read_dataset.py

import os
import pandas as pd

basedir = './'
dirs = ['1.Raw(PM)','2.Raw(PM+Met)','3.Clean(PM+Met)','4.Grid(PM+Met)']
months = ['Nov2020', 'Dec2020', 'Jan2021']
months_fmt = {'Nov2020':'2020-11-{:02d}_all', 'Dec2020':'2020-12-{:02d}_all', 'Jan2021':'2021-01-{:02d}_all'}

def read(path, D):
    if os.path.exists(path):
        d = pd.read_csv(path)
        D = pd.concat((D,d))
    return D

for dir in dirs:
    D = None
    for month in months:
        if 'Raw' not in dir:
            D = read(os.path.join(basedir, dir, month+'.csv.gz'), D)
        else:
            for file in range(1,32):
                D = read(os.path.join(basedir, dir, month, months_fmt[month].format(file) + '.csv.gz'), D)
    print('{}: Shape {}, Mean PM2.5 {}'.format(dir, D.shape, D.pm2_5.mean()))

print('Done')