AQI.py

df_ind = df_sd.copy()
df_ind['Date'] = pd.to_datetime(df_ind['Date'],format='%Y-%m-%d')
df_ind['Period'] = df_ind.apply(lambda x: 'Before' if (x['Date'] < dt.datetime(2020, 3, 23)) else 'After',axis=1)
df_ind = df_ind.query('Date>="2020-01-01"')
df_ind = df_ind.groupby(['Period','StationId'],as_index=False)['AQI','PM2.5','PM10','O3','CO','SO2','NO2'].mean()
df_ind = df_ind.merge(df_st[['StationId','StationName','State','Latitude','Longitude']],how='inner',on='StationId')

def scale(aqiSeries):
    cmax = aqiSeries.max()
    cmin = aqiSeries.min()
    dt = 1e-5
    good = min((50-cmin)/(cmax-cmin)+dt,1.0)
    satisfactory = min((100-cmin)/(cmax-cmin)+dt,1.0)
    moderate = min((200-cmin)/(cmax-cmin)+dt,1.0)
    poor = min((300-cmin)/(cmax-cmin)+dt,1.0)
    very_poor = min((400-cmin)/(cmax-cmin)+dt,1.0)
    severe = min((500-cmin)/(cmax-cmin)+dt,1.0)

    colorcode = [good,satisfactory,moderate,poor,very_poor,severe]
    colorcode = [0.0 if c<0 else c for c in colorcode]
    colors = ['#77DD77','#33AF13','#F6D20E','#F17700','#FE6B64','#F12424']
    scl = []
    prev = 0
    for i in range(len(colorcode)):
        scl.extend([[prev,colors[i]],[colorcode[i],colors[i]]])
        prev=colorcode[i]
        if colorcode[i]==1.0: break
    if scl[-1][0]!=1.0:
        scl[-1][0]=1.0
    
    return scl