viz_AQI.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualize Air Pollution in Bangkok"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Background\n",
    "\n",
    "Air pollution is a pollution threat in many Asian countries. Being informed about the current outdoor air quality helps you protect yourself. Recently, Thai people become more concern about local air pollution due to the high level of pollution in Bangkok in winter of 2019. Air polution is reported as air quality index (AQI). The high AQI mean bad air pollution. The major pollutant was the small size particle pollution (PM2.5). Still there are a lot of misconceptions and debates about the causes of the pollution and degree of hazard of particles. \n",
    "\n",
    "This notebook tried a way to answer answer some of the questions about air pollution through data visualization. Moreover, it includes how to generate these vizs are generated so that the same method can be applied to your local air quality measurement. I was inspired by the work from  a Facebook post by [Data-addict](https://scontent-sjc3-1.xx.fbcdn.net/v/t1.0-9/34398114_426447444494150_7618108699801812992_n.png?_nc_cat=108&_nc_ht=scontent-sjc3-1.xx&oh=0ac2b42de652b994d8f1994ecb835418&oe=5CF14B7F)\n",
    "\n",
    "I wrote the blog post about this result on [Medium.com](https://towardsdatascience.com/identifying-the-sources-of-winter-air-pollution-in-bangkok-part-i-d4392ea608dc)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import sys\n",
    "sys.path.append(r'C:\\Users\\Benny\\fastai\\old')\n",
    "from pathlib import Path\n",
    "from fastai.imports import *\n",
    "import seaborn as sns\n",
    "\n",
    "import statsmodels.api as sm\n",
    "import datetime \n",
    "import matplotlib.dates as mdates\n",
    "import matplotlib as mpl\n",
    "from matplotlib.gridspec import GridSpec\n",
    "\n",
    "from statsmodels.sandbox.regression.predstd import wls_prediction_std\n",
    "from mpl_toolkits.axes_grid1.axes_divider import make_axes_locatable\n",
    "from mpl_toolkits.axes_grid1.colorbar import colorbar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import Data\n",
    "\n",
    "## Import google trend data\n",
    "\n",
    "This is the data about search frquency of the keywords 'AQI' and 'pm2.5'. \n",
    "1. Convert the date columns to date-time data type\n",
    "2. Convert the trend to integer. \n",
    "3. Quick plot to check the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load google trend data on the search keyword 'aqi' and 'pm2.5'\n",
    "trend_aqi = pd.read_csv('data/google_trend_aqi.csv', infer_datetime_format=True,\n",
    "                       skiprows=3, names= ['date','aqi_trend'])\n",
    "trend_pm25 = pd.read_csv('data/google_trend_pm25.csv', infer_datetime_format=True,\n",
    "                       skiprows=3, names= ['date','pm25_trend'])\n",
    "\n",
    "# combine into one df\n",
    "trend = trend_aqi.merge(trend_pm25, on=['date'], how ='outer')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "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>date</th>\n",
       "      <th>aqi_trend</th>\n",
       "      <th>pm25_trend</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2014-02-23</td>\n",
       "      <td>0</td>\n",
       "      <td>&lt;1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2014-03-02</td>\n",
       "      <td>&lt;1</td>\n",
       "      <td>&lt;1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2014-03-09</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2014-03-16</td>\n",
       "      <td>1</td>\n",
       "      <td>&lt;1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2014-03-23</td>\n",
       "      <td>1</td>\n",
       "      <td>&lt;1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         date aqi_trend pm25_trend\n",
       "0  2014-02-23         0         <1\n",
       "1  2014-03-02        <1         <1\n",
       "2  2014-03-09         1          0\n",
       "3  2014-03-16         1         <1\n",
       "4  2014-03-23         1         <1"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trend.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "# change the date column to datetime type\n",
    "trend['date'] =  pd.to_datetime(trend['date'],yearfirst=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "# replace <1 string to 0 \n",
    "trend['aqi_trend'] = trend['aqi_trend'].str.replace('<1','0')\n",
    "trend['pm25_trend'] = trend['pm25_trend'].str.replace('<1','0')\n",
    "# convert google trend data to integer\n",
    "trend[['aqi_trend', 'pm25_trend']] = trend[[\n",
    "    'aqi_trend', 'pm25_trend']].astype(np.int32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# save the data for later use\n",
    "trend.to_csv('data/google_trend_all.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load google trend data\n",
    "trend = pd.read_csv('data/google_trend_all.csv',infer_datetime_format=True)\n",
    "trend['date'] = pd.to_datetime(trend['date'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As mentioned above, more people are unaware of the air pollution problem in the past, but there are an increased interests. How do I know that ? There are surge of Google search for the words 'PM2.5' and 'AQI' in Thailand around 2019. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Google Search Frequency in Thailand')"
      ]
     },
     "execution_count": 261,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiEAAADgCAYAAADVPLa/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3Xd8nWX9//HXJ2clbZKme9IB3RSoUNkoUmZlyfBbRKAM0a/4A7+4EMSBgymK4kJQUKGoCILKHmXKaEuhLZ10pjNNmiZpkjOv3x/3nfRkr5OclL6fj0ceOece133d93XG53yu675vc84hIiIi0tNysl0BERER2TcpCBEREZGsUBAiIiIiWaEgRERERLJCQYiIiIhkhYIQERERyQoFISIdYGbfN7O/ZLseXWFmzszGZ7seexsz+62Z3dgN5c4xs9c6ue5xZrYi7fk6Mzsxc7WrL3es/7oJZrps2bcpCJG9kpnNNrO3zGy3mW33H3/ZzCzbdWuJmR1rZm+Y2S4zKzOz183s49muV2v8oCtuZlVpf9/Mdr2ywTn3JefcDzu6npktTTt2STOrTXt+fRfr9KpzblJXyhDJJgUhstcxs68BdwG3A8OAocCXgGOAcBar1iIzKwT+DfwSGACMBH4ARLthW5n+tfpX51x+2t9tzWzTzEyfJ81wzh1Yd+yAV4GvpB3Ln2S7fiLZpA8N2auYWT/gJuDLzrlHnHOVzvOuc+5C51y0bjkz+5OZlZjZejP7Tt2XpJnl+M/X+1mUP/nl1m3jYn9eqZnd2FqK28yO9LMb5Wb2npkd30LVJwI45+Y655LOuRrn3LPOuffTyrrMzJaZ2U4ze8bMxqTNu8vMNppZhZktMLPj0uZ938weMbO/mFkFMMfMAmZ2vZl9aGaV/jr7pdXnRDNb5W/rV53JIJnZPDP7sZm9DlQD+/vH/T4z22Jmm8zsR2YW8JcPmNkdZrbDzNaY2VXpKf7Gx7lx11drx9qvyw/97FKlmT1rZoPS5h+btu5G87pAPm5m29KDNjM718wWtbC/95vZj/zHx5tZsZl9zX8NbTGzSzt6DBuVf4ffHmvN7LS06Zf6r4tK/7h9MW3e8WZW3EJ5h5vZf/193mJmd5tZOG2+M7MvNfc6aNxWwKe7sm8iLVEQInubo4AI8Hgby/0S6AfsD3wSuBio+5KY4/99yp+fD9wNYGZTgV8DFwLD/TJGNrcBMxsJ/Af4EV524+vAP8xscDOLrwSSZvaAmZ1mZv0blXU2cD1wDjAY7xfz3LRF3gGm+9t5CPi7meWmzT8LeAQoAh4ErgUuAGYBhcBleIFCndOBjwOHAJ8FTmluH9vhIuBKoABYDzwAJIDxwMeAk4Er/GW/4G/3Y8AM4Lz2bqSdx/pzeG08BC8j9nV/3dHAU3ivicF4x3GRc+4doBQ4Ka2MzwN/bme1hrHn9XE58KvG7doBRwArgEHAbcB9aYHhdrzjVujv38/M7NB2lJkE/s8v8yhgJvDlRsu09DrodFuJdIhzTn/622v+8L4ktjaa9gZQDtQAnwACeN0cU9OW+SIwz3/8Al4mpW7eJCAOBIHvAnPT5vUBYsCJ/vPvA3/xH38L+HOjujwDXNJC3acA9wPFeF/UTwBD/XlPAZenLZuDFzSMaaGsncAhaXV6pdH8FcBZLazrgGPTnv8NuK6FZb/v73952t8If9484Ka0ZYf6xz0vbdoFwEv+4xeBL6XNO9mvS9B/vq7uOHf0WPt1+U7avC8DT/uPvw081sL+fQt40H88wD/mw1tY9n7gR/7j4/3XWzBt/nbgyDZev/OAKxpNmwOsbvSac8CwFsr4J3BNWj2K0+Y1OIaN1vtq+nFo7XXQVlvpT3+Z+lMmRPY2pcCg9BS6c+5o51yRPy8H75dfGO+XeZ317MlojGhmXhDvS3QEsDGt7Gq/3OaMAc73093lZlYOHIuXQWnCObfMOTfHOTcKmOZv6+dpZd2VVk4ZYHV19tP+y8wb1FqO9wt8UFrxG2loP+DDFuoNsDXtcTVeNqglf3POFaX9bW5hu2OAELAlbT9+h5eZgEbHloZt0Jb2HOuW9qm1Y/EX4Awzy8fLBLzqnNvSzjqVOucSLWyzo+rr7r/mqCvLz5y9ad5g5nK87NagZspowMwmmtm/zWyred10P2lmvZaOWVfaSqTdFITI3ua/eL+2z2plmR14mY0xadNGA5v8x5ubmZcAtgFbgFF1M8wsDxjYwnY24v06T/+C7uucu6WtnXDOLcf7ZT0trawvNiorzzn3hnnjP76F9yXZ3w+4duEFKfVFNlO3A9qqRwakb3cjXtsMStuHQufcgf78LXgBQZ3RjcrajZcFqDOsUdmdOta0ciycc5vwXlOfwetaam9XTI8wswjwD+AOvKxZEfAkDdu+Jb8BlgMTnHOFeN197R3701ZbiWSEghDZqzjnyvHOKvm1mZ1nZvnmDTSdDvT1l0nipZZ/bGYF5g3wvBbvVy94Yy3+z8zG+b+Af4J3BkgCb1zFGWZ2tD+I7we0/MFd9yv6FH8gX64/UHBU4wXNbLKfzRjlP98Pr6viTX+R3wLfNrMD/fn9zOx8f14BXpBUAgTN7Lt44wNacy/wQzObYJ6DzaylYCoj/AzCs8BPzazQb5cDzOyT/iJ/A642s1H+2InrGhWxCJhtZiEzazwOod3HuhkP4g3E/ayZBc1soP96qfMn4JvAQcBjHd/zbhXGGwNVAiT8Aasnt3PdAqACqDKzycD/dmC7bbWVSEYoCJG9jvNOEb0W74tjO14G43d42YI3/MX+H94v6zXAa3iDOf/gz/sD3i/eV4C1QK2/PM65pf7jh/F+DVb622hyKq1zbiNeRuZ6vC+JjcA3aP59VYk3+PAtM9uNF3wsAb7ml/UYcCvwsJ86XwLUnSHxDN6YkZV4afFamna/NHYn3hfJs3hfRPcBeW2skwkX431xfoA3buUR9nSZ/B5vX94DFgKPNlr3RryMxU684O+huhkdPNYNOOc24HVhfA2vm2sR3kDMOo/hZcYec87tbvee9gDnXCVwNV5b7sQbfPtEO1f/ur98Jd6x/2sHNt1WW4lkhDnXOIsrInX8TEk5Xkp7bbbr81FiZmPxgsBQo7EV2ajLh3jdYc9nsx4i+xplQkQaMbMzzKyPmfXF64tfjHfWgXwEmdm5eGNbXsx2XUT2NboPgEhTZ+F11xgwH5jtlDL8SDKzecBU4CLnXCrL1RHZ56g7RkRERLJC3TEiIiKSFQpCREREJCt6xZiQQYMGubFjx2a7GiIiIpIBCxYs2OGca+4+Wg30iiBk7NixzJ8/P9vVEBERkQwws3Zd6l/dMSIiIpIVCkJEREQkK9oMQszsD2a23cyWpE0bYGbPmdkq/39/f7qZ2S/MbLWZvW9mh3Zn5UVERGTv1Z4xIfcDd+Pd5KnOdcALzrlbzOw6//m38O51McH/OwLvLo5HdKZi8Xic4uJiamtrO7P6R0Zubi6jRo0iFApluyoiIiIZ1WYQ4px7xb/HQ7qzgOP9xw8A8/CCkLOAP/lXl3zTzIrMbLh/d80OKS4upqCggLFjx2LW3rtPf7Q45ygtLaW4uJhx48ZluzoiIiINJBMJ3v7zDRz4mW9SWNTxG3V3dkzI0LrAwv8/xJ8+koZ39yz2pzVhZlea2Xwzm19SUtJkfm1tLQMHDtxnAxAAM2PgwIH7fDZIRER6p3XLF3DU+t+y8vXO3Wg50wNTm4sYmr0uvHPuHufcDOfcjMGDmz+VeF8OQOroGIiISG8Vq60GIBXt3I/lzgYh28xsOID/f7s/vRjYL225UcDmTm6jV3jssccwM5YvX14/benSpZxwwglMnDiRAw44gO9973ukUt69r+6//36+8pWvZKu6IiIiPSbmBx/JeE2n1u9sEPIEcIn/+BLg8bTpF/tnyRwJ7OrMeJDeZO7cuRx77LE8/PDDANTU1HDmmWdy3XXXsXLlShYvXszbb7/NXXfdleWaioiI9Ky4H4Sk4tFOrd+eU3TnAv8FJplZsZldDtwCnGRmq4CT/OcATwJrgNXA74Evd6pWvURVVRWvv/469913X30Q8tBDD3HMMcdw8sknA9CnTx/uvvtubr/99mxWVUREpMfFY14GxCU61x3TnrNjLmhh1sxmlnXAVZ2qSSt+8K+lfLC5IqNlTh1RyPfOOLDVZf75z39y6qmnMnHiRAYMGMDChQtZunQphx12WIPlDjjgAGpqaigvL89oHUVERHqzRMzLgLh4z44J2SfMnTuX2bNnAzB79mzmzp2Lc67ZwaJe/CUiIrLvSNZ1wyQ61x3TK25g15a2MhbdobS0lBdffJElS5ZgZiSTScyMG2+8kVdffbXBsmvWrGHQoEEUFRX1eD1FRESyJRHzMyCd7I5RJqQFjzzyCBdffDHr169n3bp1bNy4kXHjxjF58mRee+01nn/+ecAbqHr11Vfzgx/8IMs1FhER6VkpPwNincyEKAhpwdy5c/nMZz7TYNq5557LQw89xBNPPMGPf/xjJk6cyKBBgzjmmGO48MILs1RTERGR7KjrjrHkR7g7JhvmzZvXZNrVV19d//ill14CvMGr1157LZ/73OcYM2YMc+bMYc6cOT1USxERkeypy4TkpJQJyYqzzz6bNWvWMGbMmGxXRUREpGf5mZBAJzMhCkJERESkU5yfCQmkYp1aX0GIiIiIdIpLesGHghARERHpWX4QEtSYEBEREelJlvCCkJBTJkRERER6UI7fDRNUEJJ5gUCA6dOnM23aNM4//3yqq6sBMDMuuuii+uUSiQSDBw/m9NNPB+DBBx/k4IMP5uCDD+boo4/mvffea7b8OXPmMG7cOKZPn8706dNZtGhR9++UiIhIhpjfHRMm3qnblygIaUVeXh6LFi1iyZIlhMNhfvvb3wLQt29flixZQk2Nd/fA5557jpEjR9avN27cOF5++WXef/99brzxRq688soWt3H77bezaNEiFi1axPTp07t3h0RERDIox+0JQuJJBSHd5rjjjmP16tX1z0877TT+85//AN7VVS+4YM/Nho8++mj69+8PwJFHHklxcXHPVlZERKQHBFJxACLEqYknO7z+3nHF1Keug62LM1vmsIPgtFvatWgikeCpp57i1FNPrZ82e/ZsbrrpJk4//XTef/99LrvssiY3tgO47777OO2001os+4YbbuCmm25i5syZ3HLLLUQikY7vi4iISA9LpZx3am7AC0Iq40nIC3WoDGVCWlFTU8P06dOZMWMGo0eP5vLLL6+fd/DBB7Nu3Trmzp3LrFmzml3/pZde4r777uPWW29tdv7NN9/M8uXLeeeddygrK2txORERkd4mmkgRIgH4mZBYosNl7B2ZkHZmLDKtbkxIS84880y+/vWvM2/ePEpLSxvMe//997niiit46qmnGDhwYLPrDx8+HIBIJMKll17KHXfckbnKi4iIdKOaeJKweYFHjjlqorVAfofK2DuCkF7qsssuo1+/fhx00EENbni3YcMGzjnnHP785z8zceLEFtffsmULw4cPxznHP//5T6ZNm9YDtRYREem66liCMPH657Hamg6XoSCkC0aNGsU111zTZPpNN91EaWkpX/7ylwEIBoPMnz8fgFmzZnHvvfcyYsQILrzwQkpKSnDOMX369Pqzb0RERHq72niSCHu6YKI11R0uQ0FIK6qqqto9/fjjj+f4448H4N577+Xee+9tdt0nn3yy/vGLL77Y9UqKiIhkQXUsSZ+0ICQe63gmpEsDU83s/8xsqZktMbO5ZpZrZuPM7C0zW2VmfzWzcFe2ISIiIr1PTSxJmDgpP5SI13Y8E9LpIMTMRgJXAzOcc9OAADAbuBX4mXNuArATuLzlUkRERGRvVDcwNRHyBqPGoz2cCcHrzskzsyDQB9gCnAA84s9/ADi7i9sQERGRXqY+ExIuACAZq+1wGZ0OQpxzm4A7gA14wccuYAFQ7pyr6yQqBkY2X0K7ttHZVT8ydAxERKQ3qoknCZHERQoBSPTkmBAz6w+cBYwDRgB9geYuDdrst6iZXWlm881sfklJSZP5ubm5lJaW7tNfws45SktLyc3NzXZVREREGqj2MyGW6wUhyU4EIV05O+ZEYK1zrgTAzB4FjgaKzCzoZ0NGAZubW9k5dw9wD8CMGTOaRBqjRo2iuLiY5gKUfUlubi6jRo3KdjVEREQaqI0liFiCeJ4XhKTiHe+O6UoQsgE40sz6ADXATGA+8BJwHvAwcAnweGcKD4VCjBs3rgvVExERke4SjXpBRyCvH9C5TEhXxoS8hTcAdSGw2C/rHuBbwLVmthoYCNzX2W2IiIhI7xTzB6Lm5HpBSE9nQnDOfQ/4XqPJa4DDu1KuiIiI9G5xPxOCPybEJWIdLkN30RUREZEOS9Sdkhsp8Cf04Cm6IiIisu+qz4T4p+gqCBEREZEekYhHvQf+mBBLRDtchoIQERER6bD6K6SG8kgQwJIKQkRERKQHJOsyH4EwCQsrCBEREZGekarLhATCJHLCBFIKQkRERKQHpOpOyQ1GSFiYnKRO0RUREZEekKrvjomQDEQIKhMiIiIiPcHVByEhkjkRAillQkRERKQHWFp3TCoQJuRipFJN7kfbKgUhIiIi0i5/n7+RNz7cQSKZwuoyH4EwLhAhQpxoItWh8hSEiIiISLv8/PlV/Oql1dTEk4Qt4U0MRrwgxOLUxJMdKk9BiIiIiLRLNJFkyaYKamJJwsS9iYEwBCNEiFHbwSCkS3fRFRERkX1HNJGisjbBqu1VhPEzIYEwBHOJoEyIiIiIdJO6MR9vrS3bE4QEI34mJE5NTEGIiIiIZJhzjpgfhMxfV0bY9nTHWCiXsCWIJhSEiIiISIbFknvOfHl3QzlhEqRyQmBGTiiXCDFqYjo7RkRERDIs/fTbmrg3MNUFwgDkhPI0JkRERES6R11XzMSh+QCESHiDUoFA2BuY2tGzYxSEiIiISJvqMiGHjekPQJgELhABvCAkaClqox27f4yCEBEREWlT1M9yHDKqiECOEbY45mdCguE8AOLRmg6V2aUgxMyKzOwRM1tuZsvM7CgzG2Bmz5nZKv9//65sQ0RERLKvbmBqYV6ICUPyiZDAgo2DkOoOldnVTMhdwNPOucnAIcAy4DrgBefcBOAF/7mIiIjsxaJxLwiJBHOYNrIfIRJYyOuOCUX6AJCI1naozE4HIWZWCHwCuA/AORdzzpUDZwEP+Is9AJzd2W2IiIhI71A3JiQczOHio8YwaVAY88eE5IRyAYjHeq47Zn+gBPijmb1rZveaWV9gqHNuC4D/f0hzK5vZlWY238zml5SUdKEaIiIi0t3qzo6JBAMcPKqIsUXB+rNjCHrBSKoHg5AgcCjwG+fcx4DddKDrxTl3j3NuhnNuxuDBg7tQDREREeludVdDjQT90CERg2BdEOJlQpKxHuqOAYqBYufcW/7zR/CCkm1mNhzA/7+9C9sQERGRXiC9OwaAZAz87pj6TEi8hzIhzrmtwEYzm+RPmgl8ADwBXOJPuwR4vLPbEBERkd5hT3dMWhASrAtCvExIooOZkGAX6/T/gAfNLAysAS7FC2z+ZmaXAxuA87u4DREREcmy+u6YUMCbkIhCIOQ99oORWG3HTtHtUhDinFsEzGhm1syulCsiIiK9S10mJByoy4RE07pjvExItINBiK6YKiIiIm2qGxMSCdUFIfG0galeMNKjV0wVERGRfUO0cSYk0TQTkojVkky5dpepIERERETaFE2kONDWEdm9yZuQjKVdJ8QLQsLEKa+OtbtMBSEiIiLSpmgiyS/Dv8ReuMmbkIg26Y6JEGenghARERHJpGg8RQE1sHM9OAepeJPrhESIU1qlIEREREQyKJZMEbE4VGzyumJgTyYkJ4izHCIWUyZEREREMisaTxEhDpVbIO6fils3JsQMF4gQIU7Z7ni7y1QQIiIiIm2KxhNEiIFLQfkGb2JddwxgoVyNCREREZHMSyWie56UrfH+13XHABbMJT+Q0JgQERERyaxUvJkgJC0TQjBCQTCpTIiIiIhklku/Q27ZWu9/3b1jAIK5FAQSlO1WECIiIiKZlEi7Q25dEBJMy4RECinMqVEQIiIiIpmVSqQFF811x+T1px+7OxSEdOkuuiIiIrJvsPRMSOVm73/awFTyishPVbKzVpkQERERySCrOzsmf9ieiYH0IKQ/fZIVVMeS7S5TQYiIiIi0LelnQgYesGdaendMbhGRZBU5pNpdpIIQERERaVNO3aXaB+y/Z2KwYSYEoJDd7S8zExUTERGRj7acpN8d01ImxA9Ciqyq/WVmomIiIiLy0ZaT8oOQAelBSNp1QvKKAOinTIiIiIhkSiKZIpTyu2PSMyHB5jIhCkJEREQkQ2LJFBHz746bNwByvaxH44GpAP16Mggxs4CZvWtm//afjzOzt8xslZn91czCbZUhIiIivVc0niKCH4QEI9BvlP+46cDUEeEa2isTmZBrgGVpz28FfuacmwDsBC7PwDZEREQkS2LJ9CAkFwpHeo8DDS9WBjA01ENBiJmNAj4N3Os/N+AE4BF/kQeAs7uyDREREcmuppmQZoKQQAjC+QwOtj8I6epl238OfBMo8J8PBMqdcwn/eTEwsrkVzexK4EqA0aNHd7EaIiIi0l1iySQRi5GyEDk5AZhyBiRjYNZwwbz+DEj1wJgQMzsd2O6cW5A+uZlFXXPrO+fucc7NcM7NGDx4cGerISIiIt2s1s+EpOoyHwecAGf9qumCuUUUdeAU3a5kQo4BzjSzWUAuUIiXGSkys6CfDRkFbO7CNkRERCTLoom6ICTS+oJ5RRRUV7S73E5nQpxz33bOjXLOjQVmAy865y4EXgLO8xe7BHi8s9sQERGR7Iv5QYhrMwjpT6HL7hVTvwVca2ar8caI3NcN2xAREZEeEk0kCVt7gpAi8lOV7S63qwNTAXDOzQPm+Y/XAIdnolwRERHJvrrumAZXSG1OXn8C0fJ2l6srpoqIiEirYvVBSG7rC+YWeWfNtJOCEBEREWlVRzIhHaEgRERERFoVTSSJWBwLtZEJURAiIiIimeR1x8TaEYQUdahcBSEiIiLSqrrumBxlQkRERKQnxdobhOQqEyIiIiIZVDcmRJkQERER6VHReIrc9pyiGykAC7S7XAUhIiIi0qpYMkXE2nGKrlmHBqcqCBEREZFWRWNJwu3JhECHumQUhIiIiEir4okYAVJtZ0KgQ4NTFYSIiIhIq1yi1nugTIiIiIj0pFRMQYiIiIhkgYtHvQft6Y7RwFQRERHJFJeoC0KUCREREZGelKjx/rcnE1I4ot3FKggRERGR1nUkEzLggHYXqyBEREREWmVJPwgJhNteeKCCEBEREckQ60gmpGB4u8tVECIiIiKtqs+EtGdMiFm7y+10EGJm+5nZS2a2zMyWmtk1/vQBZvacma3y/3fslnoiIiLSq+SkOpAJ6Ui5XVg3AXzNOTcFOBK4ysymAtcBLzjnJgAv+M9FRERkL5XTkUxIR8rt7IrOuS3OuYX+40pgGTASOAt4wF/sAeDsrlZSREREssM5RyAV8570okxIPTMbC3wMeAsY6pzbAl6gAgzJxDZERESk58WSKSL00iDEzPKBfwBfdc5VdGC9K81svpnNLykp6Wo1REREpBtEEykixL0nvaU7BsDMQngByIPOuUf9ydvMbLg/fziwvbl1nXP3OOdmOOdmDB48uCvVEBERkW4SaxCE9JJMiJkZcB+wzDl3Z9qsJ4BL/MeXAI93vnoiIiKSTdFEiojFSVkAAsGMlt2V0o4BLgIWm9kif9r1wC3A38zscmADcH7XqigiIiLZEo0niRAnlRPO+MXFOh2EOOdeA1q6IsnMzpYrIiIivUfdwNRUILPjQUBXTBUREZFW1MSShEkoCBEREZGeta2ilojFsVBmB6WCghAREekhzjmiiWS2qyEdVLyzhggxguG8jJetIERERLrdsi0VfObXb3DcrS8pENnLbCqvoW9OghxlQkREZG/zh9fWcsYvX+ODLRVsr4zy7obyJsusKaniVy+txjmXhRpKazbtrKEwlMQyfI0QUBAiIiLd7O6XVnPomP68cO0nCeQYr6/e0WSZuW9v4PZnVrB0c7svvC09ZFN5DfmBZMavlgoKQkREpBvtqolTtjvGiVOGsN+APhw8ql+zQcjyrZUAPLN0a09XUdpQvLOGPjkJBSEiIrJ3WbdjNwBjBvYF4Njxg3iveBeVtfEGy61QENIrVUUT7KqJk6cgRERE9jbrSr0gZNwgLwg5+oBBJFOOt9aU1S+zc3eM7ZVR9huQx8ptVaz1AxfJvk07awC8u+hqTIiIiOxN1u2oxgxGD+gDwKFjisgN5fBaWpdMXVfMVcePB5QN6U02lVcDECauTIiIiOxd1pXuZkS/PHJDAQAiwQAfHzugwbiQFVu9wagnTB7CtJGFCkJ6kbpMSDClTIiIiOxl1u7YzdhBfRpMO3b8IFZtr2J7RS0AK7ZV0r9PiMEFEU6ZOox3N5SzzZ8HkEimWLSxnAXrd7K4eBeplE7j7Q4by6qbTCveWUM4kIMlowpCRERk77KudDdj/UGpdY4ZPwiAF5ZvB7zumEnDCjAzTp02DIBHFhTXL/+LF1dz9q9e59zfvMEZd7/G9Y8t7qHa7zveWVfGcbe9xH2vrW0wvbi8hhFFuViiVt0xIiKy9yivjlFeHW8ShBw4opAJQ/L56zsbSaUcK7dWMnlYIQAThhbwqUmD+d3LH7KrOk5JZZR7X13DCZOH8MBlh3PJUWN4+J2NPPTWhmzs0kfWfz8sBeAnTy7jzTWl9dM37axhv6IwpBLKhIiISO9WE0uyqdwbR1B3lsvYQQ2DEDNj9uGjWbSxnOeXbWN3LMmkYQX1879xymQqahP87pUP+dVLq4kmUnzn01P45MTBfPeMA/nExMF874klLNyws+d27CNuwfqdjB3YhzED+/CVhxayZZfXhpvKaxjTL+gtpEyIiIj0Zrc+vZxTfvYKFbXxtNNz+zRZ7pyPjSQcyOEnTy4DaBCETB1RyJmHjOAPr6/lobc28NkZo9h/cD4AgRzjF7OnM6xfLv/7lwWUVEZ7YK8+2lIpx8INOznqgEHcc9EMamJJvvOMs5p0AAAUJUlEQVTYEmrjSUoqo+xX6A0qViZERER6rVTK8eTiLVRFEzy+aDPrdlSTY7DfgKZBSP++YU47aBjrSr3BkBOHFjSYf+1JE0kkHRhcPXNCg3lFfcL87vMz2FUT56oHFxJPplixtZKrHlzIO+vKkI5ZXVJFZW2Cw8b0Z/yQfK46YTwvLN/Ov97bDMCoAj9UUCZERAS2V9ZSXh3LdjWkkXc3lrO9MkooYMx9a4N3em5RHpFgoNnlZ398NAD7DcgjPxJsMG/soL7cePpUfnDmgQzv1/QW8lNHFHLruQfz9roy/ud3/+XTv3iV/yzewg2PLSaps2c6ZMF6r1vrsDH9Abj06HEMLojwo/94WaoR+eYtqEyIiOzrNpZVc/LPXuGC37+lUzV7mWeXbiWYY3z1xIl8sKWCeStK6q+U2pwj9x/AxKH5fGy//s3Ov+TosVxw+OgW1z9r+kguP3YcCzeUc9b0kfzw7Gms3FbF44s2dXlf9iUL1u9kQN8wYwd6Gau8cICrTxjPrhrv0vrD6hJZyoR0zPrS3VzxwHw+0F0ZRT4SauNJvvSXBVTVJli2pYJ/vb8521USn3OOZ5Zu5agDBnLRUWPIDeWwqybOmIFNu2LqmBl//9LR3HLuQZ3e7nc+PYU3rjuBn372EC48fDQHjijkZ8+vJJpI8rf5G5nzx7frB1lK8xau38mho/tjZvXT/ufjo9lvQB6BHGNw0L+MfiDzQUiw7UX2TtWxBF/88wKWb61k+dYK/vWVY+nfN5ztamWVc47Fm3YxZXghoYAXf9bGk3xYUsWBI/pluXbdb2NZNYs37QJgQN8wR4wb0OBNJ5kTS6R4dVUJ0UQKAw4b258hBU1Tuau3V1KQG2JoYdtp3lTKcf1ji1m6uYJ7L57BHc+u4M7nVjLroOH1r2fJnpXbqlhXWs0Vx+1PYW6I0w8ewSMLir3Tc0s/hK3+tT0KhsPoI+rX65cX6tJ2zYwRRV53TU6O8Y1TJjHnj+9w4p0vs7HMCz7+9y8L+esXj2yxWyjbnHO8vbaMaSP70bdRt5RzjvnrdzJ5WAEFuW0fq9p4kldX7SCeTLW6nAEfHzeAHDPW7NjN+YeNhA9fhFrvR3t49JHcft4hvLdmM8Gnvgh5/WHkYZ3ex5Z0SxBiZqcCdwEB4F7n3C3dsZ2WOOf41j8Ws2JbJd88dRI/f24VVz/8LvdfejiBnH33S+eBN9bx/X99wGdnjOLWcw8m5eALf5rPq6t28LuLDuOUA4dlu4rdZkNpNWfc/Vp9ehHg+EmD+eFZ05odNCedt2D9Tr796Pus3FZVP60gEuSbp03mwsNHk+O/B5ds2sW5v3mDQI7x9ZMnccnRY1t8f67aVsm3H13M/PU7+eqJEzhx6lDM4PIH5vP3+cV87oiWU/bSM55ZuhUzOHnqUAA+f+QYHl1YzOF9tsJvz4N42k3pppwJp90GhcMzXo9PThzMMeMH8n7xLm4+5yD65YX48oML+f4TH3DzOZ3PuHSXtTt2c/2ji/nvmlJGFuXxw7MP5ITJ3jHcWFbNDf9cwisrSxhaGOEHZ06rv5hbc95YvYPrH1tcP9i3LYW5QU72P/fPKvsDvPzrPTPDBRw587scWf4mlCyDCx+BgqGd39EWmHOZ7VM1swCwEjgJKAbeAS5wzn3Q0jozZsxw8+fPb7PsZVsquPmp5UweVsBXT5xAn/CeGGr19kp+9J9lbCirJpF0bCir5hunTOKqT43n4bc3cN2jixlZlEcklMPQglyunzWFg0b1Y2NZNT/+zzKK+oT41qmTm82WxJMp7nllDa+sLOHakyZyxP4DO35gsuzttWV87vdvMig/wtaKWn509jS27KrhVy99yLDCXKqiCf551THkR4L85MllBHKMb8+a3Oyv13QL1pdx69Mr2FHV8DS5gkiQq2dOYOaUoeyqjnPbM8vZXhnlhllTGDuoLx9sruDmp5YxdUQh18xs2JZ1dkcT/Pz5lby4fDsO754Tlx0zlvMOG0U0keIXL6zimaVbaekVPDg/wnWnTWbSsALO+fUbbNlVy+8vnkG/vBCvr97BHc+uwDn42skTmXP0WILN/JreURXllqeWUxNPcsOsKfW/uDpj+dYKfvKk9/q9ZuaEJr94MuXllSXc/eIqTps2vMkX+xsf7uCnz65kZ6NBnf3yQlx70kSOmzC41bLLq2Pc9syK+osZ9Q0HuepT4zl12jAqauPc9vRyHnxrA8MLc/nO6VM5YHA+VdEEdz63gtdXl3LYmP7cfM5BDM6PcMbdr5FMOSYNK2DeCu9DtqVjsrGsmr6RIN/59FTOPXQkZoZzjvN++18+2FzB8KLWX6cBMz5z6Ei+cNz+zWZNynbHuO3p5bztn1mRHwly9QlesNOayto4dz63knU7dnPDp6cwfkhBq8u3RzSR5DfzPuStNWV849RJHDq6P5vLa/jxk8vIDQb49qzJDMrfkxbfXlHLT55chgOunzWlzazSuxt2cstTyynx37MThuRzw6ypjE7rNtlVE+enz65ocJO5tmzbVcvBQ0PMnfiyl/U46SbKQ8MoevBkiO2G2Q9BqA+sfApevg0sAIUjmi8sJwiHXQKHXwk5Hc9e1MaTJFOu/vV069PL+c28Dxk7sA85OcYBg/P5zqenMCbtImoVtXHufHYlG8qquX7WFMYPyW9SrnOOf7+/hfvfWMcFh4/m3ENHUhtP8YsXV/FsK59FrSneWUMkmMMXP7E/T7y3mZXbqhgzsA+BHGPTzhqCOcYXP3kATy/ZygdbKhg9oA/BQNNg3TkvoBkzsA83fnpqmz+uqqJxbn9mBW+uKeO0wHx+E7oTpn8ejrrKa695P/EyIwAzvwfHXduh/TKzBc65GW0u1w1ByFHA951zp/jPvw3gnLu5pXWmjB/t/vTT61otd3N5Da+u2kEomENtLEn/vmFOmDyESDCHbRW1zFtRQiSYw0T/XPPhhbmcNHVofbr95ZUl9eesf1hSRVVtgkNGFfHBFi/1lEw58kIBZk4Z0uCDMJVyvLKqhM3lteSFA9TEkhwxbgAHNPMC7a1SzvGf97eQGwrw1RMn8pc317FyWxXJlOOoAwZy0tSh3PnsSkLBHGpiSZKpFM5BKJjDzClDKGwhBbh2x27eXFNKv7xQk4sRbS6vYXtFlKkjCtlYVs3uWJJQjpFyjgNH9OP9TbsI+205wG/LcHDPl0M0keLF5dvZuTvGlOEFREIBdlRFKS6rYfyQvpRXx9lRFWPS8ALyQs1/SK0t2U1FbZyhhblsq6jlC8ftz5ThhfXzy6pjPLqgmKWbKxg1II9jxw8iJ617pqo2zvPLthNNpMjJgRyMmVOGUNSn4916W3bV8srKEkKBHGrjDV+/mbR8ayUL1++sf62OHtCHo8cPJMeMVdureGdtGQPzw00+oIrLqtlRFWPG2P5NTpWsUx1L8oJ/YampwwsJBoyt5bVsrahl8vACNpfXUFmb4LgJg5k1bRiRtHZxzjF/3U4ef28TtfEUA/PDlFXF+MoJExg9II9FxeUsLq7AtfAxXpgb4sQpQ5qkozeV1/Di8u2k2vgcq6yJ82HJbkYU5fKJCYPrszHgBbvPL9tGbTzF1OGFBAJW//o9ZL9+LXZVRhNJnv9gOxW1cSLBAPFkiuMnDW5X11JLEinHvBXb2V4RJS8coDae5OBR/Vi+pZIUDpeCcDCHE6cMIT83REVNnOeXbyfhp96DOd57tqUujvWl1bzx4Q4Kc0OMG9wXnPfjLoXjhElDGZgfpiaW5Pnl26iqTTB1eCGhdr5GI6lazt79d3J3F0OkH8SqYOB4KPsQLvk3jDlqz8KlH8Lrd0G0svnCKjbDxjdhxMfg41d4QUkXpJzj6SVbvcDLee+TZCrFpyYPZVB+mNq499quqE0QCQZIJFN8snFbOnh3406Wbamsf3+NH9KXndVxStv4LGpNQSTIiVOGUpgXIpH0ujE3+DeN6xsOMnPKEPr3CZNMpXhtdWn991hzhhfmcvzkIYTb2T3pnOPdtduYtuQ2wsMmw6VP7Rl86hws+QeUrYVPfB062HWdzSDkPOBU59wV/vOLgCOcc19ptNyVwJUAhw3POWz+lXvPl7qIiDRj0EQ44y4YPBme+y68+xeYdTsc/oWOleMcLH0UnroOdm/vnrrKHvnD4AsvQr+RGSsym0HI+cApjYKQw51z/6+ldQ456ED35ON/a7Xc3GCAgfl7foEmko5tld5dFiPBAIPyO/7rtLImQX5usD7AS6Vga0Vtk19jg/IjDX6xllfH2R1LdHh72dQvL9TgPPxoIkXArEFab3csSZ9QoP54OOcdj5Z+ZeaFAwxoIytQGU3QNxQkJy0wr6xJUJC3py7pbVnHMIYURJpNO1bHkoQCOYSamdecqmiiyTUIGosmUk26lAxjeL/cBsdjW2WUZKr1AV/Nae31m0kFkSCFab+Ca+MpSnd7+9U3HKSoT+sD23ZVx6lq4bVtGMMKcxu0ZZ2qaIK8UJD2jg9tT5t0h5bauaX9qqxJUBGNN53hrze0MLfBPu+oihFNJLtUx8H5uYSDe17blbXesUp/HW7Z5X1O5ZhX90y8Z7dVREmkUs3uV7v1Gw2BtHatrYDcwpaXb0u8Biq3dH79NmyvjBJPNr/PJZUxYsmGbVmYF6Ig7XXb0c+iXit/KIRbPpW6M/aq7pj2jgkRERGR3q+9QUh3nNf2DjDBzMaZWRiYDTzRDdsRERGRvVjG86HOuYSZfQV4Bu8U3T8455ZmejsiIiKyd+uWTlnn3JPAk91RtoiIiHw06DKDIiIikhUKQkRERCQrMn52TKcqYVYJrMjS5vsBu/ah7WZz24OA9l+CMbPUzj1H7bzvbLtOT7f5vtjOvaEOHWnnSc65Ni8h3FtuYLeiPafydAczu8c5d+W+st1sbtvM5qudP/rbVjvvO9tOq0OPtvm+2M69oQ4daWcza9d1N9QdA//ax7ab7W1ni9p536B23jfsi+1cpzfUIWN6S3dM1n45Sc9RO+8b1M77HrX5vqGjmZBsXaysM+7JdgWkR6id9w1q532P2nzf0JF2bteyvSITIiIiIvue3pIJERERkX1MtwQhZrafmb1kZsvMbKmZXeNPH2Bmz5nZKv9/f3/6ZDP7r5lFzezrjcpaZ2aLzWxRe0fbSs/IcDsXmdkjZrbcL++obOyTNJWpdjazSf77uO6vwsy+mq39kpZl+L39f34ZS8xsrpnlZmOfpKkMt/M1fhsv7cj7ulu6Y8xsODDcObfQzAqABcDZwBygzDl3i5ldB/R3zn3LzIYAY/xldjrn7kgrax0wwzmXresOSAsy3M4PAK865+4178aHfZxz5T29T9JUJts5rcwAsAk4wjm3vqf2RdonU21uZiOB14CpzrkaM/sb8KRz7v6e3ytpLIPtPA14GDgciAFPA//rnFvVVh26JRPinNvinFvoP64ElgEjgbOAB/zFHsDbEZxz251z7wDx7qiPdI9MtbOZFQKfAO7zl4spAOk9uun9PBP4UAFI75ThNg8CeWYWBPoAm7u5+tJOGWznKcCbzrlq51wCeBn4THvq0O1jQsxsLPAx4C1gqHNuC3g7DwxpRxEOeNbMFphZVi8SIy3rYjvvD5QAfzSzd83sXjPr243VlU7KwPu5zmxgbqbrJ5nXlTZ3zm0C7gA2AFuAXc65Z7uzvtI5XXxvLwE+YWYDzawPMAvYrz3b7dYgxMzygX8AX3XOVXSymGOcc4cCpwFXmdknMlZByYgMtHMQOBT4jXPuY8Bu4LoMVlEyIEPvZ/zutjOBv2eqbtI9utrm/liCs4BxwAigr5l9PrO1lK7qajs755YBtwLP4XXFvAck2rNutwUhZhbC26kHnXOP+pO3+X1QdX1R29sqxzm32f+/HXgMr89JeokMtXMxUOyce8t//gheUCK9RKbez77TgIXOuW2Zr6lkSoba/ERgrXOuxDkXBx4Fju6uOkvHZfC7+j7n3KHOuU8AZUCb40Gg+86OMbz+/WXOuTvTZj0BXOI/vgR4vI1y+vqDZfDT8yfjpX2kF8hUOzvntgIbzWySP2km8EGGqyudlKl2TnMB6orp1TLY5huAI82sj1/mTLxxB9ILZPK97Q9axcxGA+fQzvd4d50dcyzwKrAYSPmTr8fra/obMBrvxXm+c67MzIYB84FCf/kqYCreHfse89cPAg85536c8QpLp2SqnZ1zFWY2HbgXCANrgEudczt7cn+keRlu5z7ARmB/51y270YqLchwm/8A+B+89Py7wBXOuWhP7o80L8Pt/CowEG/Q6rXOuRfaVQddMVVERESyQVdMFRERkaxQECIiIiJZoSBEREREskJBiIiIiGSFghARERHJCgUhIpJRZvb9xnfYbDT/bDOb2pN1EpHeSUGIiPS0s/GuAyQi+zhdJ0REuszMbgAuxrsQWQneLcF3AVfiXYBuNXARMB34tz9vF3CuX8SvgMFANfAF59zynqy/iGSHghAR6RIzOwy4HzgC78rGC4HfAn90zpX6y/wI2Oac+6WZ3Q/82zn3iD/vBeBLzrlVZnYEcLNz7oSe3xMR6WnBbFdARPZ6xwGPOeeqAczsCX/6ND/4KALygWcar+jfvfNo4O/ebSwAiHR7jUWkV1AQIiKZ0FxK9X7gbOfce2Y2Bzi+mWVygHLn3PTuq5qI9FYamCoiXfUK8Bkzy/Pven2GP70A2OLfKvzCtOUr/Xk45yqAtWZ2Pnh39TSzQ3qu6iKSTRoTIiJdljYwdT1QDHwA7Aa+6U9bDBQ45+aY2THA74EocB7e3Th/AwwHQsDDzrmbenwnRKTHKQgRERGRrFB3jIiIiGSFghARERHJCgUhIiIikhUKQkRERCQrFISIiIhIVigIERERkaxQECIiIiJZoSBEREREsuL/A7iy6FpCPFd3AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 648x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "trend.plot('date', ['aqi_trend', 'pm25_trend'], kind='line',\n",
    "           label=['AQI','PM2.5'], xlim=['2015-01-01', '2019-02-07'],\n",
    "          figsize=(9,3))\n",
    "plt.title('Google Search Frequency in Thailand')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load PM 2.5 data\n",
    "\n",
    "How AQI are calculated? That depends on the law in different countries. They mostly divided into four pollutants source: ground level ozone(O3), particle pollution (PM2.5 and PM10), toxic gas (NO2, CO, SO2). The table below shows different pollutants from various stations in Thailand on February 2, 2019. There are level of CO, NO2, O3, SO2, PM10 and PM2.5. Most level are normal except, high level of PM2.5 in some region. \n",
    "\n",
    "PM2.5 are particle of size smaller than 2.5 micron. PM2.5 concentation higher than 35 ug/m3 is considered moderate and unhealthy for sensitive group with AQI 100 and above 55.4 ug/m3 is consider unhealthy with AQI 150. Since AQI is calculated from the highest level of pollutant aveage over 24 hours([1](https://airnow.gov/index.cfm?action=aqi_brochure.index)), we focus this notebook on PM 2.5 data. \n",
    "\n",
    "What cause high air pollution in Bangkok last winter? The sources of PM2.5 particle can be car engine, power plants, and agricultural waste burning. There are number of theories: still wind, old diesel cars, near by power plants, and agricultural burning from other provinces. Let's try to answer some of these questions through visualization. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![AQI report Thailand](data/aqi_2019_02_19.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The government official website only provides a month historical data. I use the PM 2.5 from Berkeley Earth [website](http://berkeleyearth.lbl.gov/air-quality/maps/cities/Thailand/). Detail about how to use Python to scrap this data can be found in this [notebook](https://github.com/worasom/aqi_thailand/blob/master/webscraping-PM25.ipynb). Also if you are interested in scrapping the data from the government website directly, you may find this [notebook](https://github.com/worasom/aqi_thailand/blob/master/webscraping-AQI.ipynb) useful.\n",
    "\n",
    "The data from Berkeley Earth is a .txt file with header. The header contain geometrical location, city name and region name. City and region actually refer to sub-province and province, accordingly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 255,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "% Country: Thailand\n",
      "\n",
      "% City: Bangkok\n",
      "\n",
      "% City (ASCII): Bangkok\n",
      "\n",
      "% Region: Bangkok\n",
      "\n",
      "% Region (ASCII): Bangkok\n",
      "\n",
      "% Population: 5104476\n",
      "\n",
      "% Latitude: 13.754\n",
      "\n",
      "% Longitude: 100.5014\n",
      "\n",
      "% Time Zone: Asia/Bangkok\n",
      "\n",
      "% Year, Month, Day, UTC Hour, PM2.5, PM10_mask, Retrospective\n",
      "\n",
      "2016\t3\t3\t8\t62.9\t1.00\t0\n",
      "\n",
      "2016\t3\t3\t9\t62.9\t1.00\t0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# inspecting the top of the files \n",
    "with open('data/Bangkok.txt','r') as f:\n",
    "    for i in range(12):\n",
    "        print(f.readline())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The PM10_mask, and Retrospective columns are uniform."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [],
   "source": [
    "#load bangkok data\n",
    "pm25 = pd.read_csv('data/Bangkok.txt', sep='\\t', \n",
    "                   header=None, skiprows=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "#assemble datetime column \n",
    "pm25['Time'] = pd.to_datetime({'year': pm25[0], 'month': pm25[1], 'day':pm25[2],'hour': pm25[3]})\n",
    "#convert to Bangkok time zone and remove the time zone information\n",
    "pm25['Time'] = pm25['Time'].dt.tz_localize('UTC').dt.tz_convert('Asia/Bangkok')\n",
    "pm25['Time'] = pm25['Time'].dt.tz_localize(None)\n",
    "# drop Year, month, day, UTC hours, PM10_mask columns\n",
    "pm25=pm25.drop([0,1,2,3,5,6],axis=1)\n",
    "pm25.columns = ['PM2.5', 'Time']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save the fie\n",
    "pm25[['Time','PM2.5']].to_csv('data/bkk_pm25.csv',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Explore Data and Feature Engineers\n",
    "\n",
    "Data preparation steps such as adding the year, day of month, day of week and the average particle levels. \n",
    "\n",
    "For easy visualization, plot the 24 hours average instead of the direct reading."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "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>Time</th>\n",
       "      <th>PM2.5</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2016-03-03 15:00:00</td>\n",
       "      <td>62.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2016-03-03 16:00:00</td>\n",
       "      <td>62.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-03-03 17:00:00</td>\n",
       "      <td>55.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2016-03-03 18:00:00</td>\n",
       "      <td>55.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2016-03-03 19:00:00</td>\n",
       "      <td>47.9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Time  PM2.5\n",
       "0 2016-03-03 15:00:00   62.9\n",
       "1 2016-03-03 16:00:00   62.9\n",
       "2 2016-03-03 17:00:00   55.5\n",
       "3 2016-03-03 18:00:00   55.5\n",
       "4 2016-03-03 19:00:00   47.9"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load the data\n",
    "bkk = pd.read_csv('data/model/bkk_pm25.csv')\n",
    "bkk['Time'] = pd.to_datetime(bkk['Time'])\n",
    "bkk.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 24153 entries, 0 to 24152\n",
      "Data columns (total 2 columns):\n",
      "Time     24153 non-null datetime64[ns]\n",
      "PM2.5    24153 non-null float64\n",
      "dtypes: datetime64[ns](1), float64(1)\n",
      "memory usage: 377.5 KB\n"
     ]
    }
   ],
   "source": [
    "bkk.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rolling average over 24 hours window\n",
    "bkk['PM2.5avg'] = bkk['PM2.5'].rolling(24).mean()\n",
    "bkk = bkk.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "pm25 = bkk.set_index('Time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# add datetime feature\n",
    "pm25['Year'] = pm25.index.year\n",
    "pm25['Month'] = pm25.index.strftime('%B')\n",
    "pm25['DayofMonth'] = pm25.index.day\n",
    "pm25['DayofWeek'] = pm25.index.strftime('%A')\n",
    "pm25['TimeofDay'] = pm25.index.time\n",
    "pm25['is_weekend'] = 'weekday'\n",
    "pm25.loc[pm25['DayofWeek'].isin(['Saturday','Sunday']),'is_weekend'] ='weekend'\n",
    "pm25['month-year'] = pm25.index.strftime('%b-%Y')\n",
    "\n",
    "# put the data into level and color based on the PM2.5 reading for easy visualization\n",
    "pm25['level'] = pd.cut(pm25['PM2.5avg'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['satisfactory', 'moderate','unhealthy'],\n",
    "                      include_lowest=True, right=False)\n",
    "pm25['color'] = pd.cut(pm25['PM2.5avg'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['green', 'orange','red'])\n",
    "\n",
    "pm25['season'] = 'other\\nseasons'\n",
    "pm25.loc['2016-09':'2017-05','season'] = 'winter\\n2016-2017'\n",
    "pm25.loc['2017-09':'2018-05','season'] = 'winter\\n2017-2018'\n",
    "pm25.loc['2018-09':'2019-05','season'] = 'winter\\n2018-2019'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 24130 entries, 2016-03-04 15:00:00 to 2019-03-07 04:00:00\n",
      "Data columns (total 12 columns):\n",
      "PM2.5         24130 non-null float64\n",
      "PM2.5avg      24130 non-null float64\n",
      "Year          24130 non-null int64\n",
      "Month         24130 non-null object\n",
      "DayofMonth    24130 non-null int64\n",
      "DayofWeek     24130 non-null object\n",
      "TimeofDay     24130 non-null object\n",
      "is_weekend    24130 non-null object\n",
      "month-year    24130 non-null object\n",
      "level         24130 non-null category\n",
      "color         24130 non-null category\n",
      "season        24130 non-null object\n",
      "dtypes: category(2), float64(2), int64(2), object(6)\n",
      "memory usage: 2.7+ MB\n"
     ]
    }
   ],
   "source": [
    "pm25.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save the data\n",
    "pm25.to_csv('data/model/bkk_pm25_proc.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  Data Visualization\n",
    "\n",
    "Try to answer questions about air pollution in Bangkok through visualization \n",
    "* Is the problem getting worse?\n",
    "* Which months have the highest level of PM2.5 in Bangkok? \n",
    "* What time has the highest pollution?\n",
    "* Is the pollution lower during the holiday?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 24130 entries, 2016-03-04 15:00:00 to 2019-03-07 04:00:00\n",
      "Data columns (total 11 columns):\n",
      "PM2.5         24130 non-null float64\n",
      "PM2.5avg      24130 non-null float64\n",
      "Year          24130 non-null int64\n",
      "Month         24130 non-null object\n",
      "DayofMonth    24130 non-null int64\n",
      "DayofWeek     24130 non-null object\n",
      "TimeofDay     24130 non-null object\n",
      "level         24130 non-null object\n",
      "color         24130 non-null object\n",
      "season        15649 non-null object\n",
      "is_weekend    24130 non-null object\n",
      "dtypes: float64(2), int64(2), object(7)\n",
      "memory usage: 2.2+ MB\n"
     ]
    }
   ],
   "source": [
    "#load the data and change time columne to datetime index\n",
    "pm25 = pd.read_csv('data/bkk_pm25_proc.csv')\n",
    "pm25['Time'] = pd.to_datetime(pm25['Time'])\n",
    "pm25 = pm25.set_index('Time')\n",
    "pm25.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {},
   "outputs": [
    {
     "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>PM2.5avg</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>DayofMonth</th>\n",
       "      <th>DayofWeek</th>\n",
       "      <th>TimeofDay</th>\n",
       "      <th>level</th>\n",
       "      <th>color</th>\n",
       "      <th>season</th>\n",
       "      <th>is_weekend</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Time</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>2016-03-04 15:00:00</th>\n",
       "      <td>34.2</td>\n",
       "      <td>43.520833</td>\n",
       "      <td>2016</td>\n",
       "      <td>March</td>\n",
       "      <td>4</td>\n",
       "      <td>Friday</td>\n",
       "      <td>15:00:00</td>\n",
       "      <td>moderate</td>\n",
       "      <td>orange</td>\n",
       "      <td>NaN</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-04 16:00:00</th>\n",
       "      <td>35.0</td>\n",
       "      <td>42.358333</td>\n",
       "      <td>2016</td>\n",
       "      <td>March</td>\n",
       "      <td>4</td>\n",
       "      <td>Friday</td>\n",
       "      <td>16:00:00</td>\n",
       "      <td>moderate</td>\n",
       "      <td>orange</td>\n",
       "      <td>NaN</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-04 17:00:00</th>\n",
       "      <td>30.0</td>\n",
       "      <td>40.987500</td>\n",
       "      <td>2016</td>\n",
       "      <td>March</td>\n",
       "      <td>4</td>\n",
       "      <td>Friday</td>\n",
       "      <td>17:00:00</td>\n",
       "      <td>moderate</td>\n",
       "      <td>orange</td>\n",
       "      <td>NaN</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-04 18:00:00</th>\n",
       "      <td>25.6</td>\n",
       "      <td>39.741667</td>\n",
       "      <td>2016</td>\n",
       "      <td>March</td>\n",
       "      <td>4</td>\n",
       "      <td>Friday</td>\n",
       "      <td>18:00:00</td>\n",
       "      <td>moderate</td>\n",
       "      <td>orange</td>\n",
       "      <td>NaN</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-03-04 19:00:00</th>\n",
       "      <td>32.1</td>\n",
       "      <td>38.766667</td>\n",
       "      <td>2016</td>\n",
       "      <td>March</td>\n",
       "      <td>4</td>\n",
       "      <td>Friday</td>\n",
       "      <td>19:00:00</td>\n",
       "      <td>moderate</td>\n",
       "      <td>orange</td>\n",
       "      <td>NaN</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     PM2.5   PM2.5avg  Year  Month  DayofMonth DayofWeek  \\\n",
       "Time                                                                       \n",
       "2016-03-04 15:00:00   34.2  43.520833  2016  March           4    Friday   \n",
       "2016-03-04 16:00:00   35.0  42.358333  2016  March           4    Friday   \n",
       "2016-03-04 17:00:00   30.0  40.987500  2016  March           4    Friday   \n",
       "2016-03-04 18:00:00   25.6  39.741667  2016  March           4    Friday   \n",
       "2016-03-04 19:00:00   32.1  38.766667  2016  March           4    Friday   \n",
       "\n",
       "                    TimeofDay     level   color season is_weekend  \n",
       "Time                                                               \n",
       "2016-03-04 15:00:00  15:00:00  moderate  orange    NaN    weekday  \n",
       "2016-03-04 16:00:00  16:00:00  moderate  orange    NaN    weekday  \n",
       "2016-03-04 17:00:00  17:00:00  moderate  orange    NaN    weekday  \n",
       "2016-03-04 18:00:00  18:00:00  moderate  orange    NaN    weekday  \n",
       "2016-03-04 19:00:00  19:00:00  moderate  orange    NaN    weekday  "
      ]
     },
     "execution_count": 278,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm25.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot the particle pollution vs time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['moderate', 'satisfactory', 'unhealthy'], dtype='<U12')"
      ]
     },
     "execution_count": 289,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check if the data contains all levels\n",
    "pm25['level'].unique().astype(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "code_folding": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(9, 6))\n",
    "ax1 = fig.add_subplot(211)\n",
    "\n",
    "# make legend and plot each level at a time \n",
    "# Each level has to be plotted separtely to generate proper legend\n",
    "for legend in ['satisfactory', 'moderate', 'unhealthy']:\n",
    "    temp = pm25[pm25['level']==legend]\n",
    "    # plot the data for each pollution level\n",
    "    ax1.scatter(temp.index, temp['PM2.5avg'], \n",
    "                c=temp['color'], marker='.', label=legend)\n",
    "\n",
    "plt.legend(loc='upper right')\n",
    "plt.title(\"PM2.5 level in Bangkok ($\\mu$g/m$^3$)\")\n",
    "\n",
    "# add vertical spand adnotation \n",
    "ax1.axvspan('2016-11','2017-04', color='gray', alpha=0.2)\n",
    "plt.text('2017-01', 75, r'Nov 2016 - Apr 2017 ',\n",
    "         {'color': 'k', 'fontsize': 16, 'ha': 'center'})\n",
    "\n",
    "# add vertical spand adnotation\n",
    "ax1.axvspan('2017-12','2018-04', color='gray', alpha=0.2)\n",
    "plt.text('2017-10', 75, r'Dec 2017 - Apr 2018 ',\n",
    "         {'color': 'k', 'fontsize': 16, 'ha': 'left'})\n",
    "# add vertical spand adnotation\n",
    "ax1.axvspan('2018-11','2019-02', color='gray', alpha=0.2)\n",
    "# format the top plot xaxis tick style\n",
    "ax1.xaxis.set_major_formatter(mdates.DateFormatter(\"%b\\n%Y\"))\n",
    "plt.xlim(pm25.index.min(), pm25.index.max())\n",
    "\n",
    "# plot google trend data\n",
    "ax2 = fig.add_subplot(212)\n",
    "trend.plot('date', ['aqi_trend', 'pm25_trend'], kind='line',\n",
    "           label=['AQI search', 'PM2.5 search'], ax=ax2)\n",
    "plt.title('Google Search Frequency in Thailand')\n",
    "# set the xaxis range to be the same \n",
    "plt.xlim(pm25.index.min(), pm25.index.max())\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The air pollution seem to be seasonal, with surge around winter seasion (Dec - Jan). In fact, the winter pollution exists since 2017, but public awareness has just increased recently.\n",
    "\n",
    "Is the problem getting worse every year? I am going to decided based on two graphs: the average pollution level and the number of day with moderate and unhealthy pollution level.\n",
    "\n",
    "First, I try grouping the data by year, then I try grouping by season instead because the problem start around November to March next year. I use the later approach. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Group data by year"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_days = pd.DataFrame()\n",
    "# number of day per year with particle lever higher than limit (AQI 150)\n",
    "num_days['unhealthy'] = pm25[pm25['PM2.5avg'] > 55.4].resample('d').mean().groupby('Year').count()['DayofMonth']\n",
    "\n",
    "# number of day per year with particle lever higher than limit (AQI 100)\n",
    "num_days['moderate'] = pm25[pm25['PM2.5avg'] > 35.4].resample('d').mean().groupby('Year').count()['DayofMonth'] - num_days['unhealthy']\n",
    "num_days.index = num_days.index.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'numbers of bad air days')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2,figsize=(9,5))\n",
    "\n",
    "# find yearly PM2.5 average\n",
    "yearavg = pm25.groupby('Year').mean()['PM2.5avg']\n",
    "# plot the yearly average \n",
    "yearavg.plot(kind='bar', color ='green',ax=ax[0])\n",
    "ax[0].set_ylabel('PM2.5 level $\\mu$g/m$^3$')\n",
    "ax[0].set_title('yearly average')\n",
    "\n",
    "# plot number of bad air day for each year\n",
    "num_days[['moderate','unhealthy']].plot(kind='bar', stacked=True, color = ['orange','red'], ax=ax[1])\n",
    "ax[1].set_title('numbers of bad air days')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Year\n",
      "2016    24.376378\n",
      "2017    25.117491\n",
      "2018    25.037562\n",
      "2019    43.118296\n",
      "Name: PM2.5avg, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "print(yearavg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The average for 2019 is high because we only have the winter data, which has high pollution.This set of data does not have data in Jan-Feb of 2016, and 2015, which explain why the number of day is lower in 2016. The year 2018 has 25 days with very high AQI (from the Jan-Feb and Dec). Without more data, it's hard to judge if the situation is getting worse. \n",
    "\n",
    "To summarize, the yearly average is not increasing, but the number of days with high AQI seem to be higher. Another way to see trend for time series data is to look at the rate of change, which is shown below. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 588,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([736420., 736511., 736603., 736695., 736785., 736876., 736968., 737060.]),\n",
       " <a list of 8 Text xticklabel objects>)"
      ]
     },
     "execution_count": 588,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(pm25['PM2.5avg'].diff().rolling(24*30*12).mean())\n",
    "plt.xticks(rotation=45)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The rate of change is around zero except at the very end. This confirm that the problem is about the same every year. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Group data by season\n",
    "\n",
    "Next, I will try another way to present the yearly trend. Since the pollution is seasonal around, winter, it's best to group the data by season rather than year."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 409,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>level</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Time</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2019-03-03</th>\n",
       "      <td>30.445833</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-03-04</th>\n",
       "      <td>25.304167</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-03-05</th>\n",
       "      <td>28.466667</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-03-06</th>\n",
       "      <td>31.725000</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-03-07</th>\n",
       "      <td>23.740000</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                PM2.5             season         level\n",
       "Time                                                  \n",
       "2019-03-03  30.445833  winter\\n2018-2019  satisfactory\n",
       "2019-03-04  25.304167  winter\\n2018-2019  satisfactory\n",
       "2019-03-05  28.466667  winter\\n2018-2019  satisfactory\n",
       "2019-03-06  31.725000  winter\\n2018-2019  satisfactory\n",
       "2019-03-07  23.740000  winter\\n2018-2019  satisfactory"
      ]
     },
     "execution_count": 409,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# select the data and resample by day\n",
    "dayavg = pm25['PM2.5'].resample('d').mean().to_frame()\n",
    "# create season, which start in Sep and end in May. \n",
    "# I select some month before and after winter starts to get some basedline\n",
    "dayavg['season'] = 'other\\nseasons'\n",
    "dayavg.loc['2016-09':'2017-05','season'] = 'winter\\n2016-2017'\n",
    "dayavg.loc['2017-09':'2018-05','season'] = 'winter\\n2017-2018'\n",
    "dayavg.loc['2018-09':'2019-05','season'] = 'winter\\n2018-2019'\n",
    "\n",
    "dayavg['level'] = pd.cut(dayavg['PM2.5'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['satisfactory', 'moderate','unhealthy'],\n",
    "                      include_lowest=True, right=False)\n",
    "dayavg.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 411,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['other\\nseasons', 'winter\\n2016-2017', 'winter\\n2017-2018', 'winter\\n2018-2019'], dtype=object)"
      ]
     },
     "execution_count": 411,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check of the separation work well\n",
    "dayavg['season'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1e40253ab70>"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#check the distribution \n",
    "dayavg['PM2.5'].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "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>moderate</th>\n",
       "      <th>unhealthy</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>season</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>winter\\n2016-2017</th>\n",
       "      <td>51</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>winter\\n2017-2018</th>\n",
       "      <td>57</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>winter\\n2018-2019</th>\n",
       "      <td>32</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   moderate  unhealthy\n",
       "season                                \n",
       "winter\\n2016-2017        51          9\n",
       "winter\\n2017-2018        57         13\n",
       "winter\\n2018-2019        32         11"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculate seasonal average and number of days with high pollutions\n",
    "numdays = pd.DataFrame()\n",
    "numdays = dayavg[['season','level','PM2.5']].groupby(['season','level'],\n",
    "                                                    as_index=False).count()\n",
    "# count the number of days with bad air\n",
    "numdays = numdays[numdays['level']!='satisfactory']\n",
    "numdays = numdays.dropna()\n",
    "# pivot into two columns based on level of pollution\n",
    "numdays = pd.pivot_table(numdays,index='season',columns='level')\n",
    "numdays.columns = ['moderate','unhealthy']\n",
    "numdays.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The number in 2019 is still low due to lack of data from the future. To estimate this missing data, I use the average during march 08 - may 31 and add to the data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 211,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "satisfactory    66\n",
       "moderate        12\n",
       "unhealthy        1\n",
       "dtype: int32"
      ]
     },
     "execution_count": 211,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp=pd.concat([dayavg.loc['2016-03-08':'2016-05-31', 'level'].value_counts(),\n",
    " dayavg.loc['2017-03-08':'2017-05-31', 'level'].value_counts(),\n",
    "          dayavg.loc['2018-03-08':'2018-05-31', 'level'].value_counts()],axis=1)\n",
    "\n",
    "temp = temp.mean(axis=1).astype(int)\n",
    "temp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {},
   "outputs": [],
   "source": [
    "numdays.iloc[-1,0] = numdays.iloc[-1,0]+ temp['moderate']\n",
    "numdays.iloc[-1,1] +=  temp['unhealthy']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 412,
   "metadata": {},
   "outputs": [
    {
     "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>season</th>\n",
       "      <th>PM2.5mean</th>\n",
       "      <th>PM2.5std</th>\n",
       "      <th>color</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>other\\nseasons</td>\n",
       "      <td>19.112784</td>\n",
       "      <td>8.002606</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>28.765613</td>\n",
       "      <td>12.395808</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>winter\\n2017-2018</td>\n",
       "      <td>28.570890</td>\n",
       "      <td>13.535275</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>29.876587</td>\n",
       "      <td>12.793924</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              season  PM2.5mean   PM2.5std  color\n",
       "0     other\\nseasons  19.112784   8.002606  green\n",
       "1  winter\\n2016-2017  28.765613  12.395808  green\n",
       "2  winter\\n2017-2018  28.570890  13.535275  green\n",
       "3  winter\\n2018-2019  29.876587  12.793924  green"
      ]
     },
     "execution_count": 412,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculate the seasonal average \n",
    "yearavg = dayavg[['season','PM2.5']].groupby('season', as_index=False).agg({'PM2.5':[np.mean,np.std]})\n",
    "yearavg['color'] = pd.cut(yearavg['PM2.5','mean'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['green', 'orange','red'], include_lowest=True, right=False).astype(str)\n",
    "yearavg.columns = [a+b for a,b in yearavg.columns]\n",
    "\n",
    "yearavg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 414,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1,2,figsize=(11,4))\n",
    "\n",
    "#plot seasonal average\n",
    "yearavg.plot(x='season', y='PM2.5mean',yerr='PM2.5std', kind='bar', \n",
    "             edgecolor='black', color='green',ax=ax[0], legend=None)\n",
    "ax[0].set_ylabel('PM2.5 level $\\mu$g/m$^3$')\n",
    "ax[0].set_title('seasonal average')\n",
    "ax[0].tick_params(rotation=0)\n",
    "ax[0].xaxis.label.set_visible(False)\n",
    "\n",
    "#plot number of bad air days using a stackbar plot\n",
    "numdays.plot(kind='bar',stacked=True, color = ['orange','red'], edgecolor='black', ax=ax[1])\n",
    "ax[1].set_title('numbers of bad air days')\n",
    "ax[1].tick_params(rotation=0)\n",
    "ax[1].xaxis.label.set_visible(False)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {},
   "outputs": [
    {
     "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>moderate</th>\n",
       "      <th>unhealthy</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>season</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>winter\\n2016-2017</th>\n",
       "      <td>51</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>winter\\n2017-2018</th>\n",
       "      <td>57</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>winter\\n2018-2019</th>\n",
       "      <td>44</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   moderate  unhealthy\n",
       "season                                \n",
       "winter\\n2016-2017        51          9\n",
       "winter\\n2017-2018        57         13\n",
       "winter\\n2018-2019        44         12"
      ]
     },
     "execution_count": 218,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numdays"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What is the best time to visit bangkok ?\n",
    "\n",
    "Bangkok is a tourist city, which month should you avoid going to Bangkok. If you are Bangkokians what is the best time to go for a jog? Is the air pollution better during the holiday when everyone is away?\n",
    "\n",
    "I am going to perform analysis for 2017 and 2018. First, separate a table for 2017 and 2018, and then create a daily average for each date of month."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 24130 entries, 2016-03-04 15:00:00 to 2019-03-07 04:00:00\n",
      "Data columns (total 9 columns):\n",
      "PM2.5         24130 non-null float64\n",
      "PM2.5avg      24130 non-null float64\n",
      "Year          24130 non-null int64\n",
      "Month         24130 non-null object\n",
      "DayofMonth    24130 non-null int64\n",
      "DayofWeek     24130 non-null object\n",
      "TimeofDay     24130 non-null object\n",
      "level         24130 non-null object\n",
      "color         24130 non-null object\n",
      "dtypes: float64(2), int64(2), object(5)\n",
      "memory usage: 2.5+ MB\n"
     ]
    }
   ],
   "source": [
    "pm25.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 428,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get the data from 2017 and 2018\n",
    "pm25_2017 = pm25['2017']\n",
    "pm25_2018 = pm25['2018']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 429,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# make a cross table with month as index, and day of month as columns\n",
    "date_month_2017 = pd.crosstab(pm25_2017['Month'], pm25_2017['DayofMonth'], pm25_2017['PM2.5avg'], aggfunc=np.mean)\n",
    "month_name=pd.date_range('2014-01','2014-12', freq='MS').strftime(\"%B\").tolist()\n",
    "date_month_2017 = date_month_2017.reindex(month_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 430,
   "metadata": {},
   "outputs": [
    {
     "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>DayofMonth</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "      <th>24</th>\n",
       "      <th>25</th>\n",
       "      <th>26</th>\n",
       "      <th>27</th>\n",
       "      <th>28</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Month</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",
       "      <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>January</th>\n",
       "      <td>26.188715</td>\n",
       "      <td>21.179167</td>\n",
       "      <td>20.485764</td>\n",
       "      <td>21.911458</td>\n",
       "      <td>26.305208</td>\n",
       "      <td>25.600694</td>\n",
       "      <td>27.596528</td>\n",
       "      <td>32.629167</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>26.099306</td>\n",
       "      <td>24.397656</td>\n",
       "      <td>30.672743</td>\n",
       "      <td>37.478646</td>\n",
       "      <td>36.815278</td>\n",
       "      <td>31.755035</td>\n",
       "      <td>30.134954</td>\n",
       "      <td>38.933553</td>\n",
       "      <td>44.263194</td>\n",
       "      <td>48.823958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>February</th>\n",
       "      <td>44.708333</td>\n",
       "      <td>43.674638</td>\n",
       "      <td>34.771875</td>\n",
       "      <td>27.608507</td>\n",
       "      <td>34.921875</td>\n",
       "      <td>41.838368</td>\n",
       "      <td>45.634201</td>\n",
       "      <td>36.740399</td>\n",
       "      <td>36.043924</td>\n",
       "      <td>46.258333</td>\n",
       "      <td>...</td>\n",
       "      <td>34.183854</td>\n",
       "      <td>24.312500</td>\n",
       "      <td>16.761979</td>\n",
       "      <td>20.480556</td>\n",
       "      <td>22.173264</td>\n",
       "      <td>25.217882</td>\n",
       "      <td>45.363542</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>March</th>\n",
       "      <td>52.884375</td>\n",
       "      <td>52.621007</td>\n",
       "      <td>59.289757</td>\n",
       "      <td>56.183102</td>\n",
       "      <td>44.355254</td>\n",
       "      <td>39.864394</td>\n",
       "      <td>37.183854</td>\n",
       "      <td>35.740972</td>\n",
       "      <td>32.018229</td>\n",
       "      <td>26.337500</td>\n",
       "      <td>...</td>\n",
       "      <td>28.008507</td>\n",
       "      <td>28.348785</td>\n",
       "      <td>17.879861</td>\n",
       "      <td>19.900725</td>\n",
       "      <td>22.532986</td>\n",
       "      <td>21.077778</td>\n",
       "      <td>29.292187</td>\n",
       "      <td>34.711285</td>\n",
       "      <td>29.931944</td>\n",
       "      <td>19.481076</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>April</th>\n",
       "      <td>14.470312</td>\n",
       "      <td>19.842935</td>\n",
       "      <td>26.610590</td>\n",
       "      <td>37.460937</td>\n",
       "      <td>33.968576</td>\n",
       "      <td>34.481076</td>\n",
       "      <td>25.777976</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>34.656597</td>\n",
       "      <td>44.060307</td>\n",
       "      <td>40.371181</td>\n",
       "      <td>33.898958</td>\n",
       "      <td>32.803299</td>\n",
       "      <td>31.099826</td>\n",
       "      <td>30.104861</td>\n",
       "      <td>28.541146</td>\n",
       "      <td>31.427951</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>May</th>\n",
       "      <td>19.723090</td>\n",
       "      <td>14.684601</td>\n",
       "      <td>19.766667</td>\n",
       "      <td>22.557246</td>\n",
       "      <td>19.998611</td>\n",
       "      <td>23.216146</td>\n",
       "      <td>21.708507</td>\n",
       "      <td>16.574132</td>\n",
       "      <td>20.622049</td>\n",
       "      <td>18.831522</td>\n",
       "      <td>...</td>\n",
       "      <td>21.952431</td>\n",
       "      <td>26.578993</td>\n",
       "      <td>25.458333</td>\n",
       "      <td>18.023785</td>\n",
       "      <td>18.687847</td>\n",
       "      <td>19.546078</td>\n",
       "      <td>23.195660</td>\n",
       "      <td>14.970833</td>\n",
       "      <td>18.881944</td>\n",
       "      <td>22.692708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>June</th>\n",
       "      <td>23.324094</td>\n",
       "      <td>19.736979</td>\n",
       "      <td>16.212153</td>\n",
       "      <td>12.963715</td>\n",
       "      <td>13.932118</td>\n",
       "      <td>12.734028</td>\n",
       "      <td>13.708854</td>\n",
       "      <td>14.748913</td>\n",
       "      <td>11.793297</td>\n",
       "      <td>18.038715</td>\n",
       "      <td>...</td>\n",
       "      <td>16.761632</td>\n",
       "      <td>15.415451</td>\n",
       "      <td>12.203993</td>\n",
       "      <td>13.359783</td>\n",
       "      <td>14.652778</td>\n",
       "      <td>14.896007</td>\n",
       "      <td>15.065972</td>\n",
       "      <td>16.526812</td>\n",
       "      <td>19.936285</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>July</th>\n",
       "      <td>19.809601</td>\n",
       "      <td>17.471875</td>\n",
       "      <td>15.485326</td>\n",
       "      <td>18.164410</td>\n",
       "      <td>19.745486</td>\n",
       "      <td>19.416493</td>\n",
       "      <td>16.140942</td>\n",
       "      <td>15.134201</td>\n",
       "      <td>14.564062</td>\n",
       "      <td>14.927381</td>\n",
       "      <td>...</td>\n",
       "      <td>17.012862</td>\n",
       "      <td>14.671920</td>\n",
       "      <td>17.653571</td>\n",
       "      <td>15.600521</td>\n",
       "      <td>13.964062</td>\n",
       "      <td>14.816304</td>\n",
       "      <td>15.596558</td>\n",
       "      <td>16.834028</td>\n",
       "      <td>21.505254</td>\n",
       "      <td>25.133333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>August</th>\n",
       "      <td>27.968924</td>\n",
       "      <td>25.758160</td>\n",
       "      <td>22.667882</td>\n",
       "      <td>21.971377</td>\n",
       "      <td>21.398264</td>\n",
       "      <td>19.938021</td>\n",
       "      <td>16.441319</td>\n",
       "      <td>17.348958</td>\n",
       "      <td>17.513368</td>\n",
       "      <td>16.924826</td>\n",
       "      <td>...</td>\n",
       "      <td>13.851042</td>\n",
       "      <td>16.296007</td>\n",
       "      <td>17.014757</td>\n",
       "      <td>15.631597</td>\n",
       "      <td>14.400000</td>\n",
       "      <td>15.793182</td>\n",
       "      <td>13.343841</td>\n",
       "      <td>19.594618</td>\n",
       "      <td>25.519792</td>\n",
       "      <td>17.697917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>September</th>\n",
       "      <td>18.014062</td>\n",
       "      <td>17.294375</td>\n",
       "      <td>15.744518</td>\n",
       "      <td>12.304340</td>\n",
       "      <td>12.365104</td>\n",
       "      <td>12.319129</td>\n",
       "      <td>12.403993</td>\n",
       "      <td>15.214015</td>\n",
       "      <td>19.090972</td>\n",
       "      <td>20.762083</td>\n",
       "      <td>...</td>\n",
       "      <td>24.179340</td>\n",
       "      <td>23.312500</td>\n",
       "      <td>21.299132</td>\n",
       "      <td>14.965799</td>\n",
       "      <td>11.197569</td>\n",
       "      <td>11.872826</td>\n",
       "      <td>14.443056</td>\n",
       "      <td>14.853819</td>\n",
       "      <td>23.052899</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>October</th>\n",
       "      <td>30.267029</td>\n",
       "      <td>24.944565</td>\n",
       "      <td>20.084601</td>\n",
       "      <td>17.035326</td>\n",
       "      <td>15.866406</td>\n",
       "      <td>18.530625</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>21.175725</td>\n",
       "      <td>17.046701</td>\n",
       "      <td>17.875347</td>\n",
       "      <td>18.820588</td>\n",
       "      <td>30.443640</td>\n",
       "      <td>29.958889</td>\n",
       "      <td>31.483152</td>\n",
       "      <td>39.876042</td>\n",
       "      <td>41.330903</td>\n",
       "      <td>37.988768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>November</th>\n",
       "      <td>36.906076</td>\n",
       "      <td>37.854861</td>\n",
       "      <td>33.858681</td>\n",
       "      <td>41.866840</td>\n",
       "      <td>37.387674</td>\n",
       "      <td>30.828993</td>\n",
       "      <td>36.672222</td>\n",
       "      <td>32.021181</td>\n",
       "      <td>30.913889</td>\n",
       "      <td>22.482609</td>\n",
       "      <td>...</td>\n",
       "      <td>15.055254</td>\n",
       "      <td>16.956061</td>\n",
       "      <td>16.667424</td>\n",
       "      <td>18.666288</td>\n",
       "      <td>23.619965</td>\n",
       "      <td>20.761775</td>\n",
       "      <td>22.338068</td>\n",
       "      <td>21.127536</td>\n",
       "      <td>21.423732</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>December</th>\n",
       "      <td>24.781341</td>\n",
       "      <td>22.414583</td>\n",
       "      <td>23.754529</td>\n",
       "      <td>22.118576</td>\n",
       "      <td>18.231875</td>\n",
       "      <td>25.921528</td>\n",
       "      <td>29.123785</td>\n",
       "      <td>35.915972</td>\n",
       "      <td>25.938021</td>\n",
       "      <td>27.693403</td>\n",
       "      <td>...</td>\n",
       "      <td>49.164205</td>\n",
       "      <td>50.840761</td>\n",
       "      <td>53.707465</td>\n",
       "      <td>31.341319</td>\n",
       "      <td>42.974653</td>\n",
       "      <td>52.931884</td>\n",
       "      <td>28.033712</td>\n",
       "      <td>20.389236</td>\n",
       "      <td>18.022569</td>\n",
       "      <td>21.482465</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "DayofMonth         1          2          3          4          5          6   \\\n",
       "Month                                                                          \n",
       "January     26.188715  21.179167  20.485764  21.911458  26.305208  25.600694   \n",
       "February    44.708333  43.674638  34.771875  27.608507  34.921875  41.838368   \n",
       "March       52.884375  52.621007  59.289757  56.183102  44.355254  39.864394   \n",
       "April       14.470312  19.842935  26.610590  37.460937  33.968576  34.481076   \n",
       "May         19.723090  14.684601  19.766667  22.557246  19.998611  23.216146   \n",
       "June        23.324094  19.736979  16.212153  12.963715  13.932118  12.734028   \n",
       "July        19.809601  17.471875  15.485326  18.164410  19.745486  19.416493   \n",
       "August      27.968924  25.758160  22.667882  21.971377  21.398264  19.938021   \n",
       "September   18.014062  17.294375  15.744518  12.304340  12.365104  12.319129   \n",
       "October     30.267029  24.944565  20.084601  17.035326  15.866406  18.530625   \n",
       "November    36.906076  37.854861  33.858681  41.866840  37.387674  30.828993   \n",
       "December    24.781341  22.414583  23.754529  22.118576  18.231875  25.921528   \n",
       "\n",
       "DayofMonth         7          8          9          10  ...         22  \\\n",
       "Month                                                   ...              \n",
       "January     27.596528  32.629167        NaN        NaN  ...  26.099306   \n",
       "February    45.634201  36.740399  36.043924  46.258333  ...  34.183854   \n",
       "March       37.183854  35.740972  32.018229  26.337500  ...  28.008507   \n",
       "April       25.777976        NaN        NaN        NaN  ...  34.656597   \n",
       "May         21.708507  16.574132  20.622049  18.831522  ...  21.952431   \n",
       "June        13.708854  14.748913  11.793297  18.038715  ...  16.761632   \n",
       "July        16.140942  15.134201  14.564062  14.927381  ...  17.012862   \n",
       "August      16.441319  17.348958  17.513368  16.924826  ...  13.851042   \n",
       "September   12.403993  15.214015  19.090972  20.762083  ...  24.179340   \n",
       "October           NaN        NaN        NaN        NaN  ...  21.175725   \n",
       "November    36.672222  32.021181  30.913889  22.482609  ...  15.055254   \n",
       "December    29.123785  35.915972  25.938021  27.693403  ...  49.164205   \n",
       "\n",
       "DayofMonth         23         24         25         26         27         28  \\\n",
       "Month                                                                          \n",
       "January     24.397656  30.672743  37.478646  36.815278  31.755035  30.134954   \n",
       "February    24.312500  16.761979  20.480556  22.173264  25.217882  45.363542   \n",
       "March       28.348785  17.879861  19.900725  22.532986  21.077778  29.292187   \n",
       "April       44.060307  40.371181  33.898958  32.803299  31.099826  30.104861   \n",
       "May         26.578993  25.458333  18.023785  18.687847  19.546078  23.195660   \n",
       "June        15.415451  12.203993  13.359783  14.652778  14.896007  15.065972   \n",
       "July        14.671920  17.653571  15.600521  13.964062  14.816304  15.596558   \n",
       "August      16.296007  17.014757  15.631597  14.400000  15.793182  13.343841   \n",
       "September   23.312500  21.299132  14.965799  11.197569  11.872826  14.443056   \n",
       "October     17.046701  17.875347  18.820588  30.443640  29.958889  31.483152   \n",
       "November    16.956061  16.667424  18.666288  23.619965  20.761775  22.338068   \n",
       "December    50.840761  53.707465  31.341319  42.974653  52.931884  28.033712   \n",
       "\n",
       "DayofMonth         29         30         31  \n",
       "Month                                        \n",
       "January     38.933553  44.263194  48.823958  \n",
       "February          NaN        NaN        NaN  \n",
       "March       34.711285  29.931944  19.481076  \n",
       "April       28.541146  31.427951        NaN  \n",
       "May         14.970833  18.881944  22.692708  \n",
       "June        16.526812  19.936285        NaN  \n",
       "July        16.834028  21.505254  25.133333  \n",
       "August      19.594618  25.519792  17.697917  \n",
       "September   14.853819  23.052899        NaN  \n",
       "October     39.876042  41.330903  37.988768  \n",
       "November    21.127536  21.423732        NaN  \n",
       "December    20.389236  18.022569  21.482465  \n",
       "\n",
       "[12 rows x 31 columns]"
      ]
     },
     "execution_count": 430,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check the data\n",
    "date_month_2017"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 431,
   "metadata": {},
   "outputs": [],
   "source": [
    "# make a cross table with month as index, and day of month as columns\n",
    "date_month_2018 = pd.crosstab(pm25_2018['Month'], pm25_2018['DayofMonth'], pm25_2018['PM2.5avg'], aggfunc=np.mean)\n",
    "month_name = pd.date_range('2014-01','2014-12', freq='MS').strftime(\"%B\").tolist()\n",
    "date_month_2018 = date_month_2018.reindex(month_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The crosstab/heatmap will be a giant table. To help the reader, I am adding a little bar plot on the side to show monthly average."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 495,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# create a montly average \n",
    "monthavg = pm25[['Month','PM2.5avg','color']].groupby('Month').mean()\n",
    "monthavg = monthavg.reindex(reversed(month_name))\n",
    "monthavg['color'] = pd.cut(monthavg['PM2.5avg'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['green', 'orange','red'],\n",
    "                      include_lowest=True, right=False).astype(str)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I am going to have four figures, two for the heatmap, one for the monthly average, then another figure just for the colorbar. This way the two heatmaps for 2017 and 2018 will have the same size. Therefore, I will have four subplots on one row with different size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 436,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create a color bar with AQI color system\n",
    "colors = ['green','orange','red']\n",
    "levels = [0, 35.4, 55.4]\n",
    "cmap, norm = mpl.colors.from_levels_and_colors(levels,colors,'max')\n",
    "\n",
    "fig = plt.figure(figsize= (10,5))\n",
    "# make a grid space for each subplot\n",
    "gs = GridSpec(1,10)\n",
    "# plot the first heatmap for 2017 data\n",
    "ax1 = fig.add_subplot(gs[0:4])\n",
    "sns.heatmap(date_month_2017, cmap=cmap, norm=norm, linewidth=0.05,vmax=70, cbar=False, ax=ax1)\n",
    "ax1.yaxis.label.set_visible(False)\n",
    "plt.title('daily average in 2017')\n",
    "plt.xlabel('day of month')\n",
    "\n",
    "# plot the first heatmap for 2018 data\n",
    "ax2 = fig.add_subplot(gs[4:8],sharey =ax1)\n",
    "ax2 = sns.heatmap(date_month_2018, cmap=cmap, norm=norm, linewidth=0.05,vmax=70, cbar=False)\n",
    "plt.title('daily average in 2018')\n",
    "plt.xlabel('day of month')\n",
    "ax2.get_yaxis().set_visible(False)\n",
    "\n",
    "# plot monthly average as a horizontal side bar, and color coded them\n",
    "ax3 = fig.add_subplot(gs[8])\n",
    "monthavg = pm25[['Month','PM2.5avg']].groupby('Month').mean()\n",
    "monthavg = monthavg.reindex(reversed(month_name))\n",
    "monthavg['color'] = pd.cut(monthavg['PM2.5avg'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['green', 'orange','red'],\n",
    "                      include_lowest=True, right=False).astype(str)\n",
    "monthavg.plot(y='PM2.5avg', kind='barh',color = monthavg['color'], legend=False, ax=ax3,\n",
    "             edgecolor='black')\n",
    "ax3.get_yaxis().set_visible(False)\n",
    "plt.title('monthly \\n average')\n",
    "\n",
    "# plot the colorbar on the further right\n",
    "cmap2 = mpl.colors.ListedColormap(colors=['green','orange','red'])\n",
    "ax4 = fig.add_subplot(gs[9])\n",
    "ax4.set_aspect(30)\n",
    "cbar = mpl.colorbar.ColorbarBase(cmap=cmap2, ax=ax4)\n",
    "cbar.set_ticks([0.2, 0.5, 0.8])\n",
    "cbar.set_ticklabels(['satisfactory', 'moderate', 'unhealthy'])\n",
    "cbar.set_label('PM 2.5 level')\n",
    "\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For both 2017 and 2018, the pollution starts around October and last until early March. So tourists should staying in Bangkok in these months. I know what you are thinking. \"Hey, those are my holiday! Where else I can go? May suggest Phuket, and Samui Island? The south of Thailand has low AQI all year long. We will see the average pollution for each province later. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Day of Week and Hourly average \n",
    "\n",
    "What is the suggestion for the local Bangkokians? Is the morning Air or weekend better? What is the best time to go outside? We need to see the day of week data and the hourly average data. \n",
    "\n",
    "After playing with the data, I found that too much averaging will obscure the hourly pattern, so I will only show a data for one month. I choose Feb 2018 because it is the month with the worse air.\n",
    "\n",
    "Note that I played around with different plots and average. You are going to see different versions of the same plots."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Day of Week and Hourly average on the same plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "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>PM2.5avg</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>DayofMonth</th>\n",
       "      <th>DayofWeek</th>\n",
       "      <th>TimeofDay</th>\n",
       "      <th>level</th>\n",
       "      <th>color</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Time</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>2018-02-01 00:00:00</th>\n",
       "      <td>18.6</td>\n",
       "      <td>24.445833</td>\n",
       "      <td>2018</td>\n",
       "      <td>February</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>00:00:00</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-02-01 01:00:00</th>\n",
       "      <td>17.5</td>\n",
       "      <td>23.945833</td>\n",
       "      <td>2018</td>\n",
       "      <td>February</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>01:00:00</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-02-01 02:00:00</th>\n",
       "      <td>16.7</td>\n",
       "      <td>23.570833</td>\n",
       "      <td>2018</td>\n",
       "      <td>February</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>02:00:00</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-02-01 03:00:00</th>\n",
       "      <td>14.6</td>\n",
       "      <td>23.045833</td>\n",
       "      <td>2018</td>\n",
       "      <td>February</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>03:00:00</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-02-01 04:00:00</th>\n",
       "      <td>23.1</td>\n",
       "      <td>23.133333</td>\n",
       "      <td>2018</td>\n",
       "      <td>February</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>04:00:00</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     PM2.5   PM2.5avg  Year     Month  DayofMonth DayofWeek  \\\n",
       "Time                                                                          \n",
       "2018-02-01 00:00:00   18.6  24.445833  2018  February           1  Thursday   \n",
       "2018-02-01 01:00:00   17.5  23.945833  2018  February           1  Thursday   \n",
       "2018-02-01 02:00:00   16.7  23.570833  2018  February           1  Thursday   \n",
       "2018-02-01 03:00:00   14.6  23.045833  2018  February           1  Thursday   \n",
       "2018-02-01 04:00:00   23.1  23.133333  2018  February           1  Thursday   \n",
       "\n",
       "                    TimeofDay         level  color  \n",
       "Time                                                \n",
       "2018-02-01 00:00:00  00:00:00  satisfactory  green  \n",
       "2018-02-01 01:00:00  01:00:00  satisfactory  green  \n",
       "2018-02-01 02:00:00  02:00:00  satisfactory  green  \n",
       "2018-02-01 03:00:00  03:00:00  satisfactory  green  \n",
       "2018-02-01 04:00:00  04:00:00  satisfactory  green  "
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# select Feb 2018 data\n",
    "pm25_winter = pm25['2018-02']\n",
    "pm25_winter.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DayofWeek</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",
       "      <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>Monday</th>\n",
       "      <td>44.295833</td>\n",
       "      <td>44.760417</td>\n",
       "      <td>45.240625</td>\n",
       "      <td>45.669792</td>\n",
       "      <td>45.933333</td>\n",
       "      <td>46.202083</td>\n",
       "      <td>46.604167</td>\n",
       "      <td>46.912500</td>\n",
       "      <td>47.011111</td>\n",
       "      <td>47.291667</td>\n",
       "      <td>...</td>\n",
       "      <td>46.645833</td>\n",
       "      <td>47.273958</td>\n",
       "      <td>47.329167</td>\n",
       "      <td>47.222917</td>\n",
       "      <td>52.098611</td>\n",
       "      <td>52.104167</td>\n",
       "      <td>51.986111</td>\n",
       "      <td>51.977778</td>\n",
       "      <td>51.701389</td>\n",
       "      <td>46.572917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tuesday</th>\n",
       "      <td>46.283333</td>\n",
       "      <td>45.945833</td>\n",
       "      <td>49.776389</td>\n",
       "      <td>44.611458</td>\n",
       "      <td>43.536111</td>\n",
       "      <td>43.402083</td>\n",
       "      <td>42.721875</td>\n",
       "      <td>42.035417</td>\n",
       "      <td>41.353125</td>\n",
       "      <td>40.691667</td>\n",
       "      <td>...</td>\n",
       "      <td>37.426042</td>\n",
       "      <td>36.972917</td>\n",
       "      <td>36.561458</td>\n",
       "      <td>36.332292</td>\n",
       "      <td>36.102083</td>\n",
       "      <td>35.993750</td>\n",
       "      <td>35.938542</td>\n",
       "      <td>35.793750</td>\n",
       "      <td>35.671875</td>\n",
       "      <td>35.623958</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wednesday</th>\n",
       "      <td>35.537500</td>\n",
       "      <td>35.553125</td>\n",
       "      <td>35.892708</td>\n",
       "      <td>38.343056</td>\n",
       "      <td>38.758333</td>\n",
       "      <td>36.902083</td>\n",
       "      <td>37.364583</td>\n",
       "      <td>31.905556</td>\n",
       "      <td>38.758333</td>\n",
       "      <td>39.859375</td>\n",
       "      <td>...</td>\n",
       "      <td>38.837500</td>\n",
       "      <td>43.107292</td>\n",
       "      <td>43.772917</td>\n",
       "      <td>44.350000</td>\n",
       "      <td>36.306944</td>\n",
       "      <td>36.800000</td>\n",
       "      <td>37.175000</td>\n",
       "      <td>45.831250</td>\n",
       "      <td>46.578125</td>\n",
       "      <td>47.288542</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Thursday</th>\n",
       "      <td>21.762500</td>\n",
       "      <td>39.286111</td>\n",
       "      <td>45.532292</td>\n",
       "      <td>45.663542</td>\n",
       "      <td>45.966667</td>\n",
       "      <td>54.919444</td>\n",
       "      <td>45.986458</td>\n",
       "      <td>46.118750</td>\n",
       "      <td>46.450000</td>\n",
       "      <td>55.890278</td>\n",
       "      <td>...</td>\n",
       "      <td>34.163889</td>\n",
       "      <td>34.254167</td>\n",
       "      <td>34.101389</td>\n",
       "      <td>45.435417</td>\n",
       "      <td>45.359375</td>\n",
       "      <td>45.482292</td>\n",
       "      <td>45.394792</td>\n",
       "      <td>45.475000</td>\n",
       "      <td>45.586458</td>\n",
       "      <td>45.615625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Friday</th>\n",
       "      <td>45.646875</td>\n",
       "      <td>45.546875</td>\n",
       "      <td>45.461458</td>\n",
       "      <td>45.420833</td>\n",
       "      <td>49.395833</td>\n",
       "      <td>44.873958</td>\n",
       "      <td>44.906250</td>\n",
       "      <td>44.826042</td>\n",
       "      <td>44.868750</td>\n",
       "      <td>44.720833</td>\n",
       "      <td>...</td>\n",
       "      <td>42.467708</td>\n",
       "      <td>42.042708</td>\n",
       "      <td>41.690625</td>\n",
       "      <td>41.488542</td>\n",
       "      <td>41.377083</td>\n",
       "      <td>41.287500</td>\n",
       "      <td>41.294792</td>\n",
       "      <td>41.401042</td>\n",
       "      <td>41.591667</td>\n",
       "      <td>41.806250</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0          1          2          3          4          5   \\\n",
       "DayofWeek                                                                     \n",
       "Monday     44.295833  44.760417  45.240625  45.669792  45.933333  46.202083   \n",
       "Tuesday    46.283333  45.945833  49.776389  44.611458  43.536111  43.402083   \n",
       "Wednesday  35.537500  35.553125  35.892708  38.343056  38.758333  36.902083   \n",
       "Thursday   21.762500  39.286111  45.532292  45.663542  45.966667  54.919444   \n",
       "Friday     45.646875  45.546875  45.461458  45.420833  49.395833  44.873958   \n",
       "\n",
       "                  6          7          8          9   ...         14  \\\n",
       "DayofWeek                                              ...              \n",
       "Monday     46.604167  46.912500  47.011111  47.291667  ...  46.645833   \n",
       "Tuesday    42.721875  42.035417  41.353125  40.691667  ...  37.426042   \n",
       "Wednesday  37.364583  31.905556  38.758333  39.859375  ...  38.837500   \n",
       "Thursday   45.986458  46.118750  46.450000  55.890278  ...  34.163889   \n",
       "Friday     44.906250  44.826042  44.868750  44.720833  ...  42.467708   \n",
       "\n",
       "                  15         16         17         18         19         20  \\\n",
       "DayofWeek                                                                     \n",
       "Monday     47.273958  47.329167  47.222917  52.098611  52.104167  51.986111   \n",
       "Tuesday    36.972917  36.561458  36.332292  36.102083  35.993750  35.938542   \n",
       "Wednesday  43.107292  43.772917  44.350000  36.306944  36.800000  37.175000   \n",
       "Thursday   34.254167  34.101389  45.435417  45.359375  45.482292  45.394792   \n",
       "Friday     42.042708  41.690625  41.488542  41.377083  41.287500  41.294792   \n",
       "\n",
       "                  21         22         23  \n",
       "DayofWeek                                   \n",
       "Monday     51.977778  51.701389  46.572917  \n",
       "Tuesday    35.793750  35.671875  35.623958  \n",
       "Wednesday  45.831250  46.578125  47.288542  \n",
       "Thursday   45.475000  45.586458  45.615625  \n",
       "Friday     41.401042  41.591667  41.806250  \n",
       "\n",
       "[5 rows x 24 columns]"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create new df with day of week as index and hour of day as columns\n",
    "date_time_map = pd.crosstab(pm25_winter['DayofWeek'],pm25_winter['TimeofDay'], pm25_winter['PM2.5avg'], aggfunc=np.mean)\n",
    "day_name=pd.date_range('2019-03-04','2019-03-10', freq='D').strftime(\"%A\").tolist()\n",
    "date_time_map = date_time_map.reindex(day_name)\n",
    "# use integer number for hours\n",
    "date_time_map.columns = [int(column[:2]) for column in date_time_map.columns ]\n",
    "date_time_map.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "day_time_map plot is going to be a heatmap. To help the reader, I am going to have another line plot showing the average for each hour on the bottom of the heatmap. I want these two plots to share the xaxis. Here, I have to turn of the colorbar and plot it as a separate figure. Also, I need to use another colorbar  so the change is more apparent."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "code_folding": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# define grid space\n",
    "fig = plt.figure(figsize= (9,6))\n",
    "gs = GridSpec(2, 2, width_ratios=[8, 1], height_ratios=[4, 2])\n",
    "\n",
    "#plot heatmap\n",
    "ax1 = fig.add_subplot(gs[0])\n",
    "ax1 = sns.heatmap(date_time_map, cmap='YlOrRd', linewidth=0.5, cbar=False)\n",
    "ax1.set_xlabel('Day of Week')\n",
    "plt.title('Hourly Average in Feb 2018 for Each day')\n",
    "ax1.yaxis.label.set_visible(False)\n",
    "\n",
    "#plot colorbar\n",
    "ax4 = fig.add_subplot(gs[1])\n",
    "cbar = colorbar(ax1.get_children()[0], ax4)\n",
    "ax4.set_ylabel('PM 2.5 level')\n",
    "ax4.set_aspect(30)\n",
    "plt.subplots_adjust(wspace=0)\n",
    "\n",
    "#plot hourly average on the bottom\n",
    "ax2 = fig.add_subplot(gs[2], sharex=ax1)\n",
    "#plt.bar(date_time_map.mean().index, height=date_time_map.mean().values)\n",
    "date_time_map.mean().plot(ax=ax2, c='red')\n",
    "plt.ylabel('PM 2.5 $\\mu$g/m$^3$')\n",
    "plt.xlabel('Hour');\n",
    "plt.title('Hourly Average');\n",
    "#plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>20</th>\n",
       "      <th>21</th>\n",
       "      <th>22</th>\n",
       "      <th>23</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DayofWeek</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",
       "      <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>Monday</th>\n",
       "      <td>38.397917</td>\n",
       "      <td>38.834470</td>\n",
       "      <td>38.887121</td>\n",
       "      <td>38.911742</td>\n",
       "      <td>38.847727</td>\n",
       "      <td>40.005903</td>\n",
       "      <td>38.664394</td>\n",
       "      <td>38.547727</td>\n",
       "      <td>38.434848</td>\n",
       "      <td>38.375379</td>\n",
       "      <td>...</td>\n",
       "      <td>37.286667</td>\n",
       "      <td>37.136250</td>\n",
       "      <td>38.437500</td>\n",
       "      <td>38.388636</td>\n",
       "      <td>38.361742</td>\n",
       "      <td>38.328788</td>\n",
       "      <td>38.277273</td>\n",
       "      <td>38.220833</td>\n",
       "      <td>38.167803</td>\n",
       "      <td>38.189773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tuesday</th>\n",
       "      <td>38.191667</td>\n",
       "      <td>39.219167</td>\n",
       "      <td>37.892045</td>\n",
       "      <td>35.173333</td>\n",
       "      <td>36.884259</td>\n",
       "      <td>39.072917</td>\n",
       "      <td>38.819167</td>\n",
       "      <td>38.559167</td>\n",
       "      <td>36.362500</td>\n",
       "      <td>36.226136</td>\n",
       "      <td>...</td>\n",
       "      <td>34.942803</td>\n",
       "      <td>34.640530</td>\n",
       "      <td>34.380682</td>\n",
       "      <td>34.105682</td>\n",
       "      <td>33.895455</td>\n",
       "      <td>33.711742</td>\n",
       "      <td>33.562121</td>\n",
       "      <td>34.463889</td>\n",
       "      <td>35.182500</td>\n",
       "      <td>35.097917</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Wednesday</th>\n",
       "      <td>33.178409</td>\n",
       "      <td>32.990152</td>\n",
       "      <td>32.782955</td>\n",
       "      <td>32.587500</td>\n",
       "      <td>32.402273</td>\n",
       "      <td>32.295076</td>\n",
       "      <td>32.239015</td>\n",
       "      <td>32.277273</td>\n",
       "      <td>32.332955</td>\n",
       "      <td>32.510606</td>\n",
       "      <td>...</td>\n",
       "      <td>33.269697</td>\n",
       "      <td>33.304167</td>\n",
       "      <td>33.332197</td>\n",
       "      <td>33.433712</td>\n",
       "      <td>33.573864</td>\n",
       "      <td>33.804545</td>\n",
       "      <td>34.073485</td>\n",
       "      <td>34.335227</td>\n",
       "      <td>34.602652</td>\n",
       "      <td>34.764015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Thursday</th>\n",
       "      <td>34.902273</td>\n",
       "      <td>35.078409</td>\n",
       "      <td>35.266288</td>\n",
       "      <td>35.501136</td>\n",
       "      <td>35.701136</td>\n",
       "      <td>37.002083</td>\n",
       "      <td>34.576667</td>\n",
       "      <td>35.083796</td>\n",
       "      <td>35.942424</td>\n",
       "      <td>35.946970</td>\n",
       "      <td>...</td>\n",
       "      <td>35.779167</td>\n",
       "      <td>35.778030</td>\n",
       "      <td>36.710185</td>\n",
       "      <td>34.672685</td>\n",
       "      <td>34.609722</td>\n",
       "      <td>33.862917</td>\n",
       "      <td>35.605303</td>\n",
       "      <td>33.871250</td>\n",
       "      <td>36.014394</td>\n",
       "      <td>34.518333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Friday</th>\n",
       "      <td>35.315278</td>\n",
       "      <td>35.786111</td>\n",
       "      <td>36.272222</td>\n",
       "      <td>36.763426</td>\n",
       "      <td>36.785417</td>\n",
       "      <td>37.360417</td>\n",
       "      <td>38.298148</td>\n",
       "      <td>38.750926</td>\n",
       "      <td>38.311458</td>\n",
       "      <td>40.642187</td>\n",
       "      <td>...</td>\n",
       "      <td>40.720833</td>\n",
       "      <td>40.765741</td>\n",
       "      <td>40.857407</td>\n",
       "      <td>45.556548</td>\n",
       "      <td>43.598958</td>\n",
       "      <td>42.589583</td>\n",
       "      <td>42.618229</td>\n",
       "      <td>41.492593</td>\n",
       "      <td>41.568981</td>\n",
       "      <td>41.623611</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  0          1          2          3          4          5   \\\n",
       "DayofWeek                                                                     \n",
       "Monday     38.397917  38.834470  38.887121  38.911742  38.847727  40.005903   \n",
       "Tuesday    38.191667  39.219167  37.892045  35.173333  36.884259  39.072917   \n",
       "Wednesday  33.178409  32.990152  32.782955  32.587500  32.402273  32.295076   \n",
       "Thursday   34.902273  35.078409  35.266288  35.501136  35.701136  37.002083   \n",
       "Friday     35.315278  35.786111  36.272222  36.763426  36.785417  37.360417   \n",
       "\n",
       "                  6          7          8          9   ...         14  \\\n",
       "DayofWeek                                              ...              \n",
       "Monday     38.664394  38.547727  38.434848  38.375379  ...  37.286667   \n",
       "Tuesday    38.819167  38.559167  36.362500  36.226136  ...  34.942803   \n",
       "Wednesday  32.239015  32.277273  32.332955  32.510606  ...  33.269697   \n",
       "Thursday   34.576667  35.083796  35.942424  35.946970  ...  35.779167   \n",
       "Friday     38.298148  38.750926  38.311458  40.642187  ...  40.720833   \n",
       "\n",
       "                  15         16         17         18         19         20  \\\n",
       "DayofWeek                                                                     \n",
       "Monday     37.136250  38.437500  38.388636  38.361742  38.328788  38.277273   \n",
       "Tuesday    34.640530  34.380682  34.105682  33.895455  33.711742  33.562121   \n",
       "Wednesday  33.304167  33.332197  33.433712  33.573864  33.804545  34.073485   \n",
       "Thursday   35.778030  36.710185  34.672685  34.609722  33.862917  35.605303   \n",
       "Friday     40.765741  40.857407  45.556548  43.598958  42.589583  42.618229   \n",
       "\n",
       "                  21         22         23  \n",
       "DayofWeek                                   \n",
       "Monday     38.220833  38.167803  38.189773  \n",
       "Tuesday    34.463889  35.182500  35.097917  \n",
       "Wednesday  34.335227  34.602652  34.764015  \n",
       "Thursday   33.871250  36.014394  34.518333  \n",
       "Friday     41.492593  41.568981  41.623611  \n",
       "\n",
       "[5 rows x 24 columns]"
      ]
     },
     "execution_count": 150,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Select the time range\n",
    "pm25_winter_2018 = pm25['2018-12':'2019-02']\n",
    "#genrate table between average time of day and day of week\n",
    "date_time_map = pd.crosstab(pm25_winter_2018['DayofWeek'],pm25_winter_2018['TimeofDay'], pm25_winter_2018['PM2.5avg'], aggfunc=np.mean)\n",
    "#create string of Monday to Sunday\n",
    "day_name=pd.date_range('2019-03-04','2019-03-10', freq='D').strftime(\"%A\").tolist()\n",
    "#reindex to give index order \n",
    "date_time_map = date_time_map.reindex(day_name)\n",
    "date_time_map.columns = [int(column[:2]) for column in date_time_map.columns.values ]\n",
    "date_time_map.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Maybe it is also nice to see the average over different day of week also? I tried adding them as a side plot on the right."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x432 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize= (9,6))\n",
    "gs = GridSpec(2, 3, width_ratios=[8, 3, 2], height_ratios=[4, 2])\n",
    "\n",
    "#plot the date_time_map as heat map\n",
    "ax1 = fig.add_subplot(gs[0])\n",
    "ax1 = sns.heatmap(date_time_map, cmap='autumn_r', linewidth=0.5, cbar=False)\n",
    "plt.title('hourly average in winter 2018')\n",
    "ax1.yaxis.label.set_visible(False)\n",
    "\n",
    "# plot color bar on the side\n",
    "ax4 = fig.add_subplot(gs[1])\n",
    "cbar = colorbar(ax1.get_children()[0], ax4)\n",
    "ax4.set_ylabel('PM 2.5 level')\n",
    "ax4.set_aspect(30)\n",
    "plt.subplots_adjust(wspace=0)\n",
    "\n",
    "#plot the average for each day of week on the right\n",
    "ax3 = fig.add_subplot(gs[2])\n",
    "ax3.plot(date_time_map.mean(axis=1), day_name,c='red')\n",
    "ax3.yaxis.set_visible(False)\n",
    "plt.title('day of week\\n variation')\n",
    "\n",
    "#plot the hourly average on the bottom\n",
    "ax2 = fig.add_subplot(gs[3], sharex=ax1)\n",
    "date_time_map.mean().plot(ax=ax2, c='red')\n",
    "plt.ylabel('PM 2.5 $\\mu$g/m$^3$')\n",
    "plt.xlabel('hour');\n",
    "#plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Morning air is not better. The better time is in in the late afternoon. Weekend is not significantly better compared to the weekdays. This graph is not very useful. Too many graphs.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Day of Week and Hourly average on different plots\n",
    "\n",
    "Instead of putting the day of week and hourly average on the same graph, I will put them on a separate plot. Here, I can also see the trend for each winter season by adding another categorical columns that indicate the month and year."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {},
   "outputs": [],
   "source": [
    "winterdata = pd.DataFrame(columns=pm25.columns)\n",
    "# select data during winter \n",
    "periods = [('2016-12','2017-02'),('2017-12','2018-02'), ('2018-12','2019-02')]\n",
    "for start,stop in periods:\n",
    "    winterdata = pd.concat([winterdata, pm25[start:stop]])\n",
    "    \n",
    "# generate a month-year columns\n",
    "winterdata['month-year'] = winterdata.index.strftime('%b-%Y')\n",
    "# create a time of day column with integer type\n",
    "winterdata['TimeofDay'] = [int(time[:2]) for time in winterdata['TimeofDay'].values]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {},
   "outputs": [
    {
     "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>PM2.5avg</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>DayofMonth</th>\n",
       "      <th>DayofWeek</th>\n",
       "      <th>TimeofDay</th>\n",
       "      <th>level</th>\n",
       "      <th>color</th>\n",
       "      <th>season</th>\n",
       "      <th>is_weekend</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-12-01 00:00:00</th>\n",
       "      <td>32.4</td>\n",
       "      <td>28.600000</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>0</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01 01:00:00</th>\n",
       "      <td>30.0</td>\n",
       "      <td>28.270833</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>1</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01 02:00:00</th>\n",
       "      <td>31.7</td>\n",
       "      <td>27.837500</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>2</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01 03:00:00</th>\n",
       "      <td>28.6</td>\n",
       "      <td>27.437500</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>3</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01 04:00:00</th>\n",
       "      <td>31.3</td>\n",
       "      <td>27.195833</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>4</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     PM2.5   PM2.5avg  Year     Month DayofMonth DayofWeek  \\\n",
       "2016-12-01 00:00:00   32.4  28.600000  2016  December          1  Thursday   \n",
       "2016-12-01 01:00:00   30.0  28.270833  2016  December          1  Thursday   \n",
       "2016-12-01 02:00:00   31.7  27.837500  2016  December          1  Thursday   \n",
       "2016-12-01 03:00:00   28.6  27.437500  2016  December          1  Thursday   \n",
       "2016-12-01 04:00:00   31.3  27.195833  2016  December          1  Thursday   \n",
       "\n",
       "                     TimeofDay         level  color             season  \\\n",
       "2016-12-01 00:00:00          0  satisfactory  green  winter\\n2016-2017   \n",
       "2016-12-01 01:00:00          1  satisfactory  green  winter\\n2016-2017   \n",
       "2016-12-01 02:00:00          2  satisfactory  green  winter\\n2016-2017   \n",
       "2016-12-01 03:00:00          3  satisfactory  green  winter\\n2016-2017   \n",
       "2016-12-01 04:00:00          4  satisfactory  green  winter\\n2016-2017   \n",
       "\n",
       "                    is_weekend  \n",
       "2016-12-01 00:00:00    weekday  \n",
       "2016-12-01 01:00:00    weekday  \n",
       "2016-12-01 02:00:00    weekday  \n",
       "2016-12-01 03:00:00    weekday  \n",
       "2016-12-01 04:00:00    weekday  "
      ]
     },
     "execution_count": 283,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "winterdata.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 421,
   "metadata": {},
   "outputs": [
    {
     "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>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>DayofWeek</th>\n",
       "      <th>season</th>\n",
       "      <th>is_weekend</th>\n",
       "      <th>month-year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2019-02-24</th>\n",
       "      <td>26.308333</td>\n",
       "      <td>2019</td>\n",
       "      <td>Feb</td>\n",
       "      <td>Sunday</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>weekend</td>\n",
       "      <td>Feb\\n2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-25</th>\n",
       "      <td>26.690909</td>\n",
       "      <td>2019</td>\n",
       "      <td>Feb</td>\n",
       "      <td>Monday</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Feb\\n2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-26</th>\n",
       "      <td>26.856522</td>\n",
       "      <td>2019</td>\n",
       "      <td>Feb</td>\n",
       "      <td>Tuesday</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Feb\\n2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-27</th>\n",
       "      <td>27.254167</td>\n",
       "      <td>2019</td>\n",
       "      <td>Feb</td>\n",
       "      <td>Wednesday</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Feb\\n2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-02-28</th>\n",
       "      <td>31.720833</td>\n",
       "      <td>2019</td>\n",
       "      <td>Feb</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>winter\\n2018-2019</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Feb\\n2019</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                PM2.5  Year Month  DayofWeek             season is_weekend  \\\n",
       "2019-02-24  26.308333  2019   Feb     Sunday  winter\\n2018-2019    weekend   \n",
       "2019-02-25  26.690909  2019   Feb     Monday  winter\\n2018-2019    weekday   \n",
       "2019-02-26  26.856522  2019   Feb    Tuesday  winter\\n2018-2019    weekday   \n",
       "2019-02-27  27.254167  2019   Feb  Wednesday  winter\\n2018-2019    weekday   \n",
       "2019-02-28  31.720833  2019   Feb   Thursday  winter\\n2018-2019    weekday   \n",
       "\n",
       "           month-year  \n",
       "2019-02-24  Feb\\n2019  \n",
       "2019-02-25  Feb\\n2019  \n",
       "2019-02-26  Feb\\n2019  \n",
       "2019-02-27  Feb\\n2019  \n",
       "2019-02-28  Feb\\n2019  "
      ]
     },
     "execution_count": 421,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# resample data by day\n",
    "dayavg_decFeb = winterdata.resample('d').agg({\n",
    "    'PM2.5':'mean','Year':'last','Month':'last','DayofWeek':'last',\n",
    "    'season':'last','is_weekend':'last'\n",
    "})\n",
    "# make a level columns\n",
    "dayavg['level'] = pd.cut(dayavg_decFeb['PM2.5'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['satisfactory', 'moderate','unhealthy'],\n",
    "                      include_lowest=True, right=False).astype(str)\n",
    "\n",
    "dayavg_decFeb['month-year'] = dayavg_decFeb.index.strftime('%b\\n%Y')\n",
    "dayavg_decFeb['Month'] = dayavg_decFeb.index.strftime('%b')\n",
    "dayavg_decFeb.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 425,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# create a table with month-year for each columns\n",
    "day_map = pd.pivot_table(dayavg_decFeb, index='DayofWeek',columns=['month-year'],values='PM2.5',\n",
    "              aggfunc='mean')\n",
    "day_name=pd.date_range('2019-03-04','2019-03-10', freq='D').strftime(\"%A\").tolist()\n",
    "\n",
    "# reorder the data\n",
    "day_map = day_map.reindex(['Dec\\n2016', 'Jan\\n2017','Feb\\n2017', 'Dec\\n2017',\n",
    "                      'Jan\\n2018', 'Feb\\n2018',  'Dec\\n2018', 'Jan\\n2019','Feb\\n2019'], axis=1)\n",
    "day_map = day_map.reindex(day_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 423,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Dec\\n2016', 'Dec\\n2017', 'Dec\\n2018', 'Feb\\n2017', 'Feb\\n2018',\n",
       "       'Feb\\n2019', 'Jan\\n2017', 'Jan\\n2018', 'Jan\\n2019'],\n",
       "      dtype='object', name='month-year')"
      ]
     },
     "execution_count": 423,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "day_map.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 427,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'daly average in winter')"
      ]
     },
     "execution_count": 427,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#make a heatmap\n",
    "fig = plt.figure(figsize= (9,4))\n",
    "gs = GridSpec(1, 2, width_ratios=[4,1])\n",
    "ax1 = fig.add_subplot(gs[0])\n",
    "sns.heatmap(day_map,cmap='autumn_r', linewidth=0.5,ax=ax1,\n",
    "           cbar_kws={'label': 'PM 2.5 level'})\n",
    "ax1.yaxis.label.set_visible(False)\n",
    "ax1.xaxis.label.set_visible(False)\n",
    "plt.title('daly average in winter')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "No pattern. To confirm that there is not trend for day of week, using lineplot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 442,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1e417132f60>"
      ]
     },
     "execution_count": 442,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=dayavg_decFeb, x='DayofWeek',y='PM2.5', hue='month-year',\n",
    "            legend=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "See if the normalized data would show the paper. I am going to divide each columns by it maximum value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 477,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#normalize\n",
    "day_map_norm = day_map/day_map.max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 478,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'daly average in winter')"
      ]
     },
     "execution_count": 478,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize= (9,4))\n",
    "gs = GridSpec(1, 2, width_ratios=[4,1])\n",
    "ax1 = fig.add_subplot(gs[0])\n",
    "sns.heatmap(day_map_norm,cmap='autumn_r', linewidth=0.5,ax=ax1,\n",
    "           cbar_kws={'label': 'normalized PM 2.5 level'})\n",
    "ax1.yaxis.label.set_visible(False)\n",
    "ax1.xaxis.label.set_visible(False)\n",
    "plt.title('daly average in winter')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot the line plot to really make sure there is not trend"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 493,
   "metadata": {},
   "outputs": [],
   "source": [
    "day_map_norm_melt = pd.melt(day_map_norm.reset_index(),id_vars='DayofWeek')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 491,
   "metadata": {},
   "outputs": [
    {
     "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>DayofWeek</th>\n",
       "      <th>month-year</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Monday</td>\n",
       "      <td>Dec\\n2016</td>\n",
       "      <td>0.889537</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Tuesday</td>\n",
       "      <td>Dec\\n2016</td>\n",
       "      <td>0.851834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Wednesday</td>\n",
       "      <td>Dec\\n2016</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Thursday</td>\n",
       "      <td>Dec\\n2016</td>\n",
       "      <td>0.704397</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Friday</td>\n",
       "      <td>Dec\\n2016</td>\n",
       "      <td>0.683375</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   DayofWeek month-year     value\n",
       "0     Monday  Dec\\n2016  0.889537\n",
       "1    Tuesday  Dec\\n2016  0.851834\n",
       "2  Wednesday  Dec\\n2016  1.000000\n",
       "3   Thursday  Dec\\n2016  0.704397\n",
       "4     Friday  Dec\\n2016  0.683375"
      ]
     },
     "execution_count": 491,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "day_map_norm_melt.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 494,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1e413f7dc18>"
      ]
     },
     "execution_count": 494,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lineplot(data=day_map_norm_melt, x='DayofWeek',y='value', hue='month-year',\n",
    "            legend=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One thing I learn from exploring this data, is too much averaging can make your trend go away. To look at a trend for time of day, I need to look at the raw plot vs time, instead of the average hours value. I will focus on the data in winter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 446,
   "metadata": {},
   "outputs": [
    {
     "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>PM2.5avg</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>DayofMonth</th>\n",
       "      <th>DayofWeek</th>\n",
       "      <th>TimeofDay</th>\n",
       "      <th>level</th>\n",
       "      <th>color</th>\n",
       "      <th>season</th>\n",
       "      <th>is_weekend</th>\n",
       "      <th>month-year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2016-12-01 00:00:00</th>\n",
       "      <td>32.4</td>\n",
       "      <td>28.600000</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>0</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Dec-2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01 01:00:00</th>\n",
       "      <td>30.0</td>\n",
       "      <td>28.270833</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>1</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Dec-2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01 02:00:00</th>\n",
       "      <td>31.7</td>\n",
       "      <td>27.837500</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>2</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Dec-2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01 03:00:00</th>\n",
       "      <td>28.6</td>\n",
       "      <td>27.437500</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>3</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Dec-2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016-12-01 04:00:00</th>\n",
       "      <td>31.3</td>\n",
       "      <td>27.195833</td>\n",
       "      <td>2016</td>\n",
       "      <td>December</td>\n",
       "      <td>1</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>4</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>green</td>\n",
       "      <td>winter\\n2016-2017</td>\n",
       "      <td>weekday</td>\n",
       "      <td>Dec-2016</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     PM2.5   PM2.5avg  Year     Month DayofMonth DayofWeek  \\\n",
       "2016-12-01 00:00:00   32.4  28.600000  2016  December          1  Thursday   \n",
       "2016-12-01 01:00:00   30.0  28.270833  2016  December          1  Thursday   \n",
       "2016-12-01 02:00:00   31.7  27.837500  2016  December          1  Thursday   \n",
       "2016-12-01 03:00:00   28.6  27.437500  2016  December          1  Thursday   \n",
       "2016-12-01 04:00:00   31.3  27.195833  2016  December          1  Thursday   \n",
       "\n",
       "                     TimeofDay         level  color             season  \\\n",
       "2016-12-01 00:00:00          0  satisfactory  green  winter\\n2016-2017   \n",
       "2016-12-01 01:00:00          1  satisfactory  green  winter\\n2016-2017   \n",
       "2016-12-01 02:00:00          2  satisfactory  green  winter\\n2016-2017   \n",
       "2016-12-01 03:00:00          3  satisfactory  green  winter\\n2016-2017   \n",
       "2016-12-01 04:00:00          4  satisfactory  green  winter\\n2016-2017   \n",
       "\n",
       "                    is_weekend month-year  \n",
       "2016-12-01 00:00:00    weekday   Dec-2016  \n",
       "2016-12-01 01:00:00    weekday   Dec-2016  \n",
       "2016-12-01 02:00:00    weekday   Dec-2016  \n",
       "2016-12-01 03:00:00    weekday   Dec-2016  \n",
       "2016-12-01 04:00:00    weekday   Dec-2016  "
      ]
     },
     "execution_count": 446,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "winterdata.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 466,
   "metadata": {},
   "outputs": [
    {
     "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>month-year</th>\n",
       "      <th>TimeofDay</th>\n",
       "      <th>PM2.5</th>\n",
       "      <th>DayofWeek</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Dec-2016</td>\n",
       "      <td>0</td>\n",
       "      <td>50.903226</td>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Dec-2016</td>\n",
       "      <td>1</td>\n",
       "      <td>49.022581</td>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Dec-2016</td>\n",
       "      <td>2</td>\n",
       "      <td>50.461290</td>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dec-2016</td>\n",
       "      <td>3</td>\n",
       "      <td>47.529032</td>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Dec-2016</td>\n",
       "      <td>4</td>\n",
       "      <td>45.174194</td>\n",
       "      <td>Saturday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  month-year  TimeofDay      PM2.5 DayofWeek\n",
       "0   Dec-2016          0  50.903226  Saturday\n",
       "1   Dec-2016          1  49.022581  Saturday\n",
       "2   Dec-2016          2  50.461290  Saturday\n",
       "3   Dec-2016          3  47.529032  Saturday\n",
       "4   Dec-2016          4  45.174194  Saturday"
      ]
     },
     "execution_count": 466,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find the hourly average data group by month-year to plot separetely \n",
    "hourlyavg_win = winterdata.groupby(['month-year','TimeofDay'],as_index=False).agg({'PM2.5':'mean','DayofWeek':'last'})\n",
    "\n",
    "monthyear_sorter = ['Dec\\n2016', 'Jan\\n2017','Feb\\n2017', 'Dec\\n2017',\n",
    "                      'Jan\\n2018', 'Feb\\n2018',  'Dec\\n2018', 'Jan\\n2019','Feb\\n2019']\n",
    "day_name=pd.date_range('2019-03-04','2019-03-10', freq='D').strftime(\"%A\").tolist()\n",
    "\n",
    "hourlyavg_win = hourlyavg_win.sort_values(['month-year','DayofWeek'],ascending=[monthyear_sorter,day_name])\n",
    "hourlyavg_win.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 464,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1e41785d710>"
      ]
     },
     "execution_count": 464,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the mean of each month\n",
    "fig = plt.figure(figsize=(7,4))\n",
    "sns.lineplot(data=hourlyavg_win, x='TimeofDay',y='PM2.5',hue='month-year')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 474,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'PM2.5 $\\\\mu g/m^3$')"
      ]
     },
     "execution_count": 474,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plot all of the data \n",
    "fig = plt.figure(figsize=(7,4))\n",
    "sns.lineplot(data=winterdata, x='TimeofDay',y='PM2.5',hue='month-year')\n",
    "plt.legend(bbox_to_anchor=(1.02, 1))\n",
    "plt.xlabel('hour')\n",
    "plt.ylabel('PM2.5 $\\mu g/m^3$')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I like the last plot the most. It shows that pollution level tend to be low around 15 am. We can also see the peak during the morning rush hour around 8 - 9 am on some month. I will use this for my blog."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Maps\n",
    "\n",
    "Looking at the pollution level of the entire countries will tell if the pollution focus on the Bangkok. Is there provinces around Bangkok with worse pollution? If so, the pollution will inevitably flow to Bangkok.What provinces has worse pollution and lowest pollution. \n",
    "\n",
    "Here is my procedure.\n",
    "\n",
    "1. Extract city name, region name and the lattitude and logtitude from all data files, and convert to Mercator system. \n",
    "2. Collect PM2.5 data from all provinces in a single DataFrame and save it\n",
    "3. Get the names of all BKK neighbor province\n",
    "4. Display the data on a map. I use tile from Stamen.\n",
    "5. Build interactive plots or gif showing the change over months time \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## 1. Extract city name, region name and the lattitude and logtitude from all data files, and convert to Mercator system. \n",
    "\n",
    "I am going to use Bokeh geographical map, which use Mercator coordinate system. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import re"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "#Extract geospatial data and city, region information\n",
    "def geo_info(filename):\n",
    "    with open(filename,'r') as f:\n",
    "        data = f.read()\n",
    "        city = re.search('City: (.+)', data).group(1)\n",
    "        region = re.search('Region: (.+)', data).group(1)\n",
    "        pop = re.search('Population: (\\w+)', data).group(1)    \n",
    "        lat = re.search('Latitude: (\\d+.\\d+)', data).group(1)\n",
    "        lng = re.search('Longitude: (\\d+.\\d+)', data).group(1)  \n",
    "    return [city, region, pop, lat, lng]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 444,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "region_info = []\n",
    "for filename in glob('data/*.txt'):\n",
    "    info = geo_info(filename)\n",
    "    region_info.append(info)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 445,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "region_df = pd.DataFrame(region_info, columns=['City','Region', 'Population','Latitude','Longitude'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 520,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# function to change to Mercator\n",
    "#from https://wiki.openstreetmap.org/wiki/Mercator\n",
    "import math\n",
    "\n",
    "def merc_x(lon):\n",
    "  r_major = 6378137.000\n",
    "  return r_major*math.radians(lon)\n",
    "\n",
    "def merc_y(lat):\n",
    "    lat += 0.08\n",
    "    if lat>89.5:lat=89.5\n",
    "    if lat<-89.5:lat=-89.5\n",
    "    r_major=6378137.000\n",
    "    r_minor=6356752.3142\n",
    "    temp=r_minor/r_major\n",
    "    eccent=math.sqrt(1-temp**2)\n",
    "    phi=math.radians(lat)\n",
    "    sinphi=math.sin(phi)\n",
    "    con=eccent*sinphi\n",
    "    com=eccent/2\n",
    "    # shift the coordinate a little bit to make them align.\n",
    "    con=((1.0-con)/(1.0+con))**com\n",
    "    ts=math.tan((math.pi/2-phi)/2)/con\n",
    "    y=0-r_major*math.log(ts)\n",
    "    return y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 683,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "region_df['x_mer']= region_df.Longitude.apply(merc_x)\n",
    "region_df['y_mer']= region_df.Latitude.apply(merc_y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 649,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "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>City</th>\n",
       "      <th>Region</th>\n",
       "      <th>Population</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>x_mer</th>\n",
       "      <th>y_mer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>56</td>\n",
       "      <td>56</td>\n",
       "      <td>5.600000e+01</td>\n",
       "      <td>56.000000</td>\n",
       "      <td>56.000000</td>\n",
       "      <td>5.600000e+01</td>\n",
       "      <td>5.600000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>56</td>\n",
       "      <td>41</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>Pak Kret</td>\n",
       "      <td>Nonthaburi</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.956620e+05</td>\n",
       "      <td>13.554377</td>\n",
       "      <td>100.738621</td>\n",
       "      <td>1.121417e+07</td>\n",
       "      <td>1.572554e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.713606e+05</td>\n",
       "      <td>3.231692</td>\n",
       "      <td>1.311508</td>\n",
       "      <td>1.459965e+05</td>\n",
       "      <td>3.675895e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.000000e+04</td>\n",
       "      <td>6.540000</td>\n",
       "      <td>98.398100</td>\n",
       "      <td>1.095363e+07</td>\n",
       "      <td>7.804353e+05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.834075e+04</td>\n",
       "      <td>12.677275</td>\n",
       "      <td>99.953025</td>\n",
       "      <td>1.112672e+07</td>\n",
       "      <td>1.470258e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.311300e+04</td>\n",
       "      <td>13.840150</td>\n",
       "      <td>100.524200</td>\n",
       "      <td>1.119030e+07</td>\n",
       "      <td>1.602694e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.237000e+05</td>\n",
       "      <td>15.731150</td>\n",
       "      <td>101.027775</td>\n",
       "      <td>1.124636e+07</td>\n",
       "      <td>1.819571e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5.104476e+06</td>\n",
       "      <td>19.908600</td>\n",
       "      <td>104.848700</td>\n",
       "      <td>1.167170e+07</td>\n",
       "      <td>2.306605e+06</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            City      Region    Population   Latitude   Longitude  \\\n",
       "count         56          56  5.600000e+01  56.000000   56.000000   \n",
       "unique        56          41           NaN        NaN         NaN   \n",
       "top     Pak Kret  Nonthaburi           NaN        NaN         NaN   \n",
       "freq           1           3           NaN        NaN         NaN   \n",
       "mean         NaN         NaN  1.956620e+05  13.554377  100.738621   \n",
       "std          NaN         NaN  6.713606e+05   3.231692    1.311508   \n",
       "min          NaN         NaN  5.000000e+04   6.540000   98.398100   \n",
       "25%          NaN         NaN  5.834075e+04  12.677275   99.953025   \n",
       "50%          NaN         NaN  8.311300e+04  13.840150  100.524200   \n",
       "75%          NaN         NaN  1.237000e+05  15.731150  101.027775   \n",
       "max          NaN         NaN  5.104476e+06  19.908600  104.848700   \n",
       "\n",
       "               x_mer         y_mer  \n",
       "count   5.600000e+01  5.600000e+01  \n",
       "unique           NaN           NaN  \n",
       "top              NaN           NaN  \n",
       "freq             NaN           NaN  \n",
       "mean    1.121417e+07  1.572554e+06  \n",
       "std     1.459965e+05  3.675895e+05  \n",
       "min     1.095363e+07  7.804353e+05  \n",
       "25%     1.112672e+07  1.470258e+06  \n",
       "50%     1.119030e+07  1.602694e+06  \n",
       "75%     1.124636e+07  1.819571e+06  \n",
       "max     1.167170e+07  2.306605e+06  "
      ]
     },
     "execution_count": 649,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "region_df.describe(include='all')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 453,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "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>City</th>\n",
       "      <th>Region</th>\n",
       "      <th>Population</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Bang Kruai</td>\n",
       "      <td>Nonthaburi</td>\n",
       "      <td>78305</td>\n",
       "      <td>13.805</td>\n",
       "      <td>100.4728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Mueang Nonthaburi</td>\n",
       "      <td>Nonthaburi</td>\n",
       "      <td>291555</td>\n",
       "      <td>13.8607</td>\n",
       "      <td>100.5148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>Pak Kret</td>\n",
       "      <td>Nonthaburi</td>\n",
       "      <td>182926</td>\n",
       "      <td>13.913</td>\n",
       "      <td>100.4988</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 City      Region Population Latitude Longitude\n",
       "1          Bang Kruai  Nonthaburi      78305   13.805  100.4728\n",
       "25  Mueang Nonthaburi  Nonthaburi     291555  13.8607  100.5148\n",
       "31           Pak Kret  Nonthaburi     182926   13.913  100.4988"
      ]
     },
     "execution_count": 453,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "region_df.query('Region == \"Nonthaburi\"')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "region_df['Region'] = region_df['Region'].str.replace('TH.S','Chonburi')\n",
    "region_df['Region'] = region_df['Region'].str.replace('Changwat','')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "region_df.to_csv('data/region_info.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "region_df = pd.read_csv('data/region_info.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## 2. Collect PM 2.5 data from all provinces\n",
    "\n",
    "Put them in in a single DataFrame and save it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "def read_pm(filename):\n",
    "    pm25 = pd.read_csv(filename, sep='\\t', header=None, skiprows=10)\n",
    "    #assemble datetime column \n",
    "    pm25['Time'] = pd.to_datetime({'year': pm25[0], 'month': pm25[1], 'day':pm25[2],'hour': pm25[3]})\n",
    "    #convert time zone\n",
    "    pm25['Time'] = pm25['Time'].dt.tz_localize('UTC').dt.tz_convert('Asia/Bangkok')\n",
    "    pm25['Time'] = pm25['Time'].dt.tz_localize(None)\n",
    "    pm25=pm25.drop([0,1,2,3,5,6],axis=1)\n",
    "    pm25.columns = ['PM2.5', 'Time']\n",
    "    \n",
    "    city = geo_info(filename)[0]\n",
    "    pm25['city'] = city\n",
    "    \n",
    "    return pm25"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "pm25_all = pd.DataFrame()\n",
    "for filename in glob('data/pm25/*.txt'):\n",
    "    pm25_region = read_pm(filename)\n",
    "    pm25_all = pm25_all.append(pm25_region,ignore_index=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "pm25_all.to_csv('data/pm25_all.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "pm25_all = pd.read_csv('data/pm25_all.csv')\n",
    "pm25_all['Time'] = pd.to_datetime(pm25_all['Time'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1249843 entries, 0 to 1249842\n",
      "Data columns (total 3 columns):\n",
      "PM2.5    1249843 non-null float64\n",
      "Time     1249843 non-null datetime64[ns]\n",
      "city     1249843 non-null object\n",
      "dtypes: datetime64[ns](1), float64(1), object(1)\n",
      "memory usage: 28.6+ MB\n"
     ]
    }
   ],
   "source": [
    "pm25_all.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## 3. Get the names of all Bangkok neighbor provinces"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 383,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Obtain all province in BKK neighbors\n",
    "bkk_nei = ['Bangkok']\n",
    "with open('data/Bangkok.neighbors.json') as json_file:  \n",
    "    data = json.load(json_file)\n",
    "    for vec in data:\n",
    "        bkk_nei.append(vec[1])\n",
    "\n",
    "#remove province not in the proximity of BKK \n",
    "bkk_nei = bkk_nei[:8]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 384,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Bangkok',\n",
       " 'Nonthaburi',\n",
       " 'Samut_Prakan',\n",
       " 'Samut_Sakhon',\n",
       " 'Pathum_Thani',\n",
       " 'Nakhon_Pathom',\n",
       " 'Phra_Nakhon_Si_Ayutthaya',\n",
       " 'Ratchaburi']"
      ]
     },
     "execution_count": 384,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bkk_nei"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# load the region info and pm25 records\n",
    "region_df = pd.read_csv('data/region_info.csv')\n",
    "\n",
    "pm25_all = pd.read_csv('data/pm25_all.csv')\n",
    "pm25_all['Time'] = pd.to_datetime(pm25_all['Time'])\n",
    "pm25_all.set_index('Time',inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "## 4. Display the data on a map. \n",
    "\n",
    "I use tile from Stamen and Bokeh plot, the data is group by month.\n",
    "First, try to plot the data for a single month, Jan 2019."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# select data for Jan 2019 for all regions\n",
    "pm25_all_jan19 = pm25_all['2019-01']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# take the average \n",
    "reg_monthavg = pm25_all_jan19.groupby('city', as_index=False).mean()\n",
    "reg_monthavg = reg_monthavg.merge(region_df, left_on ='city', right_on='City').drop('city',axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# create level based on the PM2.5 data\n",
    "reg_monthavg['level'] = pd.cut(reg_monthavg['PM2.5'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['satisfactory', 'moderate','unhealthy'],\n",
    "                      include_lowest=True, right=False)\n",
    "# try if the bubble plot work (it didn't)\n",
    "reg_monthavg['size'] = [25 if x=='Bangkok' else 20 for x in reg_monthavg.City]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "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>City</th>\n",
       "      <th>Region</th>\n",
       "      <th>Population</th>\n",
       "      <th>Latitude</th>\n",
       "      <th>Longitude</th>\n",
       "      <th>x_mer</th>\n",
       "      <th>y_mer</th>\n",
       "      <th>level</th>\n",
       "      <th>size</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>33.498706</td>\n",
       "      <td>Ban Lam Luk Ka</td>\n",
       "      <td>Pathum Thani</td>\n",
       "      <td>60700</td>\n",
       "      <td>13.9774</td>\n",
       "      <td>100.7778</td>\n",
       "      <td>1.121853e+07</td>\n",
       "      <td>1.570430e+06</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11.950485</td>\n",
       "      <td>Ban Mai</td>\n",
       "      <td>Songkhla</td>\n",
       "      <td>86899</td>\n",
       "      <td>7.2041</td>\n",
       "      <td>100.5451</td>\n",
       "      <td>1.119263e+07</td>\n",
       "      <td>8.076416e+05</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>39.884466</td>\n",
       "      <td>Ban Pong</td>\n",
       "      <td>Ratchaburi</td>\n",
       "      <td>57559</td>\n",
       "      <td>13.8163</td>\n",
       "      <td>99.8774</td>\n",
       "      <td>1.111830e+07</td>\n",
       "      <td>1.552066e+06</td>\n",
       "      <td>moderate</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>16.423948</td>\n",
       "      <td>Ban Talat Yai</td>\n",
       "      <td>Phuket</td>\n",
       "      <td>52192</td>\n",
       "      <td>7.8848</td>\n",
       "      <td>98.4001</td>\n",
       "      <td>1.095385e+07</td>\n",
       "      <td>8.835903e+05</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>37.358900</td>\n",
       "      <td>Bang Kruai</td>\n",
       "      <td>Nonthaburi</td>\n",
       "      <td>78305</td>\n",
       "      <td>13.8050</td>\n",
       "      <td>100.4728</td>\n",
       "      <td>1.118458e+07</td>\n",
       "      <td>1.550779e+06</td>\n",
       "      <td>moderate</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>35.795146</td>\n",
       "      <td>Bangkok</td>\n",
       "      <td>Bangkok</td>\n",
       "      <td>5104476</td>\n",
       "      <td>13.7540</td>\n",
       "      <td>100.5014</td>\n",
       "      <td>1.118776e+07</td>\n",
       "      <td>1.544968e+06</td>\n",
       "      <td>moderate</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>30.250647</td>\n",
       "      <td>Chaiyaphum</td>\n",
       "      <td>Chaiyaphum</td>\n",
       "      <td>58350</td>\n",
       "      <td>15.8105</td>\n",
       "      <td>102.0288</td>\n",
       "      <td>1.135779e+07</td>\n",
       "      <td>1.780353e+06</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>23.726699</td>\n",
       "      <td>Chanthaburi</td>\n",
       "      <td>Chanthaburi</td>\n",
       "      <td>99819</td>\n",
       "      <td>12.6096</td>\n",
       "      <td>102.1045</td>\n",
       "      <td>1.136622e+07</td>\n",
       "      <td>1.414911e+06</td>\n",
       "      <td>satisfactory</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       PM2.5            City        Region  Population  Latitude  Longitude  \\\n",
       "0  33.498706  Ban Lam Luk Ka  Pathum Thani       60700   13.9774   100.7778   \n",
       "1  11.950485         Ban Mai      Songkhla       86899    7.2041   100.5451   \n",
       "2  39.884466        Ban Pong    Ratchaburi       57559   13.8163    99.8774   \n",
       "3  16.423948   Ban Talat Yai        Phuket       52192    7.8848    98.4001   \n",
       "4  37.358900      Bang Kruai    Nonthaburi       78305   13.8050   100.4728   \n",
       "5  35.795146         Bangkok       Bangkok     5104476   13.7540   100.5014   \n",
       "6  30.250647      Chaiyaphum    Chaiyaphum       58350   15.8105   102.0288   \n",
       "7  23.726699     Chanthaburi   Chanthaburi       99819   12.6096   102.1045   \n",
       "\n",
       "          x_mer         y_mer         level  size  \n",
       "0  1.121853e+07  1.570430e+06  satisfactory    20  \n",
       "1  1.119263e+07  8.076416e+05  satisfactory    20  \n",
       "2  1.111830e+07  1.552066e+06      moderate    20  \n",
       "3  1.095385e+07  8.835903e+05  satisfactory    20  \n",
       "4  1.118458e+07  1.550779e+06      moderate    20  \n",
       "5  1.118776e+07  1.544968e+06      moderate    25  \n",
       "6  1.135779e+07  1.780353e+06  satisfactory    20  \n",
       "7  1.136622e+07  1.414911e+06  satisfactory    20  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reg_monthavg.head(8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from bokeh.io import output_file, output_notebook, show, reset_output,export_png\n",
    "from bokeh.models import (\n",
    "  GMapPlot, GMapOptions, ColumnDataSource, Circle, LogColorMapper, BasicTicker, ColorBar,\n",
    "    DataRange1d, PanTool, WheelZoomTool, BoxSelectTool, CategoricalColorMapper, Slider, DateRangeSlider\n",
    ")\n",
    "from bokeh.models.mappers import ColorMapper, LinearColorMapper\n",
    "from bokeh.palettes import Viridis5\n",
    "from bokeh.plotting import figure, show, output_file\n",
    "from bokeh.tile_providers import STAMEN_TERRAIN,CARTODBPOSITRON_RETINA\n",
    "from bokeh.layouts import widgetbox\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "    <div class=\"bk-root\">\n",
       "        <a href=\"https://bokeh.pydata.org\" target=\"_blank\" class=\"bk-logo bk-logo-small bk-logo-notebook\"></a>\n",
       "        <span id=\"1003\">Loading BokehJS ...</span>\n",
       "    </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "(function(root) {\n",
       "  function now() {\n",
       "    return new Date();\n",
       "  }\n",
       "\n",
       "  var force = true;\n",
       "\n",
       "  if (typeof (root._bokeh_onload_callbacks) === \"undefined\" || force === true) {\n",
       "    root._bokeh_onload_callbacks = [];\n",
       "    root._bokeh_is_loading = undefined;\n",
       "  }\n",
       "\n",
       "  var JS_MIME_TYPE = 'application/javascript';\n",
       "  var HTML_MIME_TYPE = 'text/html';\n",
       "  var EXEC_MIME_TYPE = 'application/vnd.bokehjs_exec.v0+json';\n",
       "  var CLASS_NAME = 'output_bokeh rendered_html';\n",
       "\n",
       "  /**\n",
       "   * Render data to the DOM node\n",
       "   */\n",
       "  function render(props, node) {\n",
       "    var script = document.createElement(\"script\");\n",
       "    node.appendChild(script);\n",
       "  }\n",
       "\n",
       "  /**\n",
       "   * Handle when an output is cleared or removed\n",
       "   */\n",
       "  function handleClearOutput(event, handle) {\n",
       "    var cell = handle.cell;\n",
       "\n",
       "    var id = cell.output_area._bokeh_element_id;\n",
       "    var server_id = cell.output_area._bokeh_server_id;\n",
       "    // Clean up Bokeh references\n",
       "    if (id != null && id in Bokeh.index) {\n",
       "      Bokeh.index[id].model.document.clear();\n",
       "      delete Bokeh.index[id];\n",
       "    }\n",
       "\n",
       "    if (server_id !== undefined) {\n",
       "      // Clean up Bokeh references\n",
       "      var cmd = \"from bokeh.io.state import curstate; print(curstate().uuid_to_server['\" + server_id + \"'].get_sessions()[0].document.roots[0]._id)\";\n",
       "      cell.notebook.kernel.execute(cmd, {\n",
       "        iopub: {\n",
       "          output: function(msg) {\n",
       "            var id = msg.content.text.trim();\n",
       "            if (id in Bokeh.index) {\n",
       "              Bokeh.index[id].model.document.clear();\n",
       "              delete Bokeh.index[id];\n",
       "            }\n",
       "          }\n",
       "        }\n",
       "      });\n",
       "      // Destroy server and session\n",
       "      var cmd = \"import bokeh.io.notebook as ion; ion.destroy_server('\" + server_id + \"')\";\n",
       "      cell.notebook.kernel.execute(cmd);\n",
       "    }\n",
       "  }\n",
       "\n",
       "  /**\n",
       "   * Handle when a new output is added\n",
       "   */\n",
       "  function handleAddOutput(event, handle) {\n",
       "    var output_area = handle.output_area;\n",
       "    var output = handle.output;\n",
       "\n",
       "    // limit handleAddOutput to display_data with EXEC_MIME_TYPE content only\n",
       "    if ((output.output_type != \"display_data\") || (!output.data.hasOwnProperty(EXEC_MIME_TYPE))) {\n",
       "      return\n",
       "    }\n",
       "\n",
       "    var toinsert = output_area.element.find(\".\" + CLASS_NAME.split(' ')[0]);\n",
       "\n",
       "    if (output.metadata[EXEC_MIME_TYPE][\"id\"] !== undefined) {\n",
       "      toinsert[toinsert.length - 1].firstChild.textContent = output.data[JS_MIME_TYPE];\n",
       "      // store reference to embed id on output_area\n",
       "      output_area._bokeh_element_id = output.metadata[EXEC_MIME_TYPE][\"id\"];\n",
       "    }\n",
       "    if (output.metadata[EXEC_MIME_TYPE][\"server_id\"] !== undefined) {\n",
       "      var bk_div = document.createElement(\"div\");\n",
       "      bk_div.innerHTML = output.data[HTML_MIME_TYPE];\n",
       "      var script_attrs = bk_div.children[0].attributes;\n",
       "      for (var i = 0; i < script_attrs.length; i++) {\n",
       "        toinsert[toinsert.length - 1].firstChild.setAttribute(script_attrs[i].name, script_attrs[i].value);\n",
       "      }\n",
       "      // store reference to server id on output_area\n",
       "      output_area._bokeh_server_id = output.metadata[EXEC_MIME_TYPE][\"server_id\"];\n",
       "    }\n",
       "  }\n",
       "\n",
       "  function register_renderer(events, OutputArea) {\n",
       "\n",
       "    function append_mime(data, metadata, element) {\n",
       "      // create a DOM node to render to\n",
       "      var toinsert = this.create_output_subarea(\n",
       "        metadata,\n",
       "        CLASS_NAME,\n",
       "        EXEC_MIME_TYPE\n",
       "      );\n",
       "      this.keyboard_manager.register_events(toinsert);\n",
       "      // Render to node\n",
       "      var props = {data: data, metadata: metadata[EXEC_MIME_TYPE]};\n",
       "      render(props, toinsert[toinsert.length - 1]);\n",
       "      element.append(toinsert);\n",
       "      return toinsert\n",
       "    }\n",
       "\n",
       "    /* Handle when an output is cleared or removed */\n",
       "    events.on('clear_output.CodeCell', handleClearOutput);\n",
       "    events.on('delete.Cell', handleClearOutput);\n",
       "\n",
       "    /* Handle when a new output is added */\n",
       "    events.on('output_added.OutputArea', handleAddOutput);\n",
       "\n",
       "    /**\n",
       "     * Register the mime type and append_mime function with output_area\n",
       "     */\n",
       "    OutputArea.prototype.register_mime_type(EXEC_MIME_TYPE, append_mime, {\n",
       "      /* Is output safe? */\n",
       "      safe: true,\n",
       "      /* Index of renderer in `output_area.display_order` */\n",
       "      index: 0\n",
       "    });\n",
       "  }\n",
       "\n",
       "  // register the mime type if in Jupyter Notebook environment and previously unregistered\n",
       "  if (root.Jupyter !== undefined) {\n",
       "    var events = require('base/js/events');\n",
       "    var OutputArea = require('notebook/js/outputarea').OutputArea;\n",
       "\n",
       "    if (OutputArea.prototype.mime_types().indexOf(EXEC_MIME_TYPE) == -1) {\n",
       "      register_renderer(events, OutputArea);\n",
       "    }\n",
       "  }\n",
       "\n",
       "  \n",
       "  if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n",
       "    root._bokeh_timeout = Date.now() + 5000;\n",
       "    root._bokeh_failed_load = false;\n",
       "  }\n",
       "\n",
       "  var NB_LOAD_WARNING = {'data': {'text/html':\n",
       "     \"<div style='background-color: #fdd'>\\n\"+\n",
       "     \"<p>\\n\"+\n",
       "     \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n",
       "     \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n",
       "     \"</p>\\n\"+\n",
       "     \"<ul>\\n\"+\n",
       "     \"<li>re-rerun `output_notebook()` to attempt to load from CDN again, or</li>\\n\"+\n",
       "     \"<li>use INLINE resources instead, as so:</li>\\n\"+\n",
       "     \"</ul>\\n\"+\n",
       "     \"<code>\\n\"+\n",
       "     \"from bokeh.resources import INLINE\\n\"+\n",
       "     \"output_notebook(resources=INLINE)\\n\"+\n",
       "     \"</code>\\n\"+\n",
       "     \"</div>\"}};\n",
       "\n",
       "  function display_loaded() {\n",
       "    var el = document.getElementById(\"1003\");\n",
       "    if (el != null) {\n",
       "      el.textContent = \"BokehJS is loading...\";\n",
       "    }\n",
       "    if (root.Bokeh !== undefined) {\n",
       "      if (el != null) {\n",
       "        el.textContent = \"BokehJS \" + root.Bokeh.version + \" successfully loaded.\";\n",
       "      }\n",
       "    } else if (Date.now() < root._bokeh_timeout) {\n",
       "      setTimeout(display_loaded, 100)\n",
       "    }\n",
       "  }\n",
       "\n",
       "\n",
       "  function run_callbacks() {\n",
       "    try {\n",
       "      root._bokeh_onload_callbacks.forEach(function(callback) { callback() });\n",
       "    }\n",
       "    finally {\n",
       "      delete root._bokeh_onload_callbacks\n",
       "    }\n",
       "    console.info(\"Bokeh: all callbacks have finished\");\n",
       "  }\n",
       "\n",
       "  function load_libs(js_urls, callback) {\n",
       "    root._bokeh_onload_callbacks.push(callback);\n",
       "    if (root._bokeh_is_loading > 0) {\n",
       "      console.log(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n",
       "      return null;\n",
       "    }\n",
       "    if (js_urls == null || js_urls.length === 0) {\n",
       "      run_callbacks();\n",
       "      return null;\n",
       "    }\n",
       "    console.log(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n",
       "    root._bokeh_is_loading = js_urls.length;\n",
       "    for (var i = 0; i < js_urls.length; i++) {\n",
       "      var url = js_urls[i];\n",
       "      var s = document.createElement('script');\n",
       "      s.src = url;\n",
       "      s.async = false;\n",
       "      s.onreadystatechange = s.onload = function() {\n",
       "        root._bokeh_is_loading--;\n",
       "        if (root._bokeh_is_loading === 0) {\n",
       "          console.log(\"Bokeh: all BokehJS libraries loaded\");\n",
       "          run_callbacks()\n",
       "        }\n",
       "      };\n",
       "      s.onerror = function() {\n",
       "        console.warn(\"failed to load library \" + url);\n",
       "      };\n",
       "      console.log(\"Bokeh: injecting script tag for BokehJS library: \", url);\n",
       "      document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
       "    }\n",
       "  };var element = document.getElementById(\"1003\");\n",
       "  if (element == null) {\n",
       "    console.log(\"Bokeh: ERROR: autoload.js configured with elementid '1003' but no matching script tag was found. \")\n",
       "    return false;\n",
       "  }\n",
       "\n",
       "  var js_urls = [\"https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-gl-1.0.4.min.js\"];\n",
       "\n",
       "  var inline_js = [\n",
       "    function(Bokeh) {\n",
       "      Bokeh.set_log_level(\"info\");\n",
       "    },\n",
       "    \n",
       "    function(Bokeh) {\n",
       "      \n",
       "    },\n",
       "    function(Bokeh) {\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.css\");\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.css\");\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.css\");\n",
       "    }\n",
       "  ];\n",
       "\n",
       "  function run_inline_js() {\n",
       "    \n",
       "    if ((root.Bokeh !== undefined) || (force === true)) {\n",
       "      for (var i = 0; i < inline_js.length; i++) {\n",
       "        inline_js[i].call(root, root.Bokeh);\n",
       "      }if (force === true) {\n",
       "        display_loaded();\n",
       "      }} else if (Date.now() < root._bokeh_timeout) {\n",
       "      setTimeout(run_inline_js, 100);\n",
       "    } else if (!root._bokeh_failed_load) {\n",
       "      console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n",
       "      root._bokeh_failed_load = true;\n",
       "    } else if (force !== true) {\n",
       "      var cell = $(document.getElementById(\"1003\")).parents('.cell').data().cell;\n",
       "      cell.output_area.append_execute_result(NB_LOAD_WARNING)\n",
       "    }\n",
       "\n",
       "  }\n",
       "\n",
       "  if (root._bokeh_is_loading === 0) {\n",
       "    console.log(\"Bokeh: BokehJS loaded, going straight to plotting\");\n",
       "    run_inline_js();\n",
       "  } else {\n",
       "    load_libs(js_urls, function() {\n",
       "      console.log(\"Bokeh: BokehJS plotting callback run at\", now());\n",
       "      run_inline_js();\n",
       "    });\n",
       "  }\n",
       "}(window));"
      ],
      "application/vnd.bokehjs_load.v0+json": "\n(function(root) {\n  function now() {\n    return new Date();\n  }\n\n  var force = true;\n\n  if (typeof (root._bokeh_onload_callbacks) === \"undefined\" || force === true) {\n    root._bokeh_onload_callbacks = [];\n    root._bokeh_is_loading = undefined;\n  }\n\n  \n\n  \n  if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n    root._bokeh_timeout = Date.now() + 5000;\n    root._bokeh_failed_load = false;\n  }\n\n  var NB_LOAD_WARNING = {'data': {'text/html':\n     \"<div style='background-color: #fdd'>\\n\"+\n     \"<p>\\n\"+\n     \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n     \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n     \"</p>\\n\"+\n     \"<ul>\\n\"+\n     \"<li>re-rerun `output_notebook()` to attempt to load from CDN again, or</li>\\n\"+\n     \"<li>use INLINE resources instead, as so:</li>\\n\"+\n     \"</ul>\\n\"+\n     \"<code>\\n\"+\n     \"from bokeh.resources import INLINE\\n\"+\n     \"output_notebook(resources=INLINE)\\n\"+\n     \"</code>\\n\"+\n     \"</div>\"}};\n\n  function display_loaded() {\n    var el = document.getElementById(\"1003\");\n    if (el != null) {\n      el.textContent = \"BokehJS is loading...\";\n    }\n    if (root.Bokeh !== undefined) {\n      if (el != null) {\n        el.textContent = \"BokehJS \" + root.Bokeh.version + \" successfully loaded.\";\n      }\n    } else if (Date.now() < root._bokeh_timeout) {\n      setTimeout(display_loaded, 100)\n    }\n  }\n\n\n  function run_callbacks() {\n    try {\n      root._bokeh_onload_callbacks.forEach(function(callback) { callback() });\n    }\n    finally {\n      delete root._bokeh_onload_callbacks\n    }\n    console.info(\"Bokeh: all callbacks have finished\");\n  }\n\n  function load_libs(js_urls, callback) {\n    root._bokeh_onload_callbacks.push(callback);\n    if (root._bokeh_is_loading > 0) {\n      console.log(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n      return null;\n    }\n    if (js_urls == null || js_urls.length === 0) {\n      run_callbacks();\n      return null;\n    }\n    console.log(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n    root._bokeh_is_loading = js_urls.length;\n    for (var i = 0; i < js_urls.length; i++) {\n      var url = js_urls[i];\n      var s = document.createElement('script');\n      s.src = url;\n      s.async = false;\n      s.onreadystatechange = s.onload = function() {\n        root._bokeh_is_loading--;\n        if (root._bokeh_is_loading === 0) {\n          console.log(\"Bokeh: all BokehJS libraries loaded\");\n          run_callbacks()\n        }\n      };\n      s.onerror = function() {\n        console.warn(\"failed to load library \" + url);\n      };\n      console.log(\"Bokeh: injecting script tag for BokehJS library: \", url);\n      document.getElementsByTagName(\"head\")[0].appendChild(s);\n    }\n  };var element = document.getElementById(\"1003\");\n  if (element == null) {\n    console.log(\"Bokeh: ERROR: autoload.js configured with elementid '1003' but no matching script tag was found. \")\n    return false;\n  }\n\n  var js_urls = [\"https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-gl-1.0.4.min.js\"];\n\n  var inline_js = [\n    function(Bokeh) {\n      Bokeh.set_log_level(\"info\");\n    },\n    \n    function(Bokeh) {\n      \n    },\n    function(Bokeh) {\n      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.css\");\n      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.css\");\n      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.css\");\n      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.css\");\n      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.css\");\n      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.css\");\n    }\n  ];\n\n  function run_inline_js() {\n    \n    if ((root.Bokeh !== undefined) || (force === true)) {\n      for (var i = 0; i < inline_js.length; i++) {\n        inline_js[i].call(root, root.Bokeh);\n      }if (force === true) {\n        display_loaded();\n      }} else if (Date.now() < root._bokeh_timeout) {\n      setTimeout(run_inline_js, 100);\n    } else if (!root._bokeh_failed_load) {\n      console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n      root._bokeh_failed_load = true;\n    } else if (force !== true) {\n      var cell = $(document.getElementById(\"1003\")).parents('.cell').data().cell;\n      cell.output_area.append_execute_result(NB_LOAD_WARNING)\n    }\n\n  }\n\n  if (root._bokeh_is_loading === 0) {\n    console.log(\"Bokeh: BokehJS loaded, going straight to plotting\");\n    run_inline_js();\n  } else {\n    load_libs(js_urls, function() {\n      console.log(\"Bokeh: BokehJS plotting callback run at\", now());\n      run_inline_js();\n    });\n  }\n}(window));"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "output_notebook()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "11187764.672011103 1544967.9205643306 100000.0 100000.0\n"
     ]
    }
   ],
   "source": [
    "# center the plot on bangkok, using the coordinate of BKK\n",
    "[[bkk_x, bkk_y]]= region_df.query('City == \"Bangkok\"')[['x_mer','y_mer']].values\n",
    "stepx = 1E5\n",
    "stepy = stepx\n",
    "\n",
    "print(bkk_x,bkk_y,stepx,stepy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "\n",
       "\n",
       "\n",
       "\n",
       "\n",
       "  <div class=\"bk-root\" id=\"f533b890-765f-44ce-a5fc-2e34c1f3eed1\" data-root-id=\"7062\"></div>\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "(function(root) {\n",
       "  function embed_document(root) {\n",
       "    \n",
       "  var docs_json = {\"d3c508a5-5b37-416c-a2e9-c5b9aa1e2f90\":{\"roots\":{\"references\":[{\"attributes\":{\"below\":[{\"id\":\"7072\",\"type\":\"MercatorAxis\"}],\"left\":[{\"id\":\"7081\",\"type\":\"MercatorAxis\"}],\"renderers\":[{\"id\":\"7072\",\"type\":\"MercatorAxis\"},{\"id\":\"7080\",\"type\":\"Grid\"},{\"id\":\"7081\",\"type\":\"MercatorAxis\"},{\"id\":\"7089\",\"type\":\"Grid\"},{\"id\":\"7098\",\"type\":\"BoxAnnotation\"},{\"id\":\"7105\",\"type\":\"TileRenderer\"},{\"id\":\"7120\",\"type\":\"Legend\"},{\"id\":\"7112\",\"type\":\"GlyphRenderer\"}],\"title\":{\"id\":\"7061\",\"type\":\"Title\"},\"toolbar\":{\"id\":\"7096\",\"type\":\"Toolbar\"},\"x_range\":{\"id\":\"7064\",\"type\":\"Range1d\"},\"x_scale\":{\"id\":\"7068\",\"type\":\"LinearScale\"},\"y_range\":{\"id\":\"7066\",\"type\":\"Range1d\"},\"y_scale\":{\"id\":\"7070\",\"type\":\"LinearScale\"}},\"id\":\"7062\",\"subtype\":\"Figure\",\"type\":\"Plot\"},{\"attributes\":{},\"id\":\"7094\",\"type\":\"ResetTool\"},{\"attributes\":{\"plot\":{\"id\":\"7062\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"7073\",\"type\":\"MercatorTicker\"}},\"id\":\"7080\",\"type\":\"Grid\"},{\"attributes\":{\"callback\":null,\"end\":1644967.9205643306,\"start\":1444967.9205643306},\"id\":\"7066\",\"type\":\"Range1d\"},{\"attributes\":{\"callback\":null,\"end\":11337764.672011103,\"start\":11087764.672011103},\"id\":\"7064\",\"type\":\"Range1d\"},{\"attributes\":{},\"id\":\"7070\",\"type\":\"LinearScale\"},{\"attributes\":{\"label\":{\"field\":\"level\"},\"renderers\":[{\"id\":\"7112\",\"type\":\"GlyphRenderer\"}]},\"id\":\"7121\",\"type\":\"LegendItem\"},{\"attributes\":{\"fill_alpha\":{\"value\":0.1},\"fill_color\":{\"value\":\"#1f77b4\"},\"line_alpha\":{\"value\":0.1},\"line_color\":{\"value\":\"#1f77b4\"},\"size\":{\"field\":\"size\",\"units\":\"screen\"},\"x\":{\"field\":\"x_mer\"},\"y\":{\"field\":\"y_mer\"}},\"id\":\"7111\",\"type\":\"Circle\"},{\"attributes\":{},\"id\":\"7091\",\"type\":\"WheelZoomTool\"},{\"attributes\":{\"factors\":[\"satisfactory\",\"moderate\",\"unhealthy\"],\"palette\":[\"green\",\"orange\",\"red\"]},\"id\":\"7108\",\"type\":\"CategoricalColorMapper\"},{\"attributes\":{},\"id\":\"7152\",\"type\":\"UnionRenderers\"},{\"attributes\":{},\"id\":\"7093\",\"type\":\"SaveTool\"},{\"attributes\":{\"fill_alpha\":{\"value\":0.7},\"fill_color\":{\"field\":\"level\",\"transform\":{\"id\":\"7108\",\"type\":\"CategoricalColorMapper\"}},\"line_alpha\":{\"value\":0.7},\"line_color\":{\"field\":\"level\",\"transform\":{\"id\":\"7108\",\"type\":\"CategoricalColorMapper\"}},\"size\":{\"field\":\"size\",\"units\":\"screen\"},\"x\":{\"field\":\"x_mer\"},\"y\":{\"field\":\"y_mer\"}},\"id\":\"7110\",\"type\":\"Circle\"},{\"attributes\":{\"callback\":null,\"data\":{\"City\":[\"Ban Lam Luk Ka\",\"Ban Mai\",\"Ban Pong\",\"Ban Talat Yai\",\"Bang Kruai\",\"Bangkok\",\"Chaiyaphum\",\"Chanthaburi\",\"Chiang Mai\",\"Chiang Rai\",\"Chon Buri\",\"Chumphon\",\"Hat Yai\",\"Hua Hin\",\"Kalasin\",\"Kamphaeng Phet\",\"Kanchanaburi\",\"Khlong Luang\",\"Khon Kaen\",\"Klaeng\",\"Ko Samui\",\"Krathum Baen\",\"Lampang\",\"Lop Buri\",\"Maha Sarakham\",\"Mueang Nonthaburi\",\"Nakhon Pathom\",\"Nakhon Ratchasima\",\"Nakhon Sawan\",\"Nakhon Si Thammarat\",\"Nong Khai\",\"Pak Kret\",\"Phatthaya\",\"Phetchabun\",\"Phitsanulok\",\"Phra Nakhon Si Ayutthaya\",\"Phra Phutthabat\",\"Phra Pradaeng\",\"Phuket\",\"Ratchaburi\",\"Rayong\",\"Sakon Nakhon\",\"Samut Prakan\",\"Samut Sakhon\",\"Saraburi\",\"Sattahip\",\"Si Racha\",\"Songkhla\",\"Suphan Buri\",\"Surat Thani\",\"Tha Maka\",\"Trang\",\"Ubon Ratchathani\",\"Udon Thani\",\"Uttaradit\",\"Yala\"],\"Latitude\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[56]},\"Longitude\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[56]},\"PM2.5\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[56]},\"Population\":[60700,86899,57559,52192,78305,5104476,58350,99819,200952,78756,219164,55835,191696,50456,55102,58787,63699,118551,147579,55619,50000,72819,156139,57761,51584,291555,117927,208781,91802,120836,63609,182926,97296,50656,103427,81962,57008,171333,89072,92448,106737,76237,388920,63498,67763,65444,178916,84264,53399,127201,52907,66713,122533,247231,58313,93558],\"Region\":[\"Pathum Thani\",\"Songkhla\",\"Ratchaburi\",\"Phuket\",\"Nonthaburi\",\"Bangkok\",\"Chaiyaphum\",\"Chanthaburi\",\"Chiang Mai\",\"Chiang Rai\",\"Chon Buri\",\"Chumphon\",\"Songkhla\",\"Prachuap Khiri Khan\",\"Kalasin\",\"Kamphaeng Phet\",\"Kanchanaburi\",\"Pathum Thani\",\"Khon Kaen\",\"Rayong\",\"Surat Thani\",\"Samut Sakhon\",\"Lampang\",\"Lop Buri\",\"Maha Sarakham\",\"Nonthaburi\",\"Nakhon Pathom\",\"Nakhon Ratchasima\",\"Nakhon Sawan\",\"Nakhon Si Thammarat\",\"Nong Khai\",\"Nonthaburi\",\"TH.S\",\"Phetchabun\",\"Phitsanulok\",\"Phra Nakhon Si Ayutthaya\",\"Sara Buri\",\"Samut Prakan\",\"Phuket\",\"Ratchaburi\",\"Rayong\",\"Sakon Nakhon\",\"Samut Prakan\",\"Samut Sakhon\",\"Sara Buri\",\"Chon Buri\",\"Chon Buri\",\"Songkhla\",\"Suphan Buri\",\"Surat Thani\",\"Kanchanaburi\",\"Trang\",\"Changwat Ubon Ratchathani\",\"Changwat Udon Thani\",\"Uttaradit\",\"Yala\"],\"index\":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55],\"level\":[\"moderate\",\"satisfactory\",\"unhealthy\",\"satisfactory\",\"moderate\",\"moderate\",\"moderate\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"moderate\",\"satisfactory\",\"satisfactory\",\"moderate\",\"satisfactory\",\"moderate\",\"moderate\",\"moderate\",\"satisfactory\",\"moderate\",\"satisfactory\",\"unhealthy\",\"satisfactory\",\"moderate\",\"satisfactory\",\"moderate\",\"unhealthy\",\"moderate\",\"moderate\",\"satisfactory\",\"satisfactory\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"satisfactory\",\"unhealthy\",\"moderate\",\"satisfactory\",\"moderate\",\"unhealthy\",\"moderate\",\"moderate\",\"moderate\",\"satisfactory\",\"moderate\",\"satisfactory\",\"unhealthy\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\"],\"size\":[20,20,20,20,20,25,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20,20],\"x_mer\":{\"__ndarray__\":\"OPMirMxlZUHhJq6qJlllQZ+F6anbNGVBDA3VII/kZEGcCOidOFVlQW4dgZXGVmVBF/H9QcypZUG7klOe6a1lQUR2z8xVBGVB4VwR4moyZUEmf435+nBlQbPVbjbvDmVBWy6U4m5VZUH7++tWSDllQQ4ML8Af+mVBvNV+VoMhZUGZOxdC+CJlQWl/m+afXmVBirtLpoLVZUFmP9+fHZVlQU/hjecGOGVBSOYIWKNJZUGj06z68x9lQQOA1wASX2VBkqVSqu7uZUFcUE0LgVdlQaByDBHuPWVB12bE1MatZUGrf6HD+kJlQcA0fUS2OWVBvUrwY5DQZUFEBLlnolZlQfzIvLKIa2VBQPZeT5t6ZUFCtJYJlEllQeWGgR7oWmVBuhNGk8FmZUEtkBulhlhlQYqDYkxz5GRBb6DSl5QxZUGxuPWf5n9lQZp/sKjxHGZBXlalEfZbZUGYu/HCZEplQQtbAnVZbWVB8g5u0XpsZUG60gfVIW5lQa+U3WneW2VBU3hGCipCZUF8oozDRRdlQYh9v0fXLmVBgxRoSmYmZUEBaJ78EkNmQbsuj0Hz0mVBJvMCbKRAZUHprcXXKoFlQQ==\",\"dtype\":\"float64\",\"shape\":[56]},\"y_mer\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[56]}},\"selected\":{\"id\":\"7153\",\"type\":\"Selection\"},\"selection_policy\":{\"id\":\"7152\",\"type\":\"UnionRenderers\"}},\"id\":\"7107\",\"type\":\"ColumnDataSource\"},{\"attributes\":{\"formatter\":{\"id\":\"7084\",\"type\":\"MercatorTickFormatter\"},\"plot\":{\"id\":\"7062\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"7082\",\"type\":\"MercatorTicker\"}},\"id\":\"7081\",\"type\":\"MercatorAxis\"},{\"attributes\":{\"overlay\":{\"id\":\"7098\",\"type\":\"BoxAnnotation\"}},\"id\":\"7092\",\"type\":\"BoxZoomTool\"},{\"attributes\":{},\"id\":\"7090\",\"type\":\"PanTool\"},{\"attributes\":{\"plot\":null,\"text\":\"PM 2.5 Pollution January 2019\"},\"id\":\"7061\",\"type\":\"Title\"},{\"attributes\":{\"dimension\":\"lon\"},\"id\":\"7075\",\"type\":\"MercatorTickFormatter\"},{\"attributes\":{\"attribution\":\"Map tiles by <a href=\\\"https://stamen.com\\\">Stamen Design</a>, under <a href=\\\"https://creativecommons.org/licenses/by/3.0\\\">CC BY 3.0</a>. Data by <a href=\\\"https://openstreetmap.org\\\">OpenStreetMap</a>, under <a href=\\\"https://creativecommons.org/licenses/by-sa/3.0\\\">CC BY SA</a>.\",\"url\":\"http://tile.stamen.com/terrain/{Z}/{X}/{Y}.png\"},\"id\":\"1699\",\"type\":\"WMTSTileSource\"},{\"attributes\":{},\"id\":\"7153\",\"type\":\"Selection\"},{\"attributes\":{\"tile_source\":{\"id\":\"1699\",\"type\":\"WMTSTileSource\"}},\"id\":\"7105\",\"type\":\"TileRenderer\"},{\"attributes\":{\"source\":{\"id\":\"7107\",\"type\":\"ColumnDataSource\"}},\"id\":\"7113\",\"type\":\"CDSView\"},{\"attributes\":{\"dimension\":\"lon\"},\"id\":\"7073\",\"type\":\"MercatorTicker\"},{\"attributes\":{\"dimension\":\"lat\"},\"id\":\"7082\",\"type\":\"MercatorTicker\"},{\"attributes\":{\"data_source\":{\"id\":\"7107\",\"type\":\"ColumnDataSource\"},\"glyph\":{\"id\":\"7110\",\"type\":\"Circle\"},\"hover_glyph\":null,\"muted_glyph\":null,\"nonselection_glyph\":{\"id\":\"7111\",\"type\":\"Circle\"},\"selection_glyph\":null,\"view\":{\"id\":\"7113\",\"type\":\"CDSView\"}},\"id\":\"7112\",\"type\":\"GlyphRenderer\"},{\"attributes\":{\"dimension\":\"lat\"},\"id\":\"7084\",\"type\":\"MercatorTickFormatter\"},{\"attributes\":{\"dimension\":1,\"plot\":{\"id\":\"7062\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"7082\",\"type\":\"MercatorTicker\"}},\"id\":\"7089\",\"type\":\"Grid\"},{\"attributes\":{\"bottom_units\":\"screen\",\"fill_alpha\":{\"value\":0.5},\"fill_color\":{\"value\":\"lightgrey\"},\"left_units\":\"screen\",\"level\":\"overlay\",\"line_alpha\":{\"value\":1.0},\"line_color\":{\"value\":\"black\"},\"line_dash\":[4,4],\"line_width\":{\"value\":2},\"plot\":null,\"render_mode\":\"css\",\"right_units\":\"screen\",\"top_units\":\"screen\"},\"id\":\"7098\",\"type\":\"BoxAnnotation\"},{\"attributes\":{\"formatter\":{\"id\":\"7075\",\"type\":\"MercatorTickFormatter\"},\"plot\":{\"id\":\"7062\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"7073\",\"type\":\"MercatorTicker\"}},\"id\":\"7072\",\"type\":\"MercatorAxis\"},{\"attributes\":{},\"id\":\"7068\",\"type\":\"LinearScale\"},{\"attributes\":{\"active_drag\":\"auto\",\"active_inspect\":\"auto\",\"active_multi\":null,\"active_scroll\":\"auto\",\"active_tap\":\"auto\",\"tools\":[{\"id\":\"7090\",\"type\":\"PanTool\"},{\"id\":\"7091\",\"type\":\"WheelZoomTool\"},{\"id\":\"7092\",\"type\":\"BoxZoomTool\"},{\"id\":\"7093\",\"type\":\"SaveTool\"},{\"id\":\"7094\",\"type\":\"ResetTool\"},{\"id\":\"7095\",\"type\":\"HelpTool\"}]},\"id\":\"7096\",\"type\":\"Toolbar\"},{\"attributes\":{},\"id\":\"7095\",\"type\":\"HelpTool\"},{\"attributes\":{\"items\":[{\"id\":\"7121\",\"type\":\"LegendItem\"}],\"plot\":{\"id\":\"7062\",\"subtype\":\"Figure\",\"type\":\"Plot\"}},\"id\":\"7120\",\"type\":\"Legend\"}],\"root_ids\":[\"7062\"]},\"title\":\"Bokeh Application\",\"version\":\"1.0.4\"}};\n",
       "  var render_items = [{\"docid\":\"d3c508a5-5b37-416c-a2e9-c5b9aa1e2f90\",\"roots\":{\"7062\":\"f533b890-765f-44ce-a5fc-2e34c1f3eed1\"}}];\n",
       "  root.Bokeh.embed.embed_items_notebook(docs_json, render_items);\n",
       "\n",
       "  }\n",
       "  if (root.Bokeh !== undefined) {\n",
       "    embed_document(root);\n",
       "  } else {\n",
       "    var attempts = 0;\n",
       "    var timer = setInterval(function(root) {\n",
       "      if (root.Bokeh !== undefined) {\n",
       "        embed_document(root);\n",
       "        clearInterval(timer);\n",
       "      }\n",
       "      attempts++;\n",
       "      if (attempts > 100) {\n",
       "        console.log(\"Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing\");\n",
       "        clearInterval(timer);\n",
       "      }\n",
       "    }, 10, root)\n",
       "  }\n",
       "})(window);"
      ],
      "application/vnd.bokehjs_exec.v0+json": ""
     },
     "metadata": {
      "application/vnd.bokehjs_exec.v0+json": {
       "id": "7062"
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# range bounds supplied in web mercator coordinates\n",
    "p = figure(x_range=(bkk_x-stepx,bkk_x+1.5*stepx), y_range=(bkk_y-stepy, bkk_y+stepy),\n",
    "           x_axis_type=\"mercator\", y_axis_type=\"mercator\",\n",
    "          title = 'PM 2.5 Pollution January 2019')\n",
    "p.add_tile(STAMEN_TERRAIN)\n",
    "\n",
    "source = ColumnDataSource(reg_monthavg)\n",
    "color_mapper = CategoricalColorMapper(factors=['satisfactory', 'moderate', 'unhealthy'],\n",
    "                                      palette=['green', 'orange', 'red'])\n",
    "\n",
    "p.circle(source=source, x='x_mer',y='y_mer', size='size', alpha=0.7,\n",
    "        color=dict(field='level',transform=color_mapper), legend='level')\n",
    "\n",
    "\n",
    "show(p)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'C:\\\\Users\\\\Benny\\\\Documents\\\\Fern\\\\BKK-AQI\\\\bkk_jan2019.png'"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "export_png(p,filename='bkk_jan2019.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Building interactive maps \n",
    "- Show monthly average air pollution for the entire country\n",
    "- Plot the monthly average on the side of the map.\n",
    "- Plot the yearly average of each province, rank them to find the the provinces with has the highest and lowest air pollutions?\n",
    "- Turn the maps into gif"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from bokeh.io import output_file, output_notebook, show,curdoc, reset_output,export_png\n",
    "from bokeh.models import (\n",
    "  GMapPlot, GMapOptions, ColumnDataSource, Circle, LogColorMapper, BasicTicker, ColorBar,\n",
    "    DataRange1d, PanTool, WheelZoomTool, BoxSelectTool, CategoricalColorMapper, Slider, DateRangeSlider,\n",
    "    DateSlider, SingleIntervalTicker, LinearAxis,Legend, LegendItem\n",
    ")\n",
    "from bokeh.models.mappers import ColorMapper, LinearColorMapper\n",
    "from bokeh.palettes import Viridis5\n",
    "from bokeh.plotting import figure, show, output_file\n",
    "from bokeh.tile_providers import STAMEN_TERRAIN,CARTODBPOSITRON_RETINA\n",
    "from bokeh.layouts import widgetbox,row, column, gridplot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "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>Time</th>\n",
       "      <th>city</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>62.9</td>\n",
       "      <td>2016-03-03 15:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>62.9</td>\n",
       "      <td>2016-03-03 16:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>55.5</td>\n",
       "      <td>2016-03-03 17:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>55.5</td>\n",
       "      <td>2016-03-03 18:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>47.9</td>\n",
       "      <td>2016-03-03 19:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PM2.5                Time     city\n",
       "0   62.9 2016-03-03 15:00:00  Bangkok\n",
       "1   62.9 2016-03-03 16:00:00  Bangkok\n",
       "2   55.5 2016-03-03 17:00:00  Bangkok\n",
       "3   55.5 2016-03-03 18:00:00  Bangkok\n",
       "4   47.9 2016-03-03 19:00:00  Bangkok"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "region_df = pd.read_csv('data/region_info.csv',sep=';')\n",
    "pm25_all = pd.read_csv('data/pm25_all.csv').dropna()\n",
    "pm25_all['Time'] = pd.to_datetime(pm25_all['Time'])\n",
    "pm25_all.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "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>Time</th>\n",
       "      <th>city</th>\n",
       "      <th>month</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1266694</th>\n",
       "      <td>19.1</td>\n",
       "      <td>2019-03-20 16:00:00</td>\n",
       "      <td>Yala</td>\n",
       "      <td>2019-03</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1266695</th>\n",
       "      <td>15.8</td>\n",
       "      <td>2019-03-20 17:00:00</td>\n",
       "      <td>Yala</td>\n",
       "      <td>2019-03</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1266696</th>\n",
       "      <td>14.9</td>\n",
       "      <td>2019-03-20 18:00:00</td>\n",
       "      <td>Yala</td>\n",
       "      <td>2019-03</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1266697</th>\n",
       "      <td>17.7</td>\n",
       "      <td>2019-03-20 19:00:00</td>\n",
       "      <td>Yala</td>\n",
       "      <td>2019-03</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1266698</th>\n",
       "      <td>16.8</td>\n",
       "      <td>2019-03-20 20:00:00</td>\n",
       "      <td>Yala</td>\n",
       "      <td>2019-03</td>\n",
       "      <td>2019</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         PM2.5                Time  city    month  year\n",
       "1266694   19.1 2019-03-20 16:00:00  Yala  2019-03  2019\n",
       "1266695   15.8 2019-03-20 17:00:00  Yala  2019-03  2019\n",
       "1266696   14.9 2019-03-20 18:00:00  Yala  2019-03  2019\n",
       "1266697   17.7 2019-03-20 19:00:00  Yala  2019-03  2019\n",
       "1266698   16.8 2019-03-20 20:00:00  Yala  2019-03  2019"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm25_all.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "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>Time</th>\n",
       "      <th>city</th>\n",
       "      <th>month</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>62.9</td>\n",
       "      <td>2016-03-03 15:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "      <td>2016-03</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>62.9</td>\n",
       "      <td>2016-03-03 16:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "      <td>2016-03</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>55.5</td>\n",
       "      <td>2016-03-03 17:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "      <td>2016-03</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>55.5</td>\n",
       "      <td>2016-03-03 18:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "      <td>2016-03</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>47.9</td>\n",
       "      <td>2016-03-03 19:00:00</td>\n",
       "      <td>Bangkok</td>\n",
       "      <td>2016-03</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PM2.5                Time     city    month  year\n",
       "0   62.9 2016-03-03 15:00:00  Bangkok  2016-03  2016\n",
       "1   62.9 2016-03-03 16:00:00  Bangkok  2016-03  2016\n",
       "2   55.5 2016-03-03 17:00:00  Bangkok  2016-03  2016\n",
       "3   55.5 2016-03-03 18:00:00  Bangkok  2016-03  2016\n",
       "4   47.9 2016-03-03 19:00:00  Bangkok  2016-03  2016"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pm25_all['month'] = pm25_all['Time'].dt.to_period('M')\n",
    "pm25_all['year'] = pm25_all['Time'].dt.to_period('Y')\n",
    "pm25_all.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "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>month</th>\n",
       "      <th>PM2.5</th>\n",
       "      <th>Region</th>\n",
       "      <th>x_mer</th>\n",
       "      <th>y_mer</th>\n",
       "      <th>dis2BKK</th>\n",
       "      <th>level</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1841</th>\n",
       "      <td>2019-02</td>\n",
       "      <td>17.980260</td>\n",
       "      <td>Yala</td>\n",
       "      <td>1.127458e+07</td>\n",
       "      <td>733657.663071</td>\n",
       "      <td>815942.223042</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1840</th>\n",
       "      <td>2019-01</td>\n",
       "      <td>18.564031</td>\n",
       "      <td>Yala</td>\n",
       "      <td>1.127458e+07</td>\n",
       "      <td>733657.663071</td>\n",
       "      <td>815942.223042</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1839</th>\n",
       "      <td>2018-12</td>\n",
       "      <td>14.292233</td>\n",
       "      <td>Yala</td>\n",
       "      <td>1.127458e+07</td>\n",
       "      <td>733657.663071</td>\n",
       "      <td>815942.223042</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1838</th>\n",
       "      <td>2018-11</td>\n",
       "      <td>16.203863</td>\n",
       "      <td>Yala</td>\n",
       "      <td>1.127458e+07</td>\n",
       "      <td>733657.663071</td>\n",
       "      <td>815942.223042</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1823</th>\n",
       "      <td>2017-08</td>\n",
       "      <td>16.498505</td>\n",
       "      <td>Yala</td>\n",
       "      <td>1.127458e+07</td>\n",
       "      <td>733657.663071</td>\n",
       "      <td>815942.223042</td>\n",
       "      <td>satisfactory</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        month      PM2.5 Region         x_mer          y_mer        dis2BKK  \\\n",
       "1841  2019-02  17.980260   Yala  1.127458e+07  733657.663071  815942.223042   \n",
       "1840  2019-01  18.564031   Yala  1.127458e+07  733657.663071  815942.223042   \n",
       "1839  2018-12  14.292233   Yala  1.127458e+07  733657.663071  815942.223042   \n",
       "1838  2018-11  16.203863   Yala  1.127458e+07  733657.663071  815942.223042   \n",
       "1823  2017-08  16.498505   Yala  1.127458e+07  733657.663071  815942.223042   \n",
       "\n",
       "             level  \n",
       "1841  satisfactory  \n",
       "1840  satisfactory  \n",
       "1839  satisfactory  \n",
       "1838  satisfactory  \n",
       "1823  satisfactory  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# find the monthly average \n",
    "monthly_all = pm25_all.groupby(['month','city'],as_index=False).mean()\n",
    "#monthly_all['month'] = monthly_all['month'].values.astype('datetime64[M]')\n",
    "monthly_all['city'] = monthly_all['city'].astype('category')\n",
    "monthly_all = monthly_all.merge(region_df, left_on=['city'], right_on='City')\n",
    "# group by the level \n",
    "monthly_all['level'] = pd.cut(monthly_all['PM2.5'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['satisfactory', 'moderate','unhealthy'],\n",
    "                      include_lowest=True, right=False)\n",
    "monthly_all = monthly_all.drop(['city','City','Population','Latitude','Longitude'],axis=1)\n",
    "#sort by the lattitue\n",
    "monthly_all.sort_values('y_mer',ascending=False,inplace=True)\n",
    "monthly_all.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# build a map center on Bangkok\n",
    "month_min = monthly_all['month'].min().strftime('%Y-%m')\n",
    "month_max = monthly_all['month'].max().strftime('%Y-%m')\n",
    "[[bkk_x, bkk_y]]= region_df.query('City == \"Bangkok\"')[['x_mer','y_mer']].values\n",
    "stepx = 3E5\n",
    "stepy = 3*stepx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "reset_output()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "    <div class=\"bk-root\">\n",
       "        <a href=\"https://bokeh.pydata.org\" target=\"_blank\" class=\"bk-logo bk-logo-small bk-logo-notebook\"></a>\n",
       "        <span id=\"1008\">Loading BokehJS ...</span>\n",
       "    </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "(function(root) {\n",
       "  function now() {\n",
       "    return new Date();\n",
       "  }\n",
       "\n",
       "  var force = true;\n",
       "\n",
       "  if (typeof (root._bokeh_onload_callbacks) === \"undefined\" || force === true) {\n",
       "    root._bokeh_onload_callbacks = [];\n",
       "    root._bokeh_is_loading = undefined;\n",
       "  }\n",
       "\n",
       "  var JS_MIME_TYPE = 'application/javascript';\n",
       "  var HTML_MIME_TYPE = 'text/html';\n",
       "  var EXEC_MIME_TYPE = 'application/vnd.bokehjs_exec.v0+json';\n",
       "  var CLASS_NAME = 'output_bokeh rendered_html';\n",
       "\n",
       "  /**\n",
       "   * Render data to the DOM node\n",
       "   */\n",
       "  function render(props, node) {\n",
       "    var script = document.createElement(\"script\");\n",
       "    node.appendChild(script);\n",
       "  }\n",
       "\n",
       "  /**\n",
       "   * Handle when an output is cleared or removed\n",
       "   */\n",
       "  function handleClearOutput(event, handle) {\n",
       "    var cell = handle.cell;\n",
       "\n",
       "    var id = cell.output_area._bokeh_element_id;\n",
       "    var server_id = cell.output_area._bokeh_server_id;\n",
       "    // Clean up Bokeh references\n",
       "    if (id != null && id in Bokeh.index) {\n",
       "      Bokeh.index[id].model.document.clear();\n",
       "      delete Bokeh.index[id];\n",
       "    }\n",
       "\n",
       "    if (server_id !== undefined) {\n",
       "      // Clean up Bokeh references\n",
       "      var cmd = \"from bokeh.io.state import curstate; print(curstate().uuid_to_server['\" + server_id + \"'].get_sessions()[0].document.roots[0]._id)\";\n",
       "      cell.notebook.kernel.execute(cmd, {\n",
       "        iopub: {\n",
       "          output: function(msg) {\n",
       "            var id = msg.content.text.trim();\n",
       "            if (id in Bokeh.index) {\n",
       "              Bokeh.index[id].model.document.clear();\n",
       "              delete Bokeh.index[id];\n",
       "            }\n",
       "          }\n",
       "        }\n",
       "      });\n",
       "      // Destroy server and session\n",
       "      var cmd = \"import bokeh.io.notebook as ion; ion.destroy_server('\" + server_id + \"')\";\n",
       "      cell.notebook.kernel.execute(cmd);\n",
       "    }\n",
       "  }\n",
       "\n",
       "  /**\n",
       "   * Handle when a new output is added\n",
       "   */\n",
       "  function handleAddOutput(event, handle) {\n",
       "    var output_area = handle.output_area;\n",
       "    var output = handle.output;\n",
       "\n",
       "    // limit handleAddOutput to display_data with EXEC_MIME_TYPE content only\n",
       "    if ((output.output_type != \"display_data\") || (!output.data.hasOwnProperty(EXEC_MIME_TYPE))) {\n",
       "      return\n",
       "    }\n",
       "\n",
       "    var toinsert = output_area.element.find(\".\" + CLASS_NAME.split(' ')[0]);\n",
       "\n",
       "    if (output.metadata[EXEC_MIME_TYPE][\"id\"] !== undefined) {\n",
       "      toinsert[toinsert.length - 1].firstChild.textContent = output.data[JS_MIME_TYPE];\n",
       "      // store reference to embed id on output_area\n",
       "      output_area._bokeh_element_id = output.metadata[EXEC_MIME_TYPE][\"id\"];\n",
       "    }\n",
       "    if (output.metadata[EXEC_MIME_TYPE][\"server_id\"] !== undefined) {\n",
       "      var bk_div = document.createElement(\"div\");\n",
       "      bk_div.innerHTML = output.data[HTML_MIME_TYPE];\n",
       "      var script_attrs = bk_div.children[0].attributes;\n",
       "      for (var i = 0; i < script_attrs.length; i++) {\n",
       "        toinsert[toinsert.length - 1].firstChild.setAttribute(script_attrs[i].name, script_attrs[i].value);\n",
       "      }\n",
       "      // store reference to server id on output_area\n",
       "      output_area._bokeh_server_id = output.metadata[EXEC_MIME_TYPE][\"server_id\"];\n",
       "    }\n",
       "  }\n",
       "\n",
       "  function register_renderer(events, OutputArea) {\n",
       "\n",
       "    function append_mime(data, metadata, element) {\n",
       "      // create a DOM node to render to\n",
       "      var toinsert = this.create_output_subarea(\n",
       "        metadata,\n",
       "        CLASS_NAME,\n",
       "        EXEC_MIME_TYPE\n",
       "      );\n",
       "      this.keyboard_manager.register_events(toinsert);\n",
       "      // Render to node\n",
       "      var props = {data: data, metadata: metadata[EXEC_MIME_TYPE]};\n",
       "      render(props, toinsert[toinsert.length - 1]);\n",
       "      element.append(toinsert);\n",
       "      return toinsert\n",
       "    }\n",
       "\n",
       "    /* Handle when an output is cleared or removed */\n",
       "    events.on('clear_output.CodeCell', handleClearOutput);\n",
       "    events.on('delete.Cell', handleClearOutput);\n",
       "\n",
       "    /* Handle when a new output is added */\n",
       "    events.on('output_added.OutputArea', handleAddOutput);\n",
       "\n",
       "    /**\n",
       "     * Register the mime type and append_mime function with output_area\n",
       "     */\n",
       "    OutputArea.prototype.register_mime_type(EXEC_MIME_TYPE, append_mime, {\n",
       "      /* Is output safe? */\n",
       "      safe: true,\n",
       "      /* Index of renderer in `output_area.display_order` */\n",
       "      index: 0\n",
       "    });\n",
       "  }\n",
       "\n",
       "  // register the mime type if in Jupyter Notebook environment and previously unregistered\n",
       "  if (root.Jupyter !== undefined) {\n",
       "    var events = require('base/js/events');\n",
       "    var OutputArea = require('notebook/js/outputarea').OutputArea;\n",
       "\n",
       "    if (OutputArea.prototype.mime_types().indexOf(EXEC_MIME_TYPE) == -1) {\n",
       "      register_renderer(events, OutputArea);\n",
       "    }\n",
       "  }\n",
       "\n",
       "  \n",
       "  if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n",
       "    root._bokeh_timeout = Date.now() + 5000;\n",
       "    root._bokeh_failed_load = false;\n",
       "  }\n",
       "\n",
       "  var NB_LOAD_WARNING = {'data': {'text/html':\n",
       "     \"<div style='background-color: #fdd'>\\n\"+\n",
       "     \"<p>\\n\"+\n",
       "     \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n",
       "     \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n",
       "     \"</p>\\n\"+\n",
       "     \"<ul>\\n\"+\n",
       "     \"<li>re-rerun `output_notebook()` to attempt to load from CDN again, or</li>\\n\"+\n",
       "     \"<li>use INLINE resources instead, as so:</li>\\n\"+\n",
       "     \"</ul>\\n\"+\n",
       "     \"<code>\\n\"+\n",
       "     \"from bokeh.resources import INLINE\\n\"+\n",
       "     \"output_notebook(resources=INLINE)\\n\"+\n",
       "     \"</code>\\n\"+\n",
       "     \"</div>\"}};\n",
       "\n",
       "  function display_loaded() {\n",
       "    var el = document.getElementById(\"1008\");\n",
       "    if (el != null) {\n",
       "      el.textContent = \"BokehJS is loading...\";\n",
       "    }\n",
       "    if (root.Bokeh !== undefined) {\n",
       "      if (el != null) {\n",
       "        el.textContent = \"BokehJS \" + root.Bokeh.version + \" successfully loaded.\";\n",
       "      }\n",
       "    } else if (Date.now() < root._bokeh_timeout) {\n",
       "      setTimeout(display_loaded, 100)\n",
       "    }\n",
       "  }\n",
       "\n",
       "\n",
       "  function run_callbacks() {\n",
       "    try {\n",
       "      root._bokeh_onload_callbacks.forEach(function(callback) { callback() });\n",
       "    }\n",
       "    finally {\n",
       "      delete root._bokeh_onload_callbacks\n",
       "    }\n",
       "    console.info(\"Bokeh: all callbacks have finished\");\n",
       "  }\n",
       "\n",
       "  function load_libs(js_urls, callback) {\n",
       "    root._bokeh_onload_callbacks.push(callback);\n",
       "    if (root._bokeh_is_loading > 0) {\n",
       "      console.log(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n",
       "      return null;\n",
       "    }\n",
       "    if (js_urls == null || js_urls.length === 0) {\n",
       "      run_callbacks();\n",
       "      return null;\n",
       "    }\n",
       "    console.log(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n",
       "    root._bokeh_is_loading = js_urls.length;\n",
       "    for (var i = 0; i < js_urls.length; i++) {\n",
       "      var url = js_urls[i];\n",
       "      var s = document.createElement('script');\n",
       "      s.src = url;\n",
       "      s.async = false;\n",
       "      s.onreadystatechange = s.onload = function() {\n",
       "        root._bokeh_is_loading--;\n",
       "        if (root._bokeh_is_loading === 0) {\n",
       "          console.log(\"Bokeh: all BokehJS libraries loaded\");\n",
       "          run_callbacks()\n",
       "        }\n",
       "      };\n",
       "      s.onerror = function() {\n",
       "        console.warn(\"failed to load library \" + url);\n",
       "      };\n",
       "      console.log(\"Bokeh: injecting script tag for BokehJS library: \", url);\n",
       "      document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
       "    }\n",
       "  };var element = document.getElementById(\"1008\");\n",
       "  if (element == null) {\n",
       "    console.log(\"Bokeh: ERROR: autoload.js configured with elementid '1008' but no matching script tag was found. \")\n",
       "    return false;\n",
       "  }\n",
       "\n",
       "  var js_urls = [\"https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-gl-1.0.4.min.js\"];\n",
       "\n",
       "  var inline_js = [\n",
       "    function(Bokeh) {\n",
       "      Bokeh.set_log_level(\"info\");\n",
       "    },\n",
       "    \n",
       "    function(Bokeh) {\n",
       "      \n",
       "    },\n",
       "    function(Bokeh) {\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.css\");\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.css\");\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.css\");\n",
       "    }\n",
       "  ];\n",
       "\n",
       "  function run_inline_js() {\n",
       "    \n",
       "    if ((root.Bokeh !== undefined) || (force === true)) {\n",
       "      for (var i = 0; i < inline_js.length; i++) {\n",
       "        inline_js[i].call(root, root.Bokeh);\n",
       "      }if (force === true) {\n",
       "        display_loaded();\n",
       "      }} else if (Date.now() < root._bokeh_timeout) {\n",
       "      setTimeout(run_inline_js, 100);\n",
       "    } else if (!root._bokeh_failed_load) {\n",
       "      console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n",
       "      root._bokeh_failed_load = true;\n",
       "    } else if (force !== true) {\n",
       "      var cell = $(document.getElementById(\"1008\")).parents('.cell').data().cell;\n",
       "      cell.output_area.append_execute_result(NB_LOAD_WARNING)\n",
       "    }\n",
       "\n",
       "  }\n",
       "\n",
       "  if (root._bokeh_is_loading === 0) {\n",
       "    console.log(\"Bokeh: BokehJS loaded, going straight to plotting\");\n",
       "    run_inline_js();\n",
       "  } else {\n",
       "    load_libs(js_urls, function() {\n",
       "      console.log(\"Bokeh: BokehJS plotting callback run at\", now());\n",
       "      run_inline_js();\n",
       "    });\n",
       "  }\n",
       "}(window));"
      ],
      "application/vnd.bokehjs_load.v0+json": "\n(function(root) {\n  function now() {\n    return new Date();\n  }\n\n  var force = true;\n\n  if (typeof (root._bokeh_onload_callbacks) === \"undefined\" || force === true) {\n    root._bokeh_onload_callbacks = [];\n    root._bokeh_is_loading = undefined;\n  }\n\n  \n\n  \n  if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n    root._bokeh_timeout = Date.now() + 5000;\n    root._bokeh_failed_load = false;\n  }\n\n  var NB_LOAD_WARNING = {'data': {'text/html':\n     \"<div style='background-color: #fdd'>\\n\"+\n     \"<p>\\n\"+\n     \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n     \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n     \"</p>\\n\"+\n     \"<ul>\\n\"+\n     \"<li>re-rerun `output_notebook()` to attempt to load from CDN again, or</li>\\n\"+\n     \"<li>use INLINE resources instead, as so:</li>\\n\"+\n     \"</ul>\\n\"+\n     \"<code>\\n\"+\n     \"from bokeh.resources import INLINE\\n\"+\n     \"output_notebook(resources=INLINE)\\n\"+\n     \"</code>\\n\"+\n     \"</div>\"}};\n\n  function display_loaded() {\n    var el = document.getElementById(\"1008\");\n    if (el != null) {\n      el.textContent = \"BokehJS is loading...\";\n    }\n    if (root.Bokeh !== undefined) {\n      if (el != null) {\n        el.textContent = \"BokehJS \" + root.Bokeh.version + \" successfully loaded.\";\n      }\n    } else if (Date.now() < root._bokeh_timeout) {\n      setTimeout(display_loaded, 100)\n    }\n  }\n\n\n  function run_callbacks() {\n    try {\n      root._bokeh_onload_callbacks.forEach(function(callback) { callback() });\n    }\n    finally {\n      delete root._bokeh_onload_callbacks\n    }\n    console.info(\"Bokeh: all callbacks have finished\");\n  }\n\n  function load_libs(js_urls, callback) {\n    root._bokeh_onload_callbacks.push(callback);\n    if (root._bokeh_is_loading > 0) {\n      console.log(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n      return null;\n    }\n    if (js_urls == null || js_urls.length === 0) {\n      run_callbacks();\n      return null;\n    }\n    console.log(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n    root._bokeh_is_loading = js_urls.length;\n    for (var i = 0; i < js_urls.length; i++) {\n      var url = js_urls[i];\n      var s = document.createElement('script');\n      s.src = url;\n      s.async = false;\n      s.onreadystatechange = s.onload = function() {\n        root._bokeh_is_loading--;\n        if (root._bokeh_is_loading === 0) {\n          console.log(\"Bokeh: all BokehJS libraries loaded\");\n          run_callbacks()\n        }\n      };\n      s.onerror = function() {\n        console.warn(\"failed to load library \" + url);\n      };\n      console.log(\"Bokeh: injecting script tag for BokehJS library: \", url);\n      document.getElementsByTagName(\"head\")[0].appendChild(s);\n    }\n  };var element = document.getElementById(\"1008\");\n  if (element == null) {\n    console.log(\"Bokeh: ERROR: autoload.js configured with elementid '1008' but no matching script tag was found. \")\n    return false;\n  }\n\n  var js_urls = [\"https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-gl-1.0.4.min.js\"];\n\n  var inline_js = [\n    function(Bokeh) {\n      Bokeh.set_log_level(\"info\");\n    },\n    \n    function(Bokeh) {\n      \n    },\n    function(Bokeh) {\n      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.css\");\n      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-1.0.4.min.css\");\n      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.css\");\n      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.0.4.min.css\");\n      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.css\");\n      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.0.4.min.css\");\n    }\n  ];\n\n  function run_inline_js() {\n    \n    if ((root.Bokeh !== undefined) || (force === true)) {\n      for (var i = 0; i < inline_js.length; i++) {\n        inline_js[i].call(root, root.Bokeh);\n      }if (force === true) {\n        display_loaded();\n      }} else if (Date.now() < root._bokeh_timeout) {\n      setTimeout(run_inline_js, 100);\n    } else if (!root._bokeh_failed_load) {\n      console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n      root._bokeh_failed_load = true;\n    } else if (force !== true) {\n      var cell = $(document.getElementById(\"1008\")).parents('.cell').data().cell;\n      cell.output_area.append_execute_result(NB_LOAD_WARNING)\n    }\n\n  }\n\n  if (root._bokeh_is_loading === 0) {\n    console.log(\"Bokeh: BokehJS loaded, going straight to plotting\");\n    run_inline_js();\n  } else {\n    load_libs(js_urls, function() {\n      console.log(\"Bokeh: BokehJS plotting callback run at\", now());\n      run_inline_js();\n    });\n  }\n}(window));"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "output_notebook()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# build the interactive map\n",
    "def plot_monthly(doc):\n",
    "    \n",
    "# create figure \n",
    "    p = figure(x_range=(bkk_x-0.3*stepx, bkk_x+stepx), y_range=(bkk_y-stepy, bkk_y+stepy),\n",
    "               x_axis_type=\"mercator\", y_axis_type=\"mercator\",\n",
    "               title='PM 2.5 Pollution', plot_width=450, plot_height=600)\n",
    "    # add map\n",
    "    p.add_tile(STAMEN_TERRAIN)\n",
    "\n",
    "    source = ColumnDataSource(monthly_all[monthly_all['month'] == '2019-03'])\n",
    "    # make colorbar\n",
    "    color_mapper = CategoricalColorMapper(factors=['satisfactory', 'moderate', 'unhealthy'],\n",
    "                                          palette=['green', 'orange', 'red'])\n",
    "\n",
    "    #plot the pollution reading\n",
    "    r = p.circle(source=source, x='x_mer', y='y_mer', size=15, alpha=0.7,\n",
    "                 color=dict(field='level', transform=color_mapper))\n",
    "    \n",
    "    # create legend\n",
    "    legend = Legend(items=[LegendItem(label=dict(field='level'), renderers=[r])], location=\"bottom_right\")\n",
    "\n",
    "    \n",
    "    p.add_layout(legend)\n",
    "\n",
    "    #put monthly average bar graph on the side\n",
    "    p2 = figure(y_range=monthly_all['Region'].iloc[::-1].unique(), x_range=(0, 70),\n",
    "                plot_width=200, plot_height=600, toolbar_location=None)\n",
    "\n",
    "    p2.xaxis.major_label_orientation = math.pi/2\n",
    "    p2.xaxis.ticker = [0, 20, 40, 60]\n",
    "\n",
    "    p2.hbar(y='Region', right='PM2.5', height=0.5, source=source,\n",
    "            color=dict(field='level', transform=color_mapper))\n",
    "\n",
    "    # add slider bar for selecting month\n",
    "    def callback(attr, old, new):\n",
    "        data = monthly_all[monthly_all['month'] == new]\n",
    "        source.data = ColumnDataSource(data=data).data\n",
    "\n",
    "    slider = DateSlider(title='Month of Year', start=month_min, end=month_max, step=1,\n",
    "                        value=month_min, format='%B %Y')\n",
    "    slider.on_change('value', callback)\n",
    "\n",
    "    layout = column(widgetbox(slider), row(p2, p))\n",
    "    doc.add_root(layout)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.bokehjs_exec.v0+json": "",
      "text/html": [
       "\n",
       "<script src=\"http://localhost:52726/autoload.js?bokeh-autoload-element=2079&bokeh-absolute-url=http://localhost:52726&resources=none\" id=\"2079\"></script>"
      ]
     },
     "metadata": {
      "application/vnd.bokehjs_exec.v0+json": {
       "server_id": "9b850b43722c4a0d8bd54880bde01146"
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "show(plot_monthly)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_monthly(monthly_all, month):\n",
    "    '''function to create pollution map for each month and\n",
    "    save it as png file\n",
    "    '''\n",
    "\n",
    "    # create figure\n",
    "    p = figure(x_range=(bkk_x-0.3*stepx, bkk_x+stepx), y_range=(bkk_y-stepy, bkk_y+stepy),\n",
    "               x_axis_type=\"mercator\", y_axis_type=\"mercator\",\n",
    "               title='PM 2.5 Pollution in '+month.strftime('%B %Y'), plot_width=450, plot_height=600)\n",
    "    # map\n",
    "    p.add_tile(STAMEN_TERRAIN)\n",
    "\n",
    "    source = ColumnDataSource(monthly_all[monthly_all['month'] == month])\n",
    "    color_mapper = CategoricalColorMapper(factors=['satisfactory', 'moderate', 'unhealthy'],\n",
    "                                          palette=['green', 'orange', 'red'])\n",
    "\n",
    "    #add the pollution data colored by level\n",
    "    r = p.circle(source=source, x='x_mer', y='y_mer', size=15, alpha=0.7,\n",
    "                 color=dict(field='level', transform=color_mapper))\n",
    "    \n",
    "    legend = Legend(items=[LegendItem(label=dict(field='level'), renderers=[r])], location=\"bottom_right\")\n",
    "\n",
    "    \n",
    "    p.add_layout(legend)\n",
    "\n",
    "    # add monthly average as a side bar, color coded by level\n",
    "    p2 = figure(y_range=monthly_all['Region'].iloc[::-1].unique(), x_range=(0, 70),\n",
    "                plot_width=200, plot_height=600, toolbar_location=None)\n",
    "\n",
    "    p2.xaxis.major_label_orientation = math.pi/2\n",
    "    p2.xaxis.ticker = [0, 20, 40, 60]\n",
    "\n",
    "    p2.hbar(y='Region', right='PM2.5', height=0.5, source=source,\n",
    "            color=dict(field='level', transform=color_mapper))\n",
    "\n",
    "    layout = row(p2, p)\n",
    "    month = month.strftime('%Y-%m')\n",
    "    # save the figure for each month for building gif\n",
    "    export_png(layout,filename='monthly_map/'+month+'monthly_map.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# call the function to create the map and save it\n",
    "for month in monthly_all['month'].unique():\n",
    "    show_monthly(monthly_all, month)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from glob import glob\n",
    "import imageio"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load all the pollution map fire \n",
    "filenames = glob('monthly_map/*.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# create a gif showing pollution level for each month\n",
    "images = []\n",
    "for filename in filenames:\n",
    "    images.append(imageio.imread(filename))\n",
    "imageio.mimsave('monthly_map/monthly_map.gif', images, duration=0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Finding yearly average for each province\n",
    "\n",
    "Show the data for each province in 2017 and 2018 as bar plot, rank by the pollution level."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "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>index</th>\n",
       "      <th>year</th>\n",
       "      <th>PM2.5</th>\n",
       "      <th>Time</th>\n",
       "      <th>month</th>\n",
       "      <th>Region</th>\n",
       "      <th>x_mer</th>\n",
       "      <th>y_mer</th>\n",
       "      <th>level</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>106</th>\n",
       "      <td>45</td>\n",
       "      <td>2017</td>\n",
       "      <td>59.2</td>\n",
       "      <td>2017-12-31 23:00:00</td>\n",
       "      <td>2017-12</td>\n",
       "      <td>Chumphon</td>\n",
       "      <td>1.104063e+07</td>\n",
       "      <td>1.176187e+06</td>\n",
       "      <td>unhealthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>6</td>\n",
       "      <td>2018</td>\n",
       "      <td>59.2</td>\n",
       "      <td>2018-12-31 23:00:00</td>\n",
       "      <td>2018-12</td>\n",
       "      <td>Songkhla</td>\n",
       "      <td>1.119263e+07</td>\n",
       "      <td>8.076416e+05</td>\n",
       "      <td>unhealthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108</th>\n",
       "      <td>118</td>\n",
       "      <td>2018</td>\n",
       "      <td>58.3</td>\n",
       "      <td>2018-12-31 23:00:00</td>\n",
       "      <td>2018-12</td>\n",
       "      <td>Nakhon Si Thammarat</td>\n",
       "      <td>1.112824e+07</td>\n",
       "      <td>9.448815e+05</td>\n",
       "      <td>unhealthy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>205</td>\n",
       "      <td>2017</td>\n",
       "      <td>52.8</td>\n",
       "      <td>2017-12-31 23:00:00</td>\n",
       "      <td>2017-12</td>\n",
       "      <td>Trang</td>\n",
       "      <td>1.108869e+07</td>\n",
       "      <td>8.469228e+05</td>\n",
       "      <td>moderate</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>206</td>\n",
       "      <td>2018</td>\n",
       "      <td>44.8</td>\n",
       "      <td>2018-12-31 23:00:00</td>\n",
       "      <td>2018-12</td>\n",
       "      <td>Trang</td>\n",
       "      <td>1.108869e+07</td>\n",
       "      <td>8.469228e+05</td>\n",
       "      <td>moderate</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     index  year  PM2.5                Time    month               Region  \\\n",
       "106     45  2017   59.2 2017-12-31 23:00:00  2017-12             Chumphon   \n",
       "107      6  2018   59.2 2018-12-31 23:00:00  2018-12             Songkhla   \n",
       "108    118  2018   58.3 2018-12-31 23:00:00  2018-12  Nakhon Si Thammarat   \n",
       "109    205  2017   52.8 2017-12-31 23:00:00  2017-12                Trang   \n",
       "110    206  2018   44.8 2018-12-31 23:00:00  2018-12                Trang   \n",
       "\n",
       "            x_mer         y_mer      level  \n",
       "106  1.104063e+07  1.176187e+06  unhealthy  \n",
       "107  1.119263e+07  8.076416e+05  unhealthy  \n",
       "108  1.112824e+07  9.448815e+05  unhealthy  \n",
       "109  1.108869e+07  8.469228e+05   moderate  \n",
       "110  1.108869e+07  8.469228e+05   moderate  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create yearly average group by city\n",
    "yearly_all = pm25_all.groupby(['year','city'], as_index=False).max()\n",
    "yearly_all['year'] = yearly_all['year'].values.astype(str)\n",
    "yearly_all['city'] = yearly_all['city'].astype('category')\n",
    "yearly_all = yearly_all.merge(region_df,left_on='city',right_on='City')\n",
    "\n",
    "# separate by level\n",
    "yearly_all['level'] = pd.cut(yearly_all['PM2.5'],bins = [0, 35.5, 55.5, 150.4], \n",
    "                       labels=['satisfactory', 'moderate','unhealthy'],\n",
    "                      include_lowest=True, right=False)\n",
    "yearly_all = yearly_all.drop(['city','Population','Latitude','Longitude','City'],axis=1)\n",
    "\n",
    "# drop the value in 2016 and 2019 due to incomplete data \n",
    "yearly_all = yearly_all.drop(yearly_all.loc[yearly_all['year'].isin(['2016','2019'])].index)\n",
    "yearly_all.sort_values('PM2.5',ascending=False,inplace=True)\n",
    "yearly_all.reset_index(inplace=True)\n",
    "yearly_all.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "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>year</th>\n",
       "      <th>Region</th>\n",
       "      <th>PM2.5</th>\n",
       "      <th>month</th>\n",
       "      <th>x_mer</th>\n",
       "      <th>y_mer</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>2017</td>\n",
       "      <td>Nakhon Si Thammarat</td>\n",
       "      <td>22.758319</td>\n",
       "      <td>2017-12</td>\n",
       "      <td>1.112824e+07</td>\n",
       "      <td>944881.495734</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77</th>\n",
       "      <td>2017</td>\n",
       "      <td>Trang</td>\n",
       "      <td>22.458169</td>\n",
       "      <td>2017-12</td>\n",
       "      <td>1.108869e+07</td>\n",
       "      <td>846922.840036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>118</th>\n",
       "      <td>2018</td>\n",
       "      <td>Trang</td>\n",
       "      <td>21.950633</td>\n",
       "      <td>2018-12</td>\n",
       "      <td>1.108869e+07</td>\n",
       "      <td>846922.840036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>2017</td>\n",
       "      <td>Phuket</td>\n",
       "      <td>21.274390</td>\n",
       "      <td>2017-12</td>\n",
       "      <td>1.095385e+07</td>\n",
       "      <td>884237.988691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>115</th>\n",
       "      <td>2018</td>\n",
       "      <td>Songkhla</td>\n",
       "      <td>21.247627</td>\n",
       "      <td>2018-12</td>\n",
       "      <td>1.119820e+07</td>\n",
       "      <td>807641.616273</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year               Region      PM2.5    month         x_mer  \\\n",
       "59   2017  Nakhon Si Thammarat  22.758319  2017-12  1.112824e+07   \n",
       "77   2017                Trang  22.458169  2017-12  1.108869e+07   \n",
       "118  2018                Trang  21.950633  2018-12  1.108869e+07   \n",
       "66   2017               Phuket  21.274390  2017-12  1.095385e+07   \n",
       "115  2018             Songkhla  21.247627  2018-12  1.119820e+07   \n",
       "\n",
       "             y_mer  \n",
       "59   944881.495734  \n",
       "77   846922.840036  \n",
       "118  846922.840036  \n",
       "66   884237.988691  \n",
       "115  807641.616273  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "monthly_all['year'] = monthly_all['month'].dt.strftime('%Y')\n",
    "yearly_all = monthly_all.groupby(['year','Region'], as_index=False).max()\n",
    "# sort data by pollution level\n",
    "yearly_all.sort_values('PM2.5',ascending=False,inplace=True)\n",
    "\n",
    "# drop the value in 2016 and 2019 due to incomplete data \n",
    "yearly_all = yearly_all.drop(yearly_all.loc[yearly_all['year'].isin(['2016','2019'])].index)\n",
    "\n",
    "yearly_all.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# this province has incomplete data \n",
    "yearly_all = yearly_all[yearly_all['Region']!=' Ubon Ratchathani']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Benny\\Anaconda3\\envs\\fastai\\lib\\site-packages\\matplotlib\\figure.py:98: MatplotlibDeprecationWarning: \n",
      "Adding an axes using the same arguments as a previous axes currently reuses the earlier instance.  In a future version, a new instance will always be created and returned.  Meanwhile, this warning can be suppressed, and the future behavior ensured, by passing a unique label to each axes instance.\n",
      "  \"Adding an axes using the same arguments as a previous axes \"\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f = plt.figure(figs ize=(10,3))\n",
    "\n",
    "# create the pollution level line for comparision\n",
    "plt.axhline(55, color='red', alpha=0.5, linestyle='--')\n",
    "plt.axhline(35, color='orange', alpha=0.5,linestyle='--')\n",
    "\n",
    "g = sns.barplot(x=\"Region\", y=\"PM2.5\", hue='year', data=yearly_all)\n",
    "plt.legend(loc= 'upper right')\n",
    "plt.xticks(rotation=90, fontsize=9)\n",
    "ax1 = plt.axes()\n",
    "ax1.xaxis.label.set_visible(False)\n",
    "\n",
    "# addnotation \n",
    "plt.annotate('southern provices\\n has the best air quality',xycoords='axes fraction',\n",
    "              textcoords='axes fraction',\n",
    "             xy = (0.95,0.35), xytext=(0.8,0.5), arrowprops=dict(arrowstyle='->', lw = 2))\n",
    "\n",
    "plt.annotate('northern provices\\n are worse than Bangkok',xycoords='axes fraction',\n",
    "              textcoords='axes fraction',\n",
    "             xy = (0.1,0.8), xytext=(0.15,0.85), arrowprops=dict(arrowstyle='->', lw = 2))\n",
    "\n",
    "plt.annotate('Bangkok',xycoords='axes fraction',xy = (0.72,0.6),\n",
    "            textcoords='axes fraction', xytext=(0.68,0.85),\n",
    "            arrowprops=dict(arrowstyle='->', lw = 2))\n",
    "plt.title('Maximum Particle Pollution')\n",
    "plt.show(g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 309,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "\n",
       "\n",
       "\n",
       "\n",
       "\n",
       "  <div class=\"bk-root\" id=\"d1a6b1d9-5ade-4406-9648-dd876fcc0772\" data-root-id=\"69356\"></div>\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "(function(root) {\n",
       "  function embed_document(root) {\n",
       "    \n",
       "  var docs_json = {\"fb5a668f-38bd-4431-b00d-a7b662c35bef\":{\"roots\":{\"references\":[{\"attributes\":{\"below\":[{\"id\":\"69365\",\"type\":\"LinearAxis\"}],\"left\":[{\"id\":\"69370\",\"type\":\"CategoricalAxis\"}],\"plot_width\":200,\"renderers\":[{\"id\":\"69365\",\"type\":\"LinearAxis\"},{\"id\":\"69369\",\"type\":\"Grid\"},{\"id\":\"69370\",\"type\":\"CategoricalAxis\"},{\"id\":\"69373\",\"type\":\"Grid\"},{\"id\":\"69382\",\"type\":\"BoxAnnotation\"},{\"id\":\"69393\",\"type\":\"GlyphRenderer\"}],\"title\":{\"id\":\"69602\",\"type\":\"Title\"},\"toolbar\":{\"id\":\"69380\",\"type\":\"Toolbar\"},\"toolbar_location\":null,\"x_range\":{\"id\":\"69357\",\"type\":\"DataRange1d\"},\"x_scale\":{\"id\":\"69361\",\"type\":\"LinearScale\"},\"y_range\":{\"id\":\"69359\",\"type\":\"FactorRange\"},\"y_scale\":{\"id\":\"69363\",\"type\":\"CategoricalScale\"}},\"id\":\"69356\",\"subtype\":\"Figure\",\"type\":\"Plot\"},{\"attributes\":{},\"id\":\"69363\",\"type\":\"CategoricalScale\"},{\"attributes\":{},\"id\":\"69609\",\"type\":\"Selection\"},{\"attributes\":{\"overlay\":{\"id\":\"69382\",\"type\":\"BoxAnnotation\"}},\"id\":\"69376\",\"type\":\"BoxZoomTool\"},{\"attributes\":{},\"id\":\"69374\",\"type\":\"PanTool\"},{\"attributes\":{\"plot\":{\"id\":\"69356\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"69366\",\"type\":\"BasicTicker\"}},\"id\":\"69369\",\"type\":\"Grid\"},{\"attributes\":{},\"id\":\"69378\",\"type\":\"ResetTool\"},{\"attributes\":{\"active_drag\":\"auto\",\"active_inspect\":\"auto\",\"active_multi\":null,\"active_scroll\":\"auto\",\"active_tap\":\"auto\",\"tools\":[{\"id\":\"69374\",\"type\":\"PanTool\"},{\"id\":\"69375\",\"type\":\"WheelZoomTool\"},{\"id\":\"69376\",\"type\":\"BoxZoomTool\"},{\"id\":\"69377\",\"type\":\"SaveTool\"},{\"id\":\"69378\",\"type\":\"ResetTool\"},{\"id\":\"69379\",\"type\":\"HelpTool\"}]},\"id\":\"69380\",\"type\":\"Toolbar\"},{\"attributes\":{},\"id\":\"69375\",\"type\":\"WheelZoomTool\"},{\"attributes\":{\"plot\":null,\"text\":\"\"},\"id\":\"69602\",\"type\":\"Title\"},{\"attributes\":{\"source\":{\"id\":\"69389\",\"type\":\"ColumnDataSource\"}},\"id\":\"69394\",\"type\":\"CDSView\"},{\"attributes\":{\"callback\":null},\"id\":\"69357\",\"type\":\"DataRange1d\"},{\"attributes\":{\"dimension\":1,\"plot\":{\"id\":\"69356\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"69371\",\"type\":\"CategoricalTicker\"}},\"id\":\"69373\",\"type\":\"Grid\"},{\"attributes\":{\"data_source\":{\"id\":\"69389\",\"type\":\"ColumnDataSource\"},\"glyph\":{\"id\":\"69391\",\"type\":\"HBar\"},\"hover_glyph\":null,\"muted_glyph\":null,\"nonselection_glyph\":{\"id\":\"69392\",\"type\":\"HBar\"},\"selection_glyph\":null,\"view\":{\"id\":\"69394\",\"type\":\"CDSView\"}},\"id\":\"69393\",\"type\":\"GlyphRenderer\"},{\"attributes\":{\"bottom_units\":\"screen\",\"fill_alpha\":{\"value\":0.5},\"fill_color\":{\"value\":\"lightgrey\"},\"left_units\":\"screen\",\"level\":\"overlay\",\"line_alpha\":{\"value\":1.0},\"line_color\":{\"value\":\"black\"},\"line_dash\":[4,4],\"line_width\":{\"value\":2},\"plot\":null,\"render_mode\":\"css\",\"right_units\":\"screen\",\"top_units\":\"screen\"},\"id\":\"69382\",\"type\":\"BoxAnnotation\"},{\"attributes\":{\"fill_alpha\":{\"value\":0.1},\"fill_color\":{\"value\":\"#1f77b4\"},\"height\":{\"value\":0.5},\"line_alpha\":{\"value\":0.1},\"line_color\":{\"value\":\"#1f77b4\"},\"right\":{\"field\":\"PM2.5\"},\"y\":{\"field\":\"City\"}},\"id\":\"69392\",\"type\":\"HBar\"},{\"attributes\":{\"formatter\":{\"id\":\"69604\",\"type\":\"BasicTickFormatter\"},\"plot\":{\"id\":\"69356\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"69366\",\"type\":\"BasicTicker\"}},\"id\":\"69365\",\"type\":\"LinearAxis\"},{\"attributes\":{},\"id\":\"69604\",\"type\":\"BasicTickFormatter\"},{\"attributes\":{},\"id\":\"69377\",\"type\":\"SaveTool\"},{\"attributes\":{},\"id\":\"69608\",\"type\":\"UnionRenderers\"},{\"attributes\":{\"callback\":null,\"data\":{\"City\":[\"Chiang Rai\",\"Chiang Mai\",\"Lampang\",\"Nong Khai\",\"Uttaradit\",\"Udon Thani\",\"Sakon Nakhon\",\"Phitsanulok\",\"Kamphaeng Phet\",\"Khon Kaen\",\"Kalasin\",\"Phetchabun\",\"Maha Sarakham\",\"Chaiyaphum\",\"Nakhon Sawan\",\"Ubon Ratchathani\",\"Nakhon Ratchasima\",\"Lop Buri\",\"Phra Phutthabat\",\"Saraburi\",\"Suphan Buri\",\"Phra Nakhon Si Ayutthaya\",\"Khlong Luang\",\"Kanchanaburi\",\"Ban Lam Luk Ka\",\"Pak Kret\",\"Tha Maka\",\"Mueang Nonthaburi\",\"Nakhon Pathom\",\"Ban Pong\",\"Bang Kruai\",\"Bangkok\",\"Phra Pradaeng\",\"Krathum Baen\",\"Samut Prakan\",\"Samut Sakhon\",\"Ratchaburi\",\"Chon Buri\",\"Si Racha\",\"Phatthaya\",\"Klaeng\",\"Rayong\",\"Sattahip\",\"Chanthaburi\",\"Hua Hin\",\"Chumphon\",\"Ko Samui\",\"Surat Thani\",\"Nakhon Si Thammarat\",\"Phuket\",\"Ban Talat Yai\",\"Trang\",\"Ban Mai\",\"Songkhla\",\"Hat Yai\",\"Yala\"],\"PM2.5\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[56]},\"index\":[369,332,724,1008,1805,1768,1406,1156,508,613,471,1119,792,258,940,1968,903,761,1230,1517,1628,1193,582,545,36,1045,1702,829,866,110,184,221,1267,687,1443,1480,1341,1870,1954,1082,650,1378,1554,295,1898,406,1926,1665,977,1304,147,1739,73,1591,443,1842],\"level\":[\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"unhealthy\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"moderate\",\"satisfactory\",\"satisfactory\",\"moderate\",\"moderate\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"moderate\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"moderate\",\"moderate\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\",\"satisfactory\"],\"month\":[1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0,1551398400000.0],\"x_mer\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[56]},\"y_mer\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[56]}},\"selected\":{\"id\":\"69609\",\"type\":\"Selection\"},\"selection_policy\":{\"id\":\"69608\",\"type\":\"UnionRenderers\"}},\"id\":\"69389\",\"type\":\"ColumnDataSource\"},{\"attributes\":{\"fill_color\":{\"value\":\"#1f77b4\"},\"height\":{\"value\":0.5},\"line_color\":{\"value\":\"#1f77b4\"},\"right\":{\"field\":\"PM2.5\"},\"y\":{\"field\":\"City\"}},\"id\":\"69391\",\"type\":\"HBar\"},{\"attributes\":{},\"id\":\"69366\",\"type\":\"BasicTicker\"},{\"attributes\":{\"callback\":null,\"factors\":[\"Chiang Rai\",\"Chiang Mai\",\"Lampang\",\"Nong Khai\",\"Uttaradit\",\"Udon Thani\",\"Sakon Nakhon\",\"Phitsanulok\",\"Kamphaeng Phet\",\"Khon Kaen\",\"Kalasin\",\"Phetchabun\",\"Maha Sarakham\",\"Chaiyaphum\",\"Nakhon Sawan\",\"Ubon Ratchathani\",\"Nakhon Ratchasima\",\"Lop Buri\",\"Phra Phutthabat\",\"Saraburi\",\"Suphan Buri\",\"Phra Nakhon Si Ayutthaya\",\"Khlong Luang\",\"Kanchanaburi\",\"Ban Lam Luk Ka\",\"Pak Kret\",\"Tha Maka\",\"Mueang Nonthaburi\",\"Nakhon Pathom\",\"Ban Pong\",\"Bang Kruai\",\"Bangkok\",\"Phra Pradaeng\",\"Krathum Baen\",\"Samut Prakan\",\"Samut Sakhon\",\"Ratchaburi\",\"Chon Buri\",\"Si Racha\",\"Phatthaya\",\"Klaeng\",\"Rayong\",\"Sattahip\",\"Chanthaburi\",\"Hua Hin\",\"Chumphon\",\"Ko Samui\",\"Surat Thani\",\"Nakhon Si Thammarat\",\"Phuket\",\"Ban Talat Yai\",\"Trang\",\"Ban Mai\",\"Songkhla\",\"Hat Yai\",\"Yala\"]},\"id\":\"69359\",\"type\":\"FactorRange\"},{\"attributes\":{\"formatter\":{\"id\":\"69606\",\"type\":\"CategoricalTickFormatter\"},\"plot\":{\"id\":\"69356\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"69371\",\"type\":\"CategoricalTicker\"}},\"id\":\"69370\",\"type\":\"CategoricalAxis\"},{\"attributes\":{},\"id\":\"69371\",\"type\":\"CategoricalTicker\"},{\"attributes\":{},\"id\":\"69361\",\"type\":\"LinearScale\"},{\"attributes\":{},\"id\":\"69379\",\"type\":\"HelpTool\"},{\"attributes\":{},\"id\":\"69606\",\"type\":\"CategoricalTickFormatter\"}],\"root_ids\":[\"69356\"]},\"title\":\"Bokeh Application\",\"version\":\"1.0.4\"}};\n",
       "  var render_items = [{\"docid\":\"fb5a668f-38bd-4431-b00d-a7b662c35bef\",\"roots\":{\"69356\":\"d1a6b1d9-5ade-4406-9648-dd876fcc0772\"}}];\n",
       "  root.Bokeh.embed.embed_items_notebook(docs_json, render_items);\n",
       "\n",
       "  }\n",
       "  if (root.Bokeh !== undefined) {\n",
       "    embed_document(root);\n",
       "  } else {\n",
       "    var attempts = 0;\n",
       "    var timer = setInterval(function(root) {\n",
       "      if (root.Bokeh !== undefined) {\n",
       "        embed_document(root);\n",
       "        clearInterval(timer);\n",
       "      }\n",
       "      attempts++;\n",
       "      if (attempts > 100) {\n",
       "        console.log(\"Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing\");\n",
       "        clearInterval(timer);\n",
       "      }\n",
       "    }, 10, root)\n",
       "  }\n",
       "})(window);"
      ],
      "application/vnd.bokehjs_exec.v0+json": ""
     },
     "metadata": {
      "application/vnd.bokehjs_exec.v0+json": {
       "id": "69356"
      }
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "p2 = figure(y_range=monthly_all['City'].unique(), \n",
    "                plot_width=200, plot_height=600,toolbar_location=None)\n",
    "source = ColumnDataSource(monthly_all[monthly_all['month'] == '2019-03'])\n",
    "    \n",
    "p2.hbar(y='City',right= 'PM2.5', height=0.5, source=source)\n",
    "\n",
    "\n",
    "show(p2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}