时间分组

import seaborn as sns
import pandas as pd
import numpy as np
import matplotlib
import matplotlib.pyplot as plt
from scipy import stats
from datetime import datetime

#load data
sh = pd.read_csv('all_out.csv')
ccn_data = pd.DataFrame(
     sh
#     columns = ['DATE','CCN_1.0_cm3','SS_1.0_mg','Wb_1.0', 'Hcb1.0', 'Tcb1.0', 'Ptop1.0','Ttop1.0', 'Htop1.0','Cloud_fra', 'Retop1.0', 'Recb1.0', 'T14_1.0', 'H14_1.0','PM25','PM10','rain','rain820','rain208']
     #columns = ['DATE','CCN_1.0_cm3','SS_1.0_mg','Wb_1.0','Ttop1.0', 'Htop1.0','Cloud_fra','PM25','PM10','rain820']
#     columns = ['CCN_1.0_cm3','SS_1.0_mg','Cloud_fra','PM25','rain820']
     )
ccn_data['DATE'] = pd.to_datetime(ccn_data['DATE'])
ccn_data['DATE'] = [datetime.strftime(x,'%Y') for x in ccn_data['DATE']]
#print(type(ccn_data['DATE']))
#ccn_data["DATE"] = ccn_data["DATE"].astype(str)
ccn_data["DATE"]=ccn_data["DATE"].replace(['2013','2014','2015','2016','2017','2018'],['2013Y','2014Y','2015Y','2016Y','2017Y','2018Y'])
#print(ccn_data.head())
#data.drop(['lons','lats'], axis=1,inplace=True)
#ccn_data=ccn_data.dropna()
#ccn_data=ccn_data[ccn_data["CCN_1.0_cm3"]<10000]
#ccn_data["CCN_1.0_cm3"]=ccn_data["CCN_1.0_cm3"]**(1.0/3.0)
#ccn_data=ccn_data[ccn_data["CCN_1.0_cm3"]>2]
#ccn_data=ccn_data[ccn_data["PM25"]>0]
#ccn_data=ccn_data[ccn_data["PM25"]<200]
print(ccn_data.head())
cols = [col for col in ccn_data.columns]
print(cols)
ccn_mean = ccn_data.groupby('BJC_ccn')[cols].mean()
ccn_mean.to_csv("meantime.csv")