QAR-Validação de Dados - Est+Combine (d04).ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "   # Validação de Dados Modelados com Observados"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Carregar arquivos para comparar dados monitorados com séries modelada (arquivos combine)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import xarray as xr\n",
    "from netCDF4 import Dataset\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "#from mpl_toolkits.basemap import Basemap\n",
    "import seaborn as sns\n",
    "import datetime as dt\n",
    "from bisect import bisect_left\n",
    "import math\n",
    "import utm\n",
    "%matplotlib inline\n",
    "\n",
    "def takeClosest(myList, myNumber):\n",
    "    \"\"\"\n",
    "    Assumes myList is sorted. Returns closest value to myNumber.\n",
    "    If two numbers are equally close, return the smallest number.\n",
    "    \"\"\"\n",
    "    pos = bisect_left(myList, myNumber)\n",
    "    if pos == 0:\n",
    "        return myList[0]\n",
    "    if pos == len(myList):\n",
    "        return myList[-1]\n",
    "    before = myList[pos - 1]\n",
    "    after = myList[pos]\n",
    "    if after - myNumber < myNumber - before:\n",
    "       return after\n",
    "    else:\n",
    "       return before\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## *Entradas*\n",
    "* Selecionar arquivos de origem para preparar figuras.\n",
    "* Inicializar variáveis para figuras.\n",
    "\n",
    "### Parametrização dos arquivos de entrada\n",
    "\n",
    "Ajustar diretório que contem o arquivo da estação e nomes de rodadas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "directory = str('E:/Mestrado/1.Resultados/QualidadedoAr/')\n",
    "wrfdir = str('E:/Mestrado/1.Resultados/Meteorologia/')\n",
    "mes   = ['jan','jul']\n",
    "grade = 'd04_1km'\n",
    "data  = ['20150109','20150724']\n",
    "datawrf  = ['01-09','07-24']\n",
    "periodo = ['Verão', 'Inverno']\n",
    "\n",
    "modelagens = ['-','_PT_ED-100I'] # Carregando dois cenários base (ED e PT, respectivamente)\n",
    "modnames = ['ED', 'PT_ED-100I']\n",
    "#Demais cenários:\n",
    "#ED = '-20I',           '-20T',            '-20I-20T'\n",
    "#PT = '_PT-20_ED-100I', '_PT_ED-100I-20T', '_PT-20_ED-100I-20T'\n",
    "shapes = ['pirashape','RMSP']\n",
    "\n",
    "spinup = 48\n",
    "run_len = 24*7\n",
    "gmt = 3\n",
    "\n",
    "date_index = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "#DADOS ESTACAO\n",
    "estacao     = 'PIRACICABA (A)'\n",
    "estlat,estlon = utm.to_latlon(easting=227797, northing=7487124, zone_letter='k', zone_number=23)\n",
    "est = pd.read_csv(directory+'piracicaba.csv', delimiter=';')\n",
    "#est['Hora'] = est['Hora'].apply(lambda x: '24:00' if x == '24:00:00' else x)\n",
    "est['Hora'] = est['Hora'].apply(lambda x: dt.timedelta(hours=float(x[:2])))\n",
    "est['Data'] = est['Data'].apply(lambda x: dt.datetime.strptime(x,'%d/%m/%Y'))\n",
    "est['Data'] += est['Hora']\n",
    "est.drop(columns=['Unnamed: 0','Hora','wind_dir','wind_dir_avg','temp','ur','wind_vel'], inplace=True)\n",
    "est.set_index('Data', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Abrir arquivos COMBINE ACONC e DEP"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "aconc = []\n",
    "#dep=[]\n",
    "for m in modelagens:\n",
    "    aconc.append(xr.open_dataset(directory+mes[date_index]+'.'+grade+m+'/COMBINE_ACONC_v521_intel_pira_riz_nudge_'+\n",
    "                        mes[date_index]+'_'+data[date_index]+'.nc'))\n",
    "for i in range(0,len(aconc)):\n",
    "    aconc[i] = aconc[i].isel(LAY=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Abrir arquivo WRF para buscar valores de LAT / LON"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "nc_wrf = Dataset(wrfdir+'arqWRF/'+mes[0]+'/wrf_fnl_pira_pedruzzi2016_nudge_'+\n",
    "                 mes[0]+'/wrfout_'+grade[:3]+'_2015-'+datawrf[0]+'_00%3A00%3A00')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preparar coordenadas para XArray\n",
    "* DATAS\n",
    "* ALTITUDE\n",
    "* LATITUDE\n",
    "* LONGITUDE"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "D:\\Users\\50033096\\AppData\\Local\\Continuum\\anaconda3\\lib\\site-packages\\ipykernel_launcher.py:36: FutureWarning: The inplace argument has been deprecated and will be removed in a future version of xarray.\n"
     ]
    }
   ],
   "source": [
    "# DATAS\n",
    "dates = []\n",
    "dates = [dt.datetime.strptime(str(aconc[0].attrs['SDATE']),'%Y%j')]\n",
    "for i in range(1,aconc[0].dims['TSTEP']):\n",
    "    dates.append(dates[0]+dt.timedelta(hours=i))\n",
    "    \n",
    "# LATITUDE\n",
    "# LONGITUDE\n",
    "croplat = int((nc_wrf.dimensions['south_north'].size-aconc[0].dims['ROW'])/2)\n",
    "croplon = int((nc_wrf.dimensions['south_north'].size-aconc[0].dims['ROW'])/2)\n",
    "lats      = nc_wrf.variables['XLAT'][0,croplat:-croplat,croplat:-croplat]\n",
    "longs     = nc_wrf.variables['XLONG'][0,croplon:-croplon,croplon:-croplon]\n",
    "\n",
    "# COORDENADAS PARA POSICIONAR ESTACAO\n",
    "if (estlat > np.amin(lats) and\n",
    "    estlat < np.amax(lats) and\n",
    "    estlon > np.amin(longs)and\n",
    "    estlon < np.amax(longs)):                \n",
    "    esty     = lats[:,0].tolist().index(takeClosest(lats[:,0],estlat))     #index Lat da estação no domínio\n",
    "    estx     = longs[0,:].tolist().index(takeClosest(longs[0,:],estlon))   #index Lat da estação no domínio\n",
    "\n",
    "# PLOT CORNERS\n",
    "llcrnrlon = np.min(longs)      #longitude of lower left hand corner of the selected map domain.\n",
    "llcrnrlat = np.min(lats)       #latitude of upper right hand corner of the desired map domain (degrees).\n",
    "urcrnrlon = np.max(longs)      #longitude of upper right hand corner of the selected map domain.\n",
    "urcrnrlat = np.max(lats)      #latitude of upper right hand corner of the selected map domain.\n",
    "\n",
    "coordlat  = np.linspace(llcrnrlat,urcrnrlat,num=aconc[0].dims['ROW'])\n",
    "coordlon  = np.linspace(llcrnrlon,urcrnrlon,num=aconc[0].dims['COL'])\n",
    "\n",
    "# ADJUSTMENTS\n",
    "for a in aconc:\n",
    "    a.coords['TSTEP'] = dates\n",
    "    a.coords['ROW'] = coordlat\n",
    "    a.coords['COL'] = coordlon\n",
    "    a.rename({'ROW': 'LAT', 'COL': 'LON'}, inplace=True)\n",
    "    for var in a.variables:\n",
    "        if 'units' in a.variables[var].attrs:\n",
    "            if a.variables[var].attrs['units'].split() != []:\n",
    "                a.variables[var].attrs['units'] = a.variables[var].attrs['units'].split()[0]\n",
    "\n",
    "for i in range(0,len(aconc)):\n",
    "    aconc[i] = aconc[i].isel(TSTEP=slice(spinup,spinup+run_len))\n",
    "    aconc[i].coords['TSTEP'] -= pd.Timedelta(gmt,'h')\n",
    "weekd=[]\n",
    "for d in aconc[0].TSTEP.dt.weekday:\n",
    "    if d < 5:\n",
    "        weekd.append('Weekday')\n",
    "    else:\n",
    "        weekd.append('Weekend')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cria DataFrames para Validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "pol = ['MP2_5','MP10','O3','NOx']\n",
    "val_pm25  = pd.DataFrame(est[(aconc[0]['TSTEP'][0]):(aconc[0]['TSTEP'][-1])][pol[0]])\n",
    "val_pm10  = pd.DataFrame(est[(aconc[0]['TSTEP'][0]):(aconc[0]['TSTEP'][-1])][pol[1]])\n",
    "val_o3    = pd.DataFrame(est[(aconc[0]['TSTEP'][0]):(aconc[0]['TSTEP'][-1])][pol[2]])\n",
    "val_nox   = pd.DataFrame(est[(aconc[0]['TSTEP'][0]):(aconc[0]['TSTEP'][-1])][pol[3]])\n",
    "\n",
    "val = [val_pm25, val_pm10, val_o3, val_nox]\n",
    "# MODELAGENS \n",
    "# Buscar dados das modelagens, usando mod_param\n",
    "for i in range(0,len(aconc)):\n",
    "# pm25 = PM25_TOT        ,ug/m3\n",
    "    val[0][modnames[i]] = pd.Series(aconc[i].variables['PM25_TOT'][:,esty,estx], index=val_pm25.index)\n",
    "# pm10 = PM10            ,ug/m3 \n",
    "    val[1][modnames[i]] = pd.Series(aconc[i].variables['PM10'][:,esty,estx], index=val_pm10.index)\n",
    "# o3 = O3_UGM3        ,ug/m3\n",
    "    val[2][modnames[i]] = pd.Series(aconc[i].variables['O3_UGM3'][:,esty,estx], index=val_o3.index)\n",
    "# nox = NOX_UGM3        ,ug/m3\n",
    "    val[3][modnames[i]] = pd.Series(aconc[i].variables['NOX_UGM3'][:,esty,estx], index=val_nox.index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Funções para validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "code_folding": []
   },
   "outputs": [],
   "source": [
    "#RootMeanSquareError - RMSE\n",
    "def meansqError(lmodel,lobs):\n",
    "    sqerr = 0\n",
    "    n = 0\n",
    "    for i in range(0,len(lmodel)):\n",
    "        if lobs[i] is not None:\n",
    "            sqerr += ((lmodel[i]-lobs[i])**2)\n",
    "            n += 1\n",
    "    if n != 0:\n",
    "        sqerr /= n\n",
    "        sqerr = sqerr**0.5\n",
    "        return sqerr\n",
    "    else:\n",
    "        return None\n",
    "\n",
    "#Index of Agreement - IOA\n",
    "def indAgree (lmodel, lobs):\n",
    "    ioa = 0\n",
    "    num = 0\n",
    "    den = 0\n",
    "    n   = 0\n",
    "    obsavg = 0 \n",
    "    for i in range(0,len(lobs)):\n",
    "        if lobs[i] is not None:\n",
    "            obsavg += lobs[i]\n",
    "            n += 1\n",
    "    if n != 0: \n",
    "        obsavg /= n\n",
    "    else:\n",
    "        obsavg = None\n",
    "    if obsavg != None:\n",
    "        for i in range(0,len(lmodel)):\n",
    "            if lobs[i] is not None:\n",
    "               num += (lmodel[i]-lobs[i])**2\n",
    "               den1 = np.absolute(lmodel[i]-obsavg)\n",
    "               den2 = np.absolute(lobs[i]-obsavg)\n",
    "               den += (den1+den2)**2\n",
    "        ioa = 1-(num/den)\n",
    "        return ioa\n",
    "    else:\n",
    "        return None\n",
    "\n",
    "#MeanAbsoluteGrossError - MAGE\n",
    "def grossError(lmodel,lobs):\n",
    "    age = 0\n",
    "    n = 0\n",
    "    for i in range(0,len(lmodel)):\n",
    "        if lobs[i] is not None:\n",
    "            age += np.absolute(lmodel[i]-lobs[i])\n",
    "            n += 1\n",
    "    if n != 0:\n",
    "        return age/n\n",
    "    else:\n",
    "        return None\n",
    "        \n",
    "#Coeficiente de correlação (r): \n",
    "def cor_r(lmodel, lobs):\n",
    "    dic = {'model':lmodel,'obs':lobs}\n",
    "    df = pd.DataFrame(dic)\n",
    "    df.dropna(subset=['obs'],inplace=True)\n",
    "    return np.corrcoef(df['model'],df['obs'])[0,1]\n",
    "    #cov = np.cov(df['model'],df['obs'])[0,1]\n",
    "    #stdobs = np.std(df['obs'])\n",
    "    #stdmod = np.std(df['model'])\n",
    "    #return cov/(stdobs*stdmod)\n",
    "\n",
    "#MeanBias - MB\n",
    "def meanBias(lmodel,lobs):\n",
    "    bias = 0\n",
    "    n = 0\n",
    "    for i in range(0,len(lmodel)):\n",
    "#        print (i)\n",
    "#        print (lmodel[i])\n",
    "#        print (lobs[i])\n",
    "        if lobs[i] is not None:\n",
    "            bias += (lmodel[i]-lobs[i])\n",
    "            n += 1\n",
    "    if n != 0:\n",
    "        return bias/n\n",
    "    else:\n",
    "        return None    \n",
    "\n",
    "#MeanError - ME\n",
    "def meanError(lmodel,lobs):\n",
    "    bias = 0\n",
    "    n = 0\n",
    "    for i in range(0,len(lmodel)):\n",
    "#        print (i)\n",
    "#        print (lmodel[i])\n",
    "#        print (lobs[i])\n",
    "        if lobs[i] is not None:\n",
    "            bias += np.absolute(lmodel[i]-lobs[i])\n",
    "            n += 1\n",
    "    if n != 0:\n",
    "        return bias/n\n",
    "    else:\n",
    "        return None  \n",
    "    \n",
    "\n",
    "#Normalized Mean Bias – NMB\n",
    "def normMeanBias(lmodel, lobs):\n",
    "    bias = 0\n",
    "    n = 0\n",
    "    soma = 0\n",
    "    for i in range(0,len(lmodel)):\n",
    "        if lobs[i] is not None:\n",
    "            bias += (lmodel[i]-lobs[i])\n",
    "            soma += lobs[i]\n",
    "            n += 1            \n",
    "    if n != 0:\n",
    "        return (bias/soma)\n",
    "    else:\n",
    "        return None    \n",
    "\n",
    "#Normalized Mean Error – NME\n",
    "def normMeanError(lmodel, lobs):\n",
    "    err = 0\n",
    "    n = 0\n",
    "    soma = 0\n",
    "    for i in range(0,len(lmodel)):\n",
    "        if lobs[i] is not None:\n",
    "            err += np.absolute(lmodel[i]-lobs[i])\n",
    "            soma += lobs[i]\n",
    "            n += 1            \n",
    "    if n != 0:\n",
    "        return (err/soma)\n",
    "    else:\n",
    "        return None \n",
    "\n",
    "#Mean Fractional Bias – MFB\n",
    "def meanFracBias(lmodel, lobs):\n",
    "    fracbias = 0\n",
    "    n = 0\n",
    "    soma = 0\n",
    "    for i in range(0,len(lmodel)):\n",
    "        if lobs[i] is not None:\n",
    "            fracbias += ( (lmodel[i]-lobs[i]) / ((lmodel[i]+lobs[i]) / 2))\n",
    "            n += 1            \n",
    "    if n != 0:\n",
    "        return fracbias/n\n",
    "    else:\n",
    "        return None   \n",
    "\n",
    "#Mean Fractional Error – MFE\n",
    "def meanFracError(lmodel, lobs):\n",
    "    fracerr = 0\n",
    "    n = 0\n",
    "    soma = 0\n",
    "    for i in range(0,len(lmodel)):\n",
    "        if lobs[i] is not None:\n",
    "            fracerr += ( np.absolute(lmodel[i]-lobs[i]) / ((lmodel[i]+lobs[i]) / 2))\n",
    "            n += 1            \n",
    "    if n != 0:\n",
    "        return fracerr/n\n",
    "    else:\n",
    "        return None \n",
    "    \n",
    "#Mean Fractional Bias Goal - MFBGoal\n",
    "def mfbGoal(lmodel, lobs):\n",
    "    print (np.average(lobs))\n",
    "    print (np.average(lmodel))\n",
    "    return 170 * np.exp(-0.5 * ( np.average(lobs) + np.average(lmodel) ) / 0.5) + 30 \n",
    "\n",
    "#Mean Fractional Error Goal - MFEGoal\n",
    "def mfeGoal(lmodel, lobs):\n",
    "    return 150 * np.exp(-0.5 * ( np.average(lobs) + np.average(lmodel) ) / 0.75) + 50\n",
    "\n",
    "#Mean Fractional Bias Criteria - MFBCriteria\n",
    "def mfbCriteria(lmodel, lobs):\n",
    "    return 140 * np.exp(-0.5 * ( np.average(lobs) + np.average(lmodel) ) / 0.5) + 60\n",
    "\n",
    "#Mean Fractional Error Criteria - MFECriteria\n",
    "def mfeCriteria(lmodel, lobs):\n",
    "    return 125 * np.exp(-0.5 * ( np.average(lobs) + np.average(lmodel) ) / 0.75) + 75\n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Validação"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "#DROPNAS\n",
    "for i in range(0,len(val)):\n",
    "    val[i].dropna(subset=[pol[i]],inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16.732142857142858\n",
      "2.7138033\n",
      "16.732142857142858\n",
      "29.422152\n",
      "35.20238095238095\n",
      "3.1717827\n",
      "35.20238095238095\n",
      "29.79194\n",
      "103.0728476821192\n",
      "66.16903\n",
      "103.0728476821192\n",
      "40.45242\n",
      "8.108108108108109\n",
      "0.99224347\n",
      "8.108108108108109\n",
      "92.73476\n"
     ]
    }
   ],
   "source": [
    "cols = ['Poluentes','Mes','Mod',#'MB',\n",
    "        'RMSE',#'MAGE','IOA',\n",
    "        'NMB','NME',\n",
    "        'MFB','MFBGoal','MFBCriteria',\n",
    "        'MFE', 'MFEGoal', 'MFECriteria',\n",
    "        'r']\n",
    "cpoluentes = []\n",
    "cmeses = []\n",
    "cmodelos = []\n",
    "#cMB   = []\n",
    "cRMSE = []\n",
    "#cMAGE = []\n",
    "#cIOA  = []\n",
    "cNMB  = []\n",
    "cNME  = []\n",
    "\n",
    "cMFB  = []\n",
    "cMFBgoal = []\n",
    "cMFBcriteria = []\n",
    "\n",
    "cMFE  = []\n",
    "cMFEgoal = []\n",
    "cMFEcriteria =[]\n",
    "\n",
    "cr    = []\n",
    "\n",
    "for i in range(0,len(pol)):\n",
    "    for m in modnames:\n",
    "        cpoluentes.append(pol[i])\n",
    "        cmeses.append(mes[date_index])\n",
    "        cmodelos.append(m)\n",
    "        #lmodel,lobs\n",
    "        #cMB.append(   meanBias      (val[i][m],val[i][pol[i]])) \n",
    "        cRMSE.append( meansqError   (val[i][m],val[i][pol[i]]))\n",
    "        #cMAGE.append( grossError    (val[i][m],val[i][pol[i]]))\n",
    "        #cIOA.append(  indAgree      (val[i][m],val[i][pol[i]]))\n",
    "        cNMB.append(  normMeanBias  (val[i][m],val[i][pol[i]])) \n",
    "        cNME.append(  normMeanError (val[i][m],val[i][pol[i]])) \n",
    "        \n",
    "        cMFB.append(  meanFracBias  (val[i][m],val[i][pol[i]])) \n",
    "        cMFBgoal.append( mfbGoal (val[i][m],val[i][pol[i]]))\n",
    "        cMFBcriteria.append( mfbCriteria (val[i][m],val[i][pol[i]]))\n",
    "        \n",
    "        cMFE.append(  meanFracError (val[i][m],val[i][pol[i]])) \n",
    "        cMFEgoal.append( mfeGoal (val[i][m],val[i][pol[i]]))\n",
    "        cMFEcriteria.append( mfeCriteria (val[i][m],val[i][pol[i]]))\n",
    "        \n",
    "        cr.append(    cor_r         (val[i][m],val[i][pol[i]]))\n",
    "\n",
    "d = {cols[0]:cpoluentes,\n",
    "    cols[1]:cmeses,\n",
    "    cols[2]:cmodelos,\n",
    "    #cols[]:cMB,\n",
    "    cols[3]:cRMSE,\n",
    "    #cols[]:cMAGE,\n",
    "    #cols[]:cIOA,\n",
    "    cols[4]:cNMB,\n",
    "    cols[5]:cNME,\n",
    "    cols[6]:cMFB,\n",
    "    cols[7]:cMFBgoal,\n",
    "    cols[8]:cMFBcriteria,\n",
    "    cols[9]:cMFE,\n",
    "    cols[10]:cMFEgoal,\n",
    "    cols[11]:cMFEcriteria,\n",
    "    cols[12]:cr}\n",
    "\n",
    "validf = pd.DataFrame(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>Poluentes</th>\n",
       "      <th>Mes</th>\n",
       "      <th>Mod</th>\n",
       "      <th>RMSE</th>\n",
       "      <th>NMB</th>\n",
       "      <th>NME</th>\n",
       "      <th>MFB</th>\n",
       "      <th>MFBGoal</th>\n",
       "      <th>MFBCriteria</th>\n",
       "      <th>MFE</th>\n",
       "      <th>MFEGoal</th>\n",
       "      <th>MFECriteria</th>\n",
       "      <th>r</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MP2_5</td>\n",
       "      <td>jan</td>\n",
       "      <td>ED</td>\n",
       "      <td>16.038601</td>\n",
       "      <td>-0.837809</td>\n",
       "      <td>0.842305</td>\n",
       "      <td>-1.316301</td>\n",
       "      <td>30.000001</td>\n",
       "      <td>60.000001</td>\n",
       "      <td>1.359011</td>\n",
       "      <td>50.000351</td>\n",
       "      <td>75.000293</td>\n",
       "      <td>0.169144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MP2_5</td>\n",
       "      <td>jan</td>\n",
       "      <td>PT_ED-100I</td>\n",
       "      <td>43.664946</td>\n",
       "      <td>0.758421</td>\n",
       "      <td>1.800284</td>\n",
       "      <td>-0.281106</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>1.274814</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>-0.153749</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MP10</td>\n",
       "      <td>jan</td>\n",
       "      <td>ED</td>\n",
       "      <td>34.557808</td>\n",
       "      <td>-0.909899</td>\n",
       "      <td>0.909899</td>\n",
       "      <td>-1.635130</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>1.635130</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>0.228066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>MP10</td>\n",
       "      <td>jan</td>\n",
       "      <td>PT_ED-100I</td>\n",
       "      <td>43.420072</td>\n",
       "      <td>-0.153695</td>\n",
       "      <td>1.013347</td>\n",
       "      <td>-0.702809</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>1.229575</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>-0.102974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>O3</td>\n",
       "      <td>jan</td>\n",
       "      <td>ED</td>\n",
       "      <td>56.766913</td>\n",
       "      <td>-0.358036</td>\n",
       "      <td>0.450181</td>\n",
       "      <td>-0.336958</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>0.504406</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>0.002603</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>O3</td>\n",
       "      <td>jan</td>\n",
       "      <td>PT_ED-100I</td>\n",
       "      <td>86.102992</td>\n",
       "      <td>-0.607536</td>\n",
       "      <td>0.669261</td>\n",
       "      <td>-0.827919</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>0.941536</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>-0.435027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>NOx</td>\n",
       "      <td>jan</td>\n",
       "      <td>ED</td>\n",
       "      <td>9.816419</td>\n",
       "      <td>-0.877623</td>\n",
       "      <td>0.908205</td>\n",
       "      <td>-1.162409</td>\n",
       "      <td>30.018977</td>\n",
       "      <td>60.015628</td>\n",
       "      <td>1.475593</td>\n",
       "      <td>50.347752</td>\n",
       "      <td>75.289793</td>\n",
       "      <td>-0.082429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>NOx</td>\n",
       "      <td>jan</td>\n",
       "      <td>PT_ED-100I</td>\n",
       "      <td>150.708360</td>\n",
       "      <td>10.437288</td>\n",
       "      <td>11.412601</td>\n",
       "      <td>0.222385</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>1.646892</td>\n",
       "      <td>50.000000</td>\n",
       "      <td>75.000000</td>\n",
       "      <td>-0.193325</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Poluentes  Mes         Mod        RMSE        NMB        NME       MFB  \\\n",
       "0     MP2_5  jan          ED   16.038601  -0.837809   0.842305 -1.316301   \n",
       "1     MP2_5  jan  PT_ED-100I   43.664946   0.758421   1.800284 -0.281106   \n",
       "2      MP10  jan          ED   34.557808  -0.909899   0.909899 -1.635130   \n",
       "3      MP10  jan  PT_ED-100I   43.420072  -0.153695   1.013347 -0.702809   \n",
       "4        O3  jan          ED   56.766913  -0.358036   0.450181 -0.336958   \n",
       "5        O3  jan  PT_ED-100I   86.102992  -0.607536   0.669261 -0.827919   \n",
       "6       NOx  jan          ED    9.816419  -0.877623   0.908205 -1.162409   \n",
       "7       NOx  jan  PT_ED-100I  150.708360  10.437288  11.412601  0.222385   \n",
       "\n",
       "     MFBGoal  MFBCriteria       MFE    MFEGoal  MFECriteria         r  \n",
       "0  30.000001    60.000001  1.359011  50.000351    75.000293  0.169144  \n",
       "1  30.000000    60.000000  1.274814  50.000000    75.000000 -0.153749  \n",
       "2  30.000000    60.000000  1.635130  50.000000    75.000000  0.228066  \n",
       "3  30.000000    60.000000  1.229575  50.000000    75.000000 -0.102974  \n",
       "4  30.000000    60.000000  0.504406  50.000000    75.000000  0.002603  \n",
       "5  30.000000    60.000000  0.941536  50.000000    75.000000 -0.435027  \n",
       "6  30.018977    60.015628  1.475593  50.347752    75.289793 -0.082429  \n",
       "7  30.000000    60.000000  1.646892  50.000000    75.000000 -0.193325  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['ED', 'PT_ED-100I']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "modnames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['MP2_5', 'MP10', 'O3', 'NOx']"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pol"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "list"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00260305442609114"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.corrcoef(val[2][pol[2]], val[2][modnames[0]])[0,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-0.4350266386727775"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.corrcoef(val[2]['O3'], val[2]['PT_ED-100I'])[0,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "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>O3</th>\n",
       "      <th>ED</th>\n",
       "      <th>PT_ED-100I</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Data</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-01-11 07:00:00</th>\n",
       "      <td>36.0</td>\n",
       "      <td>66.898804</td>\n",
       "      <td>66.825134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 08:00:00</th>\n",
       "      <td>37.0</td>\n",
       "      <td>66.502579</td>\n",
       "      <td>33.704323</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 09:00:00</th>\n",
       "      <td>49.0</td>\n",
       "      <td>71.291405</td>\n",
       "      <td>12.806714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 10:00:00</th>\n",
       "      <td>75.0</td>\n",
       "      <td>66.554436</td>\n",
       "      <td>13.000618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 11:00:00</th>\n",
       "      <td>110.0</td>\n",
       "      <td>60.916656</td>\n",
       "      <td>15.497415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 12:00:00</th>\n",
       "      <td>130.0</td>\n",
       "      <td>58.404469</td>\n",
       "      <td>6.338462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 13:00:00</th>\n",
       "      <td>139.0</td>\n",
       "      <td>58.988152</td>\n",
       "      <td>4.399853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 14:00:00</th>\n",
       "      <td>132.0</td>\n",
       "      <td>60.425591</td>\n",
       "      <td>15.050270</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 15:00:00</th>\n",
       "      <td>127.0</td>\n",
       "      <td>62.350060</td>\n",
       "      <td>39.153809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 16:00:00</th>\n",
       "      <td>125.0</td>\n",
       "      <td>60.934326</td>\n",
       "      <td>21.979267</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 17:00:00</th>\n",
       "      <td>123.0</td>\n",
       "      <td>60.322174</td>\n",
       "      <td>4.280704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 18:00:00</th>\n",
       "      <td>108.0</td>\n",
       "      <td>60.380344</td>\n",
       "      <td>5.623199</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 19:00:00</th>\n",
       "      <td>103.0</td>\n",
       "      <td>59.178963</td>\n",
       "      <td>37.650108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 20:00:00</th>\n",
       "      <td>84.0</td>\n",
       "      <td>59.479298</td>\n",
       "      <td>56.388103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 21:00:00</th>\n",
       "      <td>81.0</td>\n",
       "      <td>63.722382</td>\n",
       "      <td>63.726364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 22:00:00</th>\n",
       "      <td>67.0</td>\n",
       "      <td>63.100903</td>\n",
       "      <td>63.095402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-11 23:00:00</th>\n",
       "      <td>64.0</td>\n",
       "      <td>62.329220</td>\n",
       "      <td>62.322430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 00:00:00</th>\n",
       "      <td>66.0</td>\n",
       "      <td>62.048626</td>\n",
       "      <td>62.044731</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 01:00:00</th>\n",
       "      <td>58.0</td>\n",
       "      <td>62.210827</td>\n",
       "      <td>62.210403</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 02:00:00</th>\n",
       "      <td>76.0</td>\n",
       "      <td>62.314159</td>\n",
       "      <td>62.313747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 03:00:00</th>\n",
       "      <td>67.0</td>\n",
       "      <td>61.993073</td>\n",
       "      <td>61.992752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 04:00:00</th>\n",
       "      <td>81.0</td>\n",
       "      <td>62.240459</td>\n",
       "      <td>62.238091</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 05:00:00</th>\n",
       "      <td>99.0</td>\n",
       "      <td>62.867214</td>\n",
       "      <td>62.843124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 07:00:00</th>\n",
       "      <td>103.0</td>\n",
       "      <td>62.879799</td>\n",
       "      <td>62.886642</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 08:00:00</th>\n",
       "      <td>113.0</td>\n",
       "      <td>63.323242</td>\n",
       "      <td>29.633432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 09:00:00</th>\n",
       "      <td>102.0</td>\n",
       "      <td>63.308018</td>\n",
       "      <td>8.964066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 10:00:00</th>\n",
       "      <td>109.0</td>\n",
       "      <td>59.620358</td>\n",
       "      <td>8.762889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 11:00:00</th>\n",
       "      <td>128.0</td>\n",
       "      <td>58.172176</td>\n",
       "      <td>8.989385</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 12:00:00</th>\n",
       "      <td>136.0</td>\n",
       "      <td>58.645607</td>\n",
       "      <td>13.807670</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-12 13:00:00</th>\n",
       "      <td>155.0</td>\n",
       "      <td>60.517567</td>\n",
       "      <td>14.120878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 14:00:00</th>\n",
       "      <td>136.0</td>\n",
       "      <td>68.695404</td>\n",
       "      <td>17.244690</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 15:00:00</th>\n",
       "      <td>136.0</td>\n",
       "      <td>71.761696</td>\n",
       "      <td>10.685146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 16:00:00</th>\n",
       "      <td>139.0</td>\n",
       "      <td>74.280098</td>\n",
       "      <td>15.908505</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 17:00:00</th>\n",
       "      <td>139.0</td>\n",
       "      <td>79.634392</td>\n",
       "      <td>13.937986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 18:00:00</th>\n",
       "      <td>152.0</td>\n",
       "      <td>77.040054</td>\n",
       "      <td>1.189541</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 19:00:00</th>\n",
       "      <td>150.0</td>\n",
       "      <td>76.966202</td>\n",
       "      <td>26.693005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 20:00:00</th>\n",
       "      <td>130.0</td>\n",
       "      <td>74.999680</td>\n",
       "      <td>55.258286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 21:00:00</th>\n",
       "      <td>106.0</td>\n",
       "      <td>78.838821</td>\n",
       "      <td>79.158325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 22:00:00</th>\n",
       "      <td>77.0</td>\n",
       "      <td>85.194389</td>\n",
       "      <td>81.802116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-16 23:00:00</th>\n",
       "      <td>74.0</td>\n",
       "      <td>91.218941</td>\n",
       "      <td>88.228966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 00:00:00</th>\n",
       "      <td>105.0</td>\n",
       "      <td>93.873268</td>\n",
       "      <td>91.801300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 01:00:00</th>\n",
       "      <td>99.0</td>\n",
       "      <td>95.003609</td>\n",
       "      <td>93.490944</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 02:00:00</th>\n",
       "      <td>82.0</td>\n",
       "      <td>95.339752</td>\n",
       "      <td>94.587959</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 03:00:00</th>\n",
       "      <td>73.0</td>\n",
       "      <td>96.533623</td>\n",
       "      <td>96.152962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 04:00:00</th>\n",
       "      <td>57.0</td>\n",
       "      <td>91.328659</td>\n",
       "      <td>91.181595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 05:00:00</th>\n",
       "      <td>51.0</td>\n",
       "      <td>82.665459</td>\n",
       "      <td>82.543350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 07:00:00</th>\n",
       "      <td>42.0</td>\n",
       "      <td>81.357315</td>\n",
       "      <td>80.894531</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 08:00:00</th>\n",
       "      <td>37.0</td>\n",
       "      <td>77.644920</td>\n",
       "      <td>32.026337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 09:00:00</th>\n",
       "      <td>55.0</td>\n",
       "      <td>67.907494</td>\n",
       "      <td>13.667947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 10:00:00</th>\n",
       "      <td>82.0</td>\n",
       "      <td>64.437469</td>\n",
       "      <td>16.783344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 11:00:00</th>\n",
       "      <td>121.0</td>\n",
       "      <td>60.964432</td>\n",
       "      <td>9.511026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 12:00:00</th>\n",
       "      <td>136.0</td>\n",
       "      <td>59.010521</td>\n",
       "      <td>7.153689</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 13:00:00</th>\n",
       "      <td>155.0</td>\n",
       "      <td>58.489700</td>\n",
       "      <td>4.154863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 14:00:00</th>\n",
       "      <td>159.0</td>\n",
       "      <td>58.872272</td>\n",
       "      <td>7.086881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 15:00:00</th>\n",
       "      <td>149.0</td>\n",
       "      <td>59.487816</td>\n",
       "      <td>6.458574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 16:00:00</th>\n",
       "      <td>149.0</td>\n",
       "      <td>60.475124</td>\n",
       "      <td>6.830217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 17:00:00</th>\n",
       "      <td>153.0</td>\n",
       "      <td>59.088963</td>\n",
       "      <td>6.373579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 18:00:00</th>\n",
       "      <td>151.0</td>\n",
       "      <td>58.455376</td>\n",
       "      <td>2.698661</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 19:00:00</th>\n",
       "      <td>152.0</td>\n",
       "      <td>57.749027</td>\n",
       "      <td>0.000994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-01-17 20:00:00</th>\n",
       "      <td>149.0</td>\n",
       "      <td>55.860813</td>\n",
       "      <td>5.335095</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>151 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                        O3         ED  PT_ED-100I\n",
       "Data                                             \n",
       "2015-01-11 07:00:00   36.0  66.898804   66.825134\n",
       "2015-01-11 08:00:00   37.0  66.502579   33.704323\n",
       "2015-01-11 09:00:00   49.0  71.291405   12.806714\n",
       "2015-01-11 10:00:00   75.0  66.554436   13.000618\n",
       "2015-01-11 11:00:00  110.0  60.916656   15.497415\n",
       "2015-01-11 12:00:00  130.0  58.404469    6.338462\n",
       "2015-01-11 13:00:00  139.0  58.988152    4.399853\n",
       "2015-01-11 14:00:00  132.0  60.425591   15.050270\n",
       "2015-01-11 15:00:00  127.0  62.350060   39.153809\n",
       "2015-01-11 16:00:00  125.0  60.934326   21.979267\n",
       "2015-01-11 17:00:00  123.0  60.322174    4.280704\n",
       "2015-01-11 18:00:00  108.0  60.380344    5.623199\n",
       "2015-01-11 19:00:00  103.0  59.178963   37.650108\n",
       "2015-01-11 20:00:00   84.0  59.479298   56.388103\n",
       "2015-01-11 21:00:00   81.0  63.722382   63.726364\n",
       "2015-01-11 22:00:00   67.0  63.100903   63.095402\n",
       "2015-01-11 23:00:00   64.0  62.329220   62.322430\n",
       "2015-01-12 00:00:00   66.0  62.048626   62.044731\n",
       "2015-01-12 01:00:00   58.0  62.210827   62.210403\n",
       "2015-01-12 02:00:00   76.0  62.314159   62.313747\n",
       "2015-01-12 03:00:00   67.0  61.993073   61.992752\n",
       "2015-01-12 04:00:00   81.0  62.240459   62.238091\n",
       "2015-01-12 05:00:00   99.0  62.867214   62.843124\n",
       "2015-01-12 07:00:00  103.0  62.879799   62.886642\n",
       "2015-01-12 08:00:00  113.0  63.323242   29.633432\n",
       "2015-01-12 09:00:00  102.0  63.308018    8.964066\n",
       "2015-01-12 10:00:00  109.0  59.620358    8.762889\n",
       "2015-01-12 11:00:00  128.0  58.172176    8.989385\n",
       "2015-01-12 12:00:00  136.0  58.645607   13.807670\n",
       "2015-01-12 13:00:00  155.0  60.517567   14.120878\n",
       "...                    ...        ...         ...\n",
       "2015-01-16 14:00:00  136.0  68.695404   17.244690\n",
       "2015-01-16 15:00:00  136.0  71.761696   10.685146\n",
       "2015-01-16 16:00:00  139.0  74.280098   15.908505\n",
       "2015-01-16 17:00:00  139.0  79.634392   13.937986\n",
       "2015-01-16 18:00:00  152.0  77.040054    1.189541\n",
       "2015-01-16 19:00:00  150.0  76.966202   26.693005\n",
       "2015-01-16 20:00:00  130.0  74.999680   55.258286\n",
       "2015-01-16 21:00:00  106.0  78.838821   79.158325\n",
       "2015-01-16 22:00:00   77.0  85.194389   81.802116\n",
       "2015-01-16 23:00:00   74.0  91.218941   88.228966\n",
       "2015-01-17 00:00:00  105.0  93.873268   91.801300\n",
       "2015-01-17 01:00:00   99.0  95.003609   93.490944\n",
       "2015-01-17 02:00:00   82.0  95.339752   94.587959\n",
       "2015-01-17 03:00:00   73.0  96.533623   96.152962\n",
       "2015-01-17 04:00:00   57.0  91.328659   91.181595\n",
       "2015-01-17 05:00:00   51.0  82.665459   82.543350\n",
       "2015-01-17 07:00:00   42.0  81.357315   80.894531\n",
       "2015-01-17 08:00:00   37.0  77.644920   32.026337\n",
       "2015-01-17 09:00:00   55.0  67.907494   13.667947\n",
       "2015-01-17 10:00:00   82.0  64.437469   16.783344\n",
       "2015-01-17 11:00:00  121.0  60.964432    9.511026\n",
       "2015-01-17 12:00:00  136.0  59.010521    7.153689\n",
       "2015-01-17 13:00:00  155.0  58.489700    4.154863\n",
       "2015-01-17 14:00:00  159.0  58.872272    7.086881\n",
       "2015-01-17 15:00:00  149.0  59.487816    6.458574\n",
       "2015-01-17 16:00:00  149.0  60.475124    6.830217\n",
       "2015-01-17 17:00:00  153.0  59.088963    6.373579\n",
       "2015-01-17 18:00:00  151.0  58.455376    2.698661\n",
       "2015-01-17 19:00:00  152.0  57.749027    0.000994\n",
       "2015-01-17 20:00:00  149.0  55.860813    5.335095\n",
       "\n",
       "[151 rows x 3 columns]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val[2][pol[2]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Data\n",
       "2015-01-11 07:00:00     36.0\n",
       "2015-01-11 08:00:00     37.0\n",
       "2015-01-11 09:00:00     49.0\n",
       "2015-01-11 10:00:00     75.0\n",
       "2015-01-11 11:00:00    110.0\n",
       "2015-01-11 12:00:00    130.0\n",
       "2015-01-11 13:00:00    139.0\n",
       "2015-01-11 14:00:00    132.0\n",
       "2015-01-11 15:00:00    127.0\n",
       "2015-01-11 16:00:00    125.0\n",
       "2015-01-11 17:00:00    123.0\n",
       "2015-01-11 18:00:00    108.0\n",
       "2015-01-11 19:00:00    103.0\n",
       "2015-01-11 20:00:00     84.0\n",
       "2015-01-11 21:00:00     81.0\n",
       "2015-01-11 22:00:00     67.0\n",
       "2015-01-11 23:00:00     64.0\n",
       "2015-01-12 00:00:00     66.0\n",
       "2015-01-12 01:00:00     58.0\n",
       "2015-01-12 02:00:00     76.0\n",
       "2015-01-12 03:00:00     67.0\n",
       "2015-01-12 04:00:00     81.0\n",
       "2015-01-12 05:00:00     99.0\n",
       "2015-01-12 07:00:00    103.0\n",
       "2015-01-12 08:00:00    113.0\n",
       "2015-01-12 09:00:00    102.0\n",
       "2015-01-12 10:00:00    109.0\n",
       "2015-01-12 11:00:00    128.0\n",
       "2015-01-12 12:00:00    136.0\n",
       "2015-01-12 13:00:00    155.0\n",
       "                       ...  \n",
       "2015-01-16 14:00:00    136.0\n",
       "2015-01-16 15:00:00    136.0\n",
       "2015-01-16 16:00:00    139.0\n",
       "2015-01-16 17:00:00    139.0\n",
       "2015-01-16 18:00:00    152.0\n",
       "2015-01-16 19:00:00    150.0\n",
       "2015-01-16 20:00:00    130.0\n",
       "2015-01-16 21:00:00    106.0\n",
       "2015-01-16 22:00:00     77.0\n",
       "2015-01-16 23:00:00     74.0\n",
       "2015-01-17 00:00:00    105.0\n",
       "2015-01-17 01:00:00     99.0\n",
       "2015-01-17 02:00:00     82.0\n",
       "2015-01-17 03:00:00     73.0\n",
       "2015-01-17 04:00:00     57.0\n",
       "2015-01-17 05:00:00     51.0\n",
       "2015-01-17 07:00:00     42.0\n",
       "2015-01-17 08:00:00     37.0\n",
       "2015-01-17 09:00:00     55.0\n",
       "2015-01-17 10:00:00     82.0\n",
       "2015-01-17 11:00:00    121.0\n",
       "2015-01-17 12:00:00    136.0\n",
       "2015-01-17 13:00:00    155.0\n",
       "2015-01-17 14:00:00    159.0\n",
       "2015-01-17 15:00:00    149.0\n",
       "2015-01-17 16:00:00    149.0\n",
       "2015-01-17 17:00:00    153.0\n",
       "2015-01-17 18:00:00    151.0\n",
       "2015-01-17 19:00:00    152.0\n",
       "2015-01-17 20:00:00    149.0\n",
       "Name: O3, Length: 151, dtype: float64"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val[2][pol[2]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Data\n",
       "2015-01-11 07:00:00    66.825134\n",
       "2015-01-11 08:00:00    33.704323\n",
       "2015-01-11 09:00:00    12.806714\n",
       "2015-01-11 10:00:00    13.000618\n",
       "2015-01-11 11:00:00    15.497415\n",
       "2015-01-11 12:00:00     6.338462\n",
       "2015-01-11 13:00:00     4.399853\n",
       "2015-01-11 14:00:00    15.050270\n",
       "2015-01-11 15:00:00    39.153809\n",
       "2015-01-11 16:00:00    21.979267\n",
       "2015-01-11 17:00:00     4.280704\n",
       "2015-01-11 18:00:00     5.623199\n",
       "2015-01-11 19:00:00    37.650108\n",
       "2015-01-11 20:00:00    56.388103\n",
       "2015-01-11 21:00:00    63.726364\n",
       "2015-01-11 22:00:00    63.095402\n",
       "2015-01-11 23:00:00    62.322430\n",
       "2015-01-12 00:00:00    62.044731\n",
       "2015-01-12 01:00:00    62.210403\n",
       "2015-01-12 02:00:00    62.313747\n",
       "2015-01-12 03:00:00    61.992752\n",
       "2015-01-12 04:00:00    62.238091\n",
       "2015-01-12 05:00:00    62.843124\n",
       "2015-01-12 07:00:00    62.886642\n",
       "2015-01-12 08:00:00    29.633432\n",
       "2015-01-12 09:00:00     8.964066\n",
       "2015-01-12 10:00:00     8.762889\n",
       "2015-01-12 11:00:00     8.989385\n",
       "2015-01-12 12:00:00    13.807670\n",
       "2015-01-12 13:00:00    14.120878\n",
       "                         ...    \n",
       "2015-01-16 14:00:00    17.244690\n",
       "2015-01-16 15:00:00    10.685146\n",
       "2015-01-16 16:00:00    15.908505\n",
       "2015-01-16 17:00:00    13.937986\n",
       "2015-01-16 18:00:00     1.189541\n",
       "2015-01-16 19:00:00    26.693005\n",
       "2015-01-16 20:00:00    55.258286\n",
       "2015-01-16 21:00:00    79.158325\n",
       "2015-01-16 22:00:00    81.802116\n",
       "2015-01-16 23:00:00    88.228966\n",
       "2015-01-17 00:00:00    91.801300\n",
       "2015-01-17 01:00:00    93.490944\n",
       "2015-01-17 02:00:00    94.587959\n",
       "2015-01-17 03:00:00    96.152962\n",
       "2015-01-17 04:00:00    91.181595\n",
       "2015-01-17 05:00:00    82.543350\n",
       "2015-01-17 07:00:00    80.894531\n",
       "2015-01-17 08:00:00    32.026337\n",
       "2015-01-17 09:00:00    13.667947\n",
       "2015-01-17 10:00:00    16.783344\n",
       "2015-01-17 11:00:00     9.511026\n",
       "2015-01-17 12:00:00     7.153689\n",
       "2015-01-17 13:00:00     4.154863\n",
       "2015-01-17 14:00:00     7.086881\n",
       "2015-01-17 15:00:00     6.458574\n",
       "2015-01-17 16:00:00     6.830217\n",
       "2015-01-17 17:00:00     6.373579\n",
       "2015-01-17 18:00:00     2.698661\n",
       "2015-01-17 19:00:00     0.000994\n",
       "2015-01-17 20:00:00     5.335095\n",
       "Name: PT_ED-100I, Length: 151, dtype: float32"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val[2][modnames[1]]"
   ]
  },
  {
   "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": "3.7.3"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "oldHeight": 459,
   "position": {
    "height": "40px",
    "left": "1080px",
    "right": "20px",
    "top": "79px",
    "width": "266px"
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "varInspector_section_display": "none",
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}