</th>\n",
              "      <td>28.900000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                          Temp     lat      long\n",
              "datetime                                        \n",
              "2019-03-18 01:00:00  29.200000  13.754  100.5014\n",
              "2019-03-18 02:00:00  29.033333  13.754  100.5014\n",
              "2019-03-18 03:00:00  28.866667  13.754  100.5014\n",
              "2019-03-18 04:00:00  28.700000  13.754  100.5014\n",
              "2019-03-18 05:00:00  28.900000  13.754  100.5014"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EUw7TUJn3_JC",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 375
        },
        "outputId": "86559d2b-0413-497d-c50a-0c9ca46ce2ba"
      },
      "source": [
        "plt.figure(figsize=(20,6))\n",
        "plt.plot(bkk_pm25_train['PM2.5'])\n",
        "plt.plot(bkk_temp_train['Temp'], 'r')\n",
        "plt.show()"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "g_wAnfJzQ-qz"
      },
      "source": [
        "## Wind data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wzG7zqDBvy2e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "6b258be1-dfa5-4f06-c301-27b3f856e9b8"
      },
      "source": [
        "bkk_wind_train = prepareTempAndWind('Bangkok/Train/3H_wind_Bangkok.csv')\n",
        "bkk_wind_train.head(5)"
      ],
      "execution_count": 38,
      "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>WindDir</th>\n",
              "      <th>Wind Speed(km/h)</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2016-03-03 01:00:00</th>\n",
              "      <td>45.000000</td>\n",
              "      <td>13.000000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 02:00:00</th>\n",
              "      <td>50.000000</td>\n",
              "      <td>12.333333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 03:00:00</th>\n",
              "      <td>55.000000</td>\n",
              "      <td>11.666667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 04:00:00</th>\n",
              "      <td>60.000000</td>\n",
              "      <td>11.000000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 05:00:00</th>\n",
              "      <td>61.666667</td>\n",
              "      <td>12.333333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                       WindDir  Wind Speed(km/h)     lat      long\n",
              "datetime                                                          \n",
              "2016-03-03 01:00:00  45.000000         13.000000  13.754  100.5014\n",
              "2016-03-03 02:00:00  50.000000         12.333333  13.754  100.5014\n",
              "2016-03-03 03:00:00  55.000000         11.666667  13.754  100.5014\n",
              "2016-03-03 04:00:00  60.000000         11.000000  13.754  100.5014\n",
              "2016-03-03 05:00:00  61.666667         12.333333  13.754  100.5014"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "448HV4TAv4ka",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "f6c51356-48aa-441c-94d3-7d2544c822ac"
      },
      "source": [
        "bkk_wind_test = prepareTempAndWind('Bangkok/Test/3H_wind_Bangkok.csv')\n",
        "bkk_wind_test.head(5)"
      ],
      "execution_count": 39,
      "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>WindDir</th>\n",
              "      <th>Wind Speed(km/h)</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2019-03-18 01:00:00</th>\n",
              "      <td>100.0</td>\n",
              "      <td>16.000000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 02:00:00</th>\n",
              "      <td>105.0</td>\n",
              "      <td>16.000000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 03:00:00</th>\n",
              "      <td>110.0</td>\n",
              "      <td>16.000000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 04:00:00</th>\n",
              "      <td>115.0</td>\n",
              "      <td>16.000000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 05:00:00</th>\n",
              "      <td>120.0</td>\n",
              "      <td>16.333333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     WindDir  Wind Speed(km/h)     lat      long\n",
              "datetime                                                        \n",
              "2019-03-18 01:00:00    100.0         16.000000  13.754  100.5014\n",
              "2019-03-18 02:00:00    105.0         16.000000  13.754  100.5014\n",
              "2019-03-18 03:00:00    110.0         16.000000  13.754  100.5014\n",
              "2019-03-18 04:00:00    115.0         16.000000  13.754  100.5014\n",
              "2019-03-18 05:00:00    120.0         16.333333  13.754  100.5014"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OoY7DslMPYHY"
      },
      "source": [
        "# plt.figure(figsize=(20,6))\n",
        "# plt.plot(bkk_pm25_train['PM2.5'])\n",
        "# plt.plot(bkk_wind_train['WindDir'], 'r')\n",
        "# plt.plot(bkk_wind_train['Wind Speed(km/h)'], 'pink')\n",
        "# plt.show()"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SqZ4eie-Je7h"
      },
      "source": [
        "## Fire data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QHm6mS8Fl5yL",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 235
        },
        "outputId": "66228471-2d20-495c-b477-b94957194924"
      },
      "source": [
        "fire = prepareFire()\n",
        "fire.head(5)"
      ],
      "execution_count": 41,
      "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>hotspot_count_Thailand</th>\n",
              "      <th>hotspot_count_Cambodia</th>\n",
              "      <th>hotspot_count_Lao_PDR</th>\n",
              "      <th>hotspot_count_Myanmar</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2016-01-01 14:00:00</th>\n",
              "      <td>253.0</td>\n",
              "      <td>1861.0</td>\n",
              "      <td>46.0</td>\n",
              "      <td>173.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-01-02 01:00:00</th>\n",
              "      <td>4.0</td>\n",
              "      <td>176.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-01-02 03:00:00</th>\n",
              "      <td>62.0</td>\n",
              "      <td>104.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>33.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-01-02 14:00:00</th>\n",
              "      <td>512.0</td>\n",
              "      <td>1447.0</td>\n",
              "      <td>114.0</td>\n",
              "      <td>137.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-01-03 03:00:00</th>\n",
              "      <td>137.0</td>\n",
              "      <td>92.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>37.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     hotspot_count_Thailand  ...  hotspot_count_Myanmar\n",
              "datetime                                     ...                       \n",
              "2016-01-01 14:00:00                   253.0  ...                  173.0\n",
              "2016-01-02 01:00:00                     4.0  ...                    0.0\n",
              "2016-01-02 03:00:00                    62.0  ...                   33.0\n",
              "2016-01-02 14:00:00                   512.0  ...                  137.0\n",
              "2016-01-03 03:00:00                   137.0  ...                   37.0\n",
              "\n",
              "[5 rows x 4 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yv2onS9TIWIc"
      },
      "source": [
        "## Merge data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ns16cyrhw-OD",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        },
        "outputId": "e43d2a38-a12e-49a5-8634-2b4247f9c0fc"
      },
      "source": [
        "bkk_train = prepareData(bkk_pm25_train, bkk_temp_train, bkk_wind_train, fire)\n",
        "bkk_train.head(5)"
      ],
      "execution_count": 42,
      "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>PM2.5</th>\n",
              "      <th>day_of_year</th>\n",
              "      <th>day_of_week</th>\n",
              "      <th>Temp</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "      <th>WindDir</th>\n",
              "      <th>Wind Speed(km/h)</th>\n",
              "      <th>hotspot_count_Thailand</th>\n",
              "      <th>hotspot_count_Cambodia</th>\n",
              "      <th>hotspot_count_Lao_PDR</th>\n",
              "      <th>hotspot_count_Myanmar</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</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",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2016-03-03 15:00:00</th>\n",
              "      <td>62.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>34.433333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>63.333333</td>\n",
              "      <td>12.000000</td>\n",
              "      <td>214.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>718.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 16:00:00</th>\n",
              "      <td>62.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>34.600000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>60.000000</td>\n",
              "      <td>12.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 17:00:00</th>\n",
              "      <td>55.5</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>32.766667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>63.333333</td>\n",
              "      <td>11.666667</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 18:00:00</th>\n",
              "      <td>55.5</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>30.933333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>66.666667</td>\n",
              "      <td>11.333333</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 19:00:00</th>\n",
              "      <td>47.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>29.100000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>70.000000</td>\n",
              "      <td>11.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     PM2.5  ...  hotspot_count_Myanmar\n",
              "datetime                    ...                       \n",
              "2016-03-03 15:00:00   62.9  ...                  718.0\n",
              "2016-03-03 16:00:00   62.9  ...                    0.0\n",
              "2016-03-03 17:00:00   55.5  ...                    0.0\n",
              "2016-03-03 18:00:00   55.5  ...                    0.0\n",
              "2016-03-03 19:00:00   47.9  ...                    0.0\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Za5e_wuXJkpj"
      },
      "source": [
        "# bkk_train.describe()"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JYNpB3oyIkMj"
      },
      "source": [
        "# plt.figure(figsize=(20,6))\n",
        "# plt.plot(bkk_train)\n",
        "# plt.show()"
      ],
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t9MicomDxN8l",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        },
        "outputId": "24e48182-c319-4ab2-97ce-e9f5dacb8ee4"
      },
      "source": [
        "bkk_test = prepareData(bkk_pm25_test, bkk_temp_test, bkk_wind_test, fire)\n",
        "bkk_test.head(5)"
      ],
      "execution_count": 45,
      "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>PM2.5</th>\n",
              "      <th>day_of_year</th>\n",
              "      <th>day_of_week</th>\n",
              "      <th>Temp</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "      <th>WindDir</th>\n",
              "      <th>Wind Speed(km/h)</th>\n",
              "      <th>hotspot_count_Thailand</th>\n",
              "      <th>hotspot_count_Cambodia</th>\n",
              "      <th>hotspot_count_Lao_PDR</th>\n",
              "      <th>hotspot_count_Myanmar</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</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",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2019-03-18 08:00:00</th>\n",
              "      <td>45.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>30.666667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>126.666667</td>\n",
              "      <td>16.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 09:00:00</th>\n",
              "      <td>46.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>32.033333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>123.333333</td>\n",
              "      <td>15.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 10:00:00</th>\n",
              "      <td>47.7</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>33.400000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>120.000000</td>\n",
              "      <td>14.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 11:00:00</th>\n",
              "      <td>43.7</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>33.766667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>118.333333</td>\n",
              "      <td>14.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 12:00:00</th>\n",
              "      <td>41.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>34.133333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>116.666667</td>\n",
              "      <td>14.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     PM2.5  ...  hotspot_count_Myanmar\n",
              "datetime                    ...                       \n",
              "2019-03-18 08:00:00   45.6  ...                    0.0\n",
              "2019-03-18 09:00:00   46.6  ...                    0.0\n",
              "2019-03-18 10:00:00   47.7  ...                    0.0\n",
              "2019-03-18 11:00:00   43.7  ...                    0.0\n",
              "2019-03-18 12:00:00   41.6  ...                    0.0\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7E0JjaTo2piL"
      },
      "source": [
        "# plt.figure(figsize=(20,6))\n",
        "# plt.plot(bkk_train['PM2.5']/max(bkk_train['PM2.5']))\n",
        "# plt.plot(bkk_train['hotspot_count']/max(bkk_train['hotspot_count']))\n",
        "# plt.show()"
      ],
      "execution_count": 46,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Tbl8kIxun0zj"
      },
      "source": [
        "## Add last year PM2.5 column"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Gx7KXwojGwa5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 442
        },
        "outputId": "59f85bd0-ebdc-468d-ca95-244ac31783ff"
      },
      "source": [
        "bkk_train = addPM25Lastyear(bkk_train)\n",
        "bkk_train.head(5)"
      ],
      "execution_count": 47,
      "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>PM2.5</th>\n",
              "      <th>day_of_year</th>\n",
              "      <th>day_of_week</th>\n",
              "      <th>Temp</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "      <th>WindDir</th>\n",
              "      <th>Wind Speed(km/h)</th>\n",
              "      <th>hotspot_count_Thailand</th>\n",
              "      <th>hotspot_count_Cambodia</th>\n",
              "      <th>hotspot_count_Lao_PDR</th>\n",
              "      <th>hotspot_count_Myanmar</th>\n",
              "      <th>PM2.5_lastyear</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</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",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2016-03-03 15:00:00</th>\n",
              "      <td>62.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>34.433333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>63.333333</td>\n",
              "      <td>12.000000</td>\n",
              "      <td>214.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>718.0</td>\n",
              "      <td>62.9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 16:00:00</th>\n",
              "      <td>62.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>34.600000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>60.000000</td>\n",
              "      <td>12.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>62.9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 17:00:00</th>\n",
              "      <td>55.5</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>32.766667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>63.333333</td>\n",
              "      <td>11.666667</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>55.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 18:00:00</th>\n",
              "      <td>55.5</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>30.933333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>66.666667</td>\n",
              "      <td>11.333333</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>55.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2016-03-03 19:00:00</th>\n",
              "      <td>47.9</td>\n",
              "      <td>63</td>\n",
              "      <td>3</td>\n",
              "      <td>29.100000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>70.000000</td>\n",
              "      <td>11.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>47.9</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     PM2.5  day_of_year  ...  hotspot_count_Myanmar  PM2.5_lastyear\n",
              "datetime                                 ...                                       \n",
              "2016-03-03 15:00:00   62.9           63  ...                  718.0            62.9\n",
              "2016-03-03 16:00:00   62.9           63  ...                    0.0            62.9\n",
              "2016-03-03 17:00:00   55.5           63  ...                    0.0            55.5\n",
              "2016-03-03 18:00:00   55.5           63  ...                    0.0            55.5\n",
              "2016-03-03 19:00:00   47.9           63  ...                    0.0            47.9\n",
              "\n",
              "[5 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 47
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bre8nkZhIKAa"
      },
      "source": [
        "# bkk_train.describe()"
      ],
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NF7p1L1YHGqe",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 827
        },
        "outputId": "615563ef-4f91-4c73-bddc-0ef439ef8a41"
      },
      "source": [
        "bkk_test = addPM25LastyearForTest(bkk_test, bkk_train)\n",
        "bkk_test"
      ],
      "execution_count": 49,
      "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>PM2.5</th>\n",
              "      <th>day_of_year</th>\n",
              "      <th>day_of_week</th>\n",
              "      <th>Temp</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "      <th>WindDir</th>\n",
              "      <th>Wind Speed(km/h)</th>\n",
              "      <th>hotspot_count_Thailand</th>\n",
              "      <th>hotspot_count_Cambodia</th>\n",
              "      <th>hotspot_count_Lao_PDR</th>\n",
              "      <th>hotspot_count_Myanmar</th>\n",
              "      <th>PM2.5_lastyear</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</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",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2019-03-18 08:00:00</th>\n",
              "      <td>45.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>30.666667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>126.666667</td>\n",
              "      <td>16.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>26.40</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 09:00:00</th>\n",
              "      <td>46.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>32.033333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>123.333333</td>\n",
              "      <td>15.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>26.14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 10:00:00</th>\n",
              "      <td>47.7</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>33.400000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>120.000000</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.88</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 11:00:00</th>\n",
              "      <td>43.7</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>33.766667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>118.333333</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.62</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 12:00:00</th>\n",
              "      <td>41.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>34.133333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>116.666667</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.36</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>2020-03-18 16:00:00</th>\n",
              "      <td>23.9</td>\n",
              "      <td>78</td>\n",
              "      <td>2</td>\n",
              "      <td>33.000000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>210.000000</td>\n",
              "      <td>7.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>32.70</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-03-18 17:00:00</th>\n",
              "      <td>24.4</td>\n",
              "      <td>78</td>\n",
              "      <td>2</td>\n",
              "      <td>32.133333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>226.666667</td>\n",
              "      <td>6.666667</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>29.40</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-03-18 18:00:00</th>\n",
              "      <td>24.9</td>\n",
              "      <td>78</td>\n",
              "      <td>2</td>\n",
              "      <td>31.266667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>243.333333</td>\n",
              "      <td>6.333333</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>32.60</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-03-18 19:00:00</th>\n",
              "      <td>27.0</td>\n",
              "      <td>78</td>\n",
              "      <td>2</td>\n",
              "      <td>30.400000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>260.000000</td>\n",
              "      <td>6.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>34.50</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2020-03-18 20:00:00</th>\n",
              "      <td>26.1</td>\n",
              "      <td>78</td>\n",
              "      <td>2</td>\n",
              "      <td>29.966667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>230.000000</td>\n",
              "      <td>6.666667</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>33.00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>8797 rows × 13 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                     PM2.5  day_of_year  ...  hotspot_count_Myanmar  PM2.5_lastyear\n",
              "datetime                                 ...                                       \n",
              "2019-03-18 08:00:00   45.6           77  ...                    0.0           26.40\n",
              "2019-03-18 09:00:00   46.6           77  ...                    0.0           26.14\n",
              "2019-03-18 10:00:00   47.7           77  ...                    0.0           25.88\n",
              "2019-03-18 11:00:00   43.7           77  ...                    0.0           25.62\n",
              "2019-03-18 12:00:00   41.6           77  ...                    0.0           25.36\n",
              "...                    ...          ...  ...                    ...             ...\n",
              "2020-03-18 16:00:00   23.9           78  ...                    0.0           32.70\n",
              "2020-03-18 17:00:00   24.4           78  ...                    0.0           29.40\n",
              "2020-03-18 18:00:00   24.9           78  ...                    0.0           32.60\n",
              "2020-03-18 19:00:00   27.0           78  ...                    0.0           34.50\n",
              "2020-03-18 20:00:00   26.1           78  ...                    0.0           33.00\n",
              "\n",
              "[8797 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wYH_y3OXpq4p"
      },
      "source": [
        "# plt.figure(figsize=(20,6))\n",
        "# plt.plot(bkk_train['PM2.5_lastyear'], 'r')\n",
        "# plt.plot(bkk_train['PM2.5'])\n",
        "# plt.show()"
      ],
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZN7068o9aoe6"
      },
      "source": [
        "# plt.figure(figsize=(20,6))\n",
        "# plt.plot(bkk_test['PM2.5_lastyear'], 'r')\n",
        "# plt.plot(bkk_test['PM2.5'])\n",
        "# plt.show()"
      ],
      "execution_count": 51,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gi28avNtM46_"
      },
      "source": [
        "## Prepare data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fc-nsNvqjhpn"
      },
      "source": [
        "def load_data(df, n_in=1, n_out=72, features=['PM2.5']):\n",
        "    n_features = len(features)\n",
        "    data = df[features].to_numpy()\n",
        "    X = []\n",
        "    y = []\n",
        "    for i in range(len(data)-n_in-n_out):\n",
        "        X.append(data[i: i+n_in+2])\n",
        "        temp = []\n",
        "        for j in range(i+n_in+1, i+n_in+1 + n_out):\n",
        "            temp.append(data[j][0])\n",
        "        y.append(temp)\n",
        "    X = np.array(X)\n",
        "    y = np.array(y)\n",
        "    X = X[:, :-1]\n",
        "    return X, y\n",
        "\n",
        "def load_data_test(df, df_train, n_in=1, n_out=72, features=['PM2.5']):\n",
        "    df_all = df_train.append(df)\n",
        "    X, y = load_data(df_all, n_in, n_out, features=features)\n",
        "    X, y = X[df_train.shape[0]-n_in+4::6], y[df_train.shape[0]-n_in+4::6]\n",
        "    return X, y"
      ],
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L0Bi9-gzZEVX"
      },
      "source": [
        "from sklearn.preprocessing import MinMaxScaler, StandardScaler\n",
        "\n",
        "def normalize(df_train, df_test):\n",
        "    scaler_dict = dict()\n",
        "    for col in [col for col in df_train.columns if col != 'PM2.5' and col != 'PM2.5_lastyear']:\n",
        "        scaler = MinMaxScaler()\n",
        "        df_train[col] = scaler.fit_transform(df_train[col].values.reshape(-1,1))\n",
        "        scaler_dict[col] = scaler\n",
        "    \n",
        "    for col in [col for col in df_test.columns if col != 'PM2.5' and col != 'PM2.5_lastyear']:\n",
        "        df_test[col] = scaler_dict[col].transform(df_test[col].values.reshape(-1,1))\n",
        "      \n",
        "    scaler = MinMaxScaler()\n",
        "    df_train['PM2.5'] = scaler.fit_transform(df_train[['PM2.5']])\n",
        "    df_train['PM2.5_lastyear'] = scaler.transform(df_train[['PM2.5_lastyear']])\n",
        "    df_test['PM2.5'] = scaler.transform(df_test[['PM2.5']])\n",
        "    df_test['PM2.5_lastyear'] = scaler.transform(df_test[['PM2.5_lastyear']])\n",
        "    return df_train, df_test, scaler"
      ],
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8lKKPS1vuMPq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "36265039-b581-4a26-bd41-5a41ac5339ac"
      },
      "source": [
        "features = ['PM2.5', 'PM2.5_lastyear', 'day_of_year', 'day_of_week', 'Temp', \n",
        "            'hotspot_count_Thailand']\n",
        "# features = ['PM2.5', 'PM2.5_lastyear', 'day_of_year',\t'day_of_week', 'Temp', 'Wind Speed(km/h)', 'bright_ti4', 'bright_ti5', 'frp', 'hotspot_count', 'distance', 'direction_diff']\n",
        "prev_hour = 3\n",
        "n_features = len(features)\n",
        "X_train, y_train = load_data(bkk_train, n_in=prev_hour, features=features)\n",
        "print(X_train[0])\n",
        "print(y_train[0])\n",
        "print(X_train.shape, y_train.shape)"
      ],
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[ 62.9         62.9         63.           3.          34.43333333\n",
            "  214.        ]\n",
            " [ 62.9         62.9         63.           3.          34.6\n",
            "    0.        ]\n",
            " [ 55.5         55.5         63.           3.          32.76666667\n",
            "    0.        ]\n",
            " [ 55.5         55.5         63.           3.          30.93333333\n",
            "    0.        ]]\n",
            "[47.9 43.6 28.6 33.6 34.8 31.3 36.7 42.1 38.6 45.5 49.9 44.1 42.1 43.1\n",
            " 44.8 37.5 39.6 46.  43.8 36.6 34.2 35.  30.  25.6 32.1 24.4 32.8 32.9\n",
            " 30.7 29.2 28.1 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.3\n",
            " 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.3 33.8 20.  24.3\n",
            " 26.3 25.9 25.5 28.9 24.  24.1 25.2 23.8 27.1 41.9 41.  39.5 35.3 34.1\n",
            " 30.9 27.6]\n",
            "(26549, 4, 6) (26549, 72)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZDfX4iS36nGP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "32d7a128-e566-4f5b-80d5-15b1dda9cd88"
      },
      "source": [
        "X_test, y_test = load_data_test(bkk_test, bkk_train, n_in=prev_hour, features=features)\n",
        "print(X_test[0])\n",
        "print(y_test[0])"
      ],
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[46.6        26.14       77.          0.         32.03333333  0.        ]\n",
            " [47.7        25.88       77.          0.         33.4         0.        ]\n",
            " [43.7        25.62       77.          0.         33.76666667  0.        ]\n",
            " [41.6        25.36       77.          0.         34.13333333  0.        ]]\n",
            "[42.1 42.1 40.  39.8 43.  40.4 38.  36.6 34.8 32.7 33.9 33.1 33.1 32.2\n",
            " 30.8 28.9 30.8 31.3 32.9 36.1 37.9 40.8 35.6 31.9 31.6 30.3 30.6 32.7\n",
            " 29.4 32.6 34.5 33.  33.  32.7 33.5 34.9 33.5 33.2 32.6 31.1 29.9 30.1\n",
            " 29.8 32.8 33.3 35.1 36.5 32.5 34.3 33.1 32.1 30.7 28.6 30.  26.4 26.6\n",
            "  0.  28.8  0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.   0.\n",
            "  0.   0. ]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 572
        },
        "id": "6GJY-dXs4eq3",
        "outputId": "e7dcefd7-8e03-4dd7-8d42-6c2a74d3aacb"
      },
      "source": [
        "bkk_test.head(7)"
      ],
      "execution_count": 65,
      "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>PM2.5</th>\n",
              "      <th>day_of_year</th>\n",
              "      <th>day_of_week</th>\n",
              "      <th>Temp</th>\n",
              "      <th>lat</th>\n",
              "      <th>long</th>\n",
              "      <th>WindDir</th>\n",
              "      <th>Wind Speed(km/h)</th>\n",
              "      <th>hotspot_count_Thailand</th>\n",
              "      <th>hotspot_count_Cambodia</th>\n",
              "      <th>hotspot_count_Lao_PDR</th>\n",
              "      <th>hotspot_count_Myanmar</th>\n",
              "      <th>PM2.5_lastyear</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>datetime</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",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2019-03-18 08:00:00</th>\n",
              "      <td>45.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>30.666667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>126.666667</td>\n",
              "      <td>16.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>26.40</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 09:00:00</th>\n",
              "      <td>46.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>32.033333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>123.333333</td>\n",
              "      <td>15.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>26.14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 10:00:00</th>\n",
              "      <td>47.7</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>33.400000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>120.000000</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.88</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 11:00:00</th>\n",
              "      <td>43.7</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>33.766667</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>118.333333</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.62</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 12:00:00</th>\n",
              "      <td>41.6</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>34.133333</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>116.666667</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>25.36</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 13:00:00</th>\n",
              "      <td>42.1</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>34.500000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>115.000000</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>79.0</td>\n",
              "      <td>582.0</td>\n",
              "      <td>939.0</td>\n",
              "      <td>100.0</td>\n",
              "      <td>25.10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2019-03-18 14:00:00</th>\n",
              "      <td>42.1</td>\n",
              "      <td>77</td>\n",
              "      <td>0</td>\n",
              "      <td>34.500000</td>\n",
              "      <td>13.754</td>\n",
              "      <td>100.5014</td>\n",
              "      <td>113.333333</td>\n",
              "      <td>13.666667</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>24.84</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     PM2.5  day_of_year  ...  hotspot_count_Myanmar  PM2.5_lastyear\n",
              "datetime                                 ...                                       \n",
              "2019-03-18 08:00:00   45.6           77  ...                    0.0           26.40\n",
              "2019-03-18 09:00:00   46.6           77  ...                    0.0           26.14\n",
              "2019-03-18 10:00:00   47.7           77  ...                    0.0           25.88\n",
              "2019-03-18 11:00:00   43.7           77  ...                    0.0           25.62\n",
              "2019-03-18 12:00:00   41.6           77  ...                    0.0           25.36\n",
              "2019-03-18 13:00:00   42.1           77  ...                  100.0           25.10\n",
              "2019-03-18 14:00:00   42.1           77  ...                    0.0           24.84\n",
              "\n",
              "[7 rows x 13 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 65
        }
      ]
    }
  ]
}