QuantumDamage

import numpy as np

from sklearn.model_selection import train_test_split
from xgboost import XGBClassifier

# NOTE: Make sure that the class is labeled 'class' in the data file
tpot_data = np.recfromcsv('PATH/TO/DATA/FILE', delimiter='COLUMN_SEPARATOR', dtype=np.float64)
features = np.delete(tpot_data.view(np.float64).reshape(tpot_data.size, -1), tpot_data.dtype.names.index('class'), axis=1)
training_features, testing_features, training_classes, testing_classes = \
    train_test_split(features, tpot_data['class'], random_state=42)

QuantumDamage

r = requests.get('http://api.gios.gov.pl/pjp-api/rest/station/findAll')
allStations = json_normalize(r.json())
print(allStations[allStations["city.name"] == u"Gdańsk"])

QuantumDamage

                  addressStreet  city city.commune.communeName  \
2   ul. Powstańców Warszawskich   NaN                   Gdańsk   
14                 ul. Kaczeńce   NaN                   Gdańsk   
26               ul. Wyzwolenia   NaN                   Gdańsk   
67                 ul. Leczkowa   NaN                   Gdańsk   
75                ul. Ostrzycka   NaN                   Gdańsk   

   city.commune.districtName city.commune.provinceName  city.id city.name  \
2                     Gdańsk                 POMORSKIE    218.0    Gdańsk   
14                    Gdańsk                 POMORSKIE    218.0    Gdańsk   

QuantumDamage

stationId = 733
r = requests.get('http://api.gios.gov.pl/pjp-api/rest/station/sensors/' + str(stationId))
sensors = json_normalize(r.json())
print(sensors)

QuantumDamage

     id  param.idParam param.paramCode param.paramFormula  \
0  4720              8              CO                 CO   
1  4727              3            PM10               PM10   
2  4723              6             NO2                NO2   
3  4725              5              O3                 O3   
4  4730              1             SO2                SO2   

       param.paramName sensorDateEnd      sensorDateStart  stationId  
0         tlenek węgla          None  1998-05-01 12:00:00        733  
1  pył zawieszony PM10          None  1998-05-01 12:00:00        733  

QuantumDamage

sensorId = 4727
r = requests.get('http://api.gios.gov.pl/pjp-api/rest/data/getData/' + str(sensorId))
concentration = json_normalize(r.json())
concentrationFrame = pd.DataFrame()
concentrationFrame["dates"] = [d[u'date'] for d in concentration["values"].values.item()]
concentrationFrame["values"] = [d[u'value'] for d in concentration["values"].values.item()]
concentrationFrame.set_index(["dates"], inplace=True)
concentrationFrame = concentrationFrame.iloc[::-1]

QuantumDamage

                       values
dates                        
2017-04-16 01:00:00  18.64540
2017-04-16 02:00:00   8.53258
2017-04-16 03:00:00   3.52958
2017-04-16 04:00:00   2.12867
2017-04-16 05:00:00   1.00000
2017-04-16 06:00:00  12.96610
2017-04-16 07:00:00  13.20580
2017-04-16 08:00:00   2.32258

QuantumDamage

oneHour = bigDataFrame['2015-06-12 08:00:00':'2015-06-12 08:00:00']
oneHour.index = oneHour.index.droplevel(level=0)
oneHour["coords"] = oneHour.index.map(lambda name: map(tuple,stations[stations["Kod stacji"] == name][[u'WGS84 φ N', u'WGS84 λ E']].values)[0])

QuantumDamage

pm10vals = oneHour[["PM10", "coords"]]
pm10vals.dropna(inplace=True)

QuantumDamage

oneHour[["lat", "long"]] = oneHour["coords"].apply(pd.Series)
oneHour.drop("coords", axis=1, inplace=True)