pandas 对日期进行的处理

import pandas as pd 
import os
os.chdir(r"C:\Users\Hans\Desktop\data_analysis\test_data")
df = pd.read_csv("Beijing_2014-02.csv")
df.head()

	Date	Temperature(Celsius)(avg)	Dew Point(Celsius)(avg)	Humidity(%)(avg)	Sea Level Press(hPa's)(avg)	Precipitation(mm)(high)	Precipitation(mm)(sum)
0	2014/1/1	4	-17	19	1015	47	0.0
1	2014/1/2	0	-9	50	1018	-	0.0
2	2014/1/3	3	-13	32	1022	-	0.0
3	2014/1/4	-2	-7	68	1023	-	0.0
4	2014/1/5	0	-11	51	1025	-	0.0

df.set_index(pd.to_datetime(df['Date']),inplace = True)
df.head()

	Date	Temperature(Celsius)(avg)	Dew Point(Celsius)(avg)	Humidity(%)(avg)	Sea Level Press(hPa's)(avg)	Precipitation(mm)(high)	Precipitation(mm)(sum)
Date
2014-01-01	2014/1/1	4	-17	19	1015	47	0.0
2014-01-02	2014/1/2	0	-9	50	1018	-	0.0
2014-01-03	2014/1/3	3	-13	32	1022	-	0.0
2014-01-04	2014/1/4	-2	-7	68	1023	-	0.0
2014-01-05	2014/1/5	0	-11	51	1025	-	0.0

#时间序列格式，可以按照前缀进行取值
df.loc['2014-03'].head()
#也可以用df.loc['2014']

	Date	Temperature(Celsius)(avg)	Dew Point(Celsius)(avg)	Humidity(%)(avg)	Sea Level Press(hPa's)(avg)	Precipitation(mm)(high)	Precipitation(mm)(sum)
Date
2014-03-01	2014/3/1	4	-15	31	1023	55	0.0
2014-03-02	2014/3/2	4	-8	46	1022	-	0.0
2014-03-03	2014/3/3	0	-6	60	1022	-	0.0
2014-03-04	2014/3/4	4	-19	23	1026	50	0.0
2014-03-05	2014/3/5	1	-21	18	1028	35	0.0

#可以按照区间取值
df.loc['2014-03-01':'2014-03-02']

	Date	Temperature(Celsius)(avg)	Dew Point(Celsius)(avg)	Humidity(%)(avg)	Sea Level Press(hPa's)(avg)	Precipitation(mm)(high)	Precipitation(mm)(sum)
Date
2014-03-01	2014/3/1	4	-15	31	1023	55	0.0
2014-03-02	2014/3/2	4	-8	46	1022	-	0.0

#获取周、月、季度
df.index.week

Int64Index([ 1,  1,  1,  1,  1,  2,  2,  2,  2,  2,
            ...
            52, 52, 52, 52, 52, 52, 52,  1,  1,  1],
           dtype='int64', name='Date', length=365)

df.index.month

Int64Index([ 1,  1,  1,  1,  1,  1,  1,  1,  1,  1,
            ...
            12, 12, 12, 12, 12, 12, 12, 12, 12, 12],
           dtype='int64', name='Date', length=365)

df.index.quarter

Int64Index([1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
            ...
            4, 4, 4, 4, 4, 4, 4, 4, 4, 4],
           dtype='int64', name='Date', length=365)

#统计每周的数据
df.groupby(df.index.week)['Humidity(%)(avg)'].max().head(10)
#前10周，每周的最高湿度

Date
1     68
2     72
3     61
4     50
5     67
6     76
7     65
8     73
9     66
10    60
Name: Humidity(%)(avg), dtype: int64

#按照每月进行统计
df.groupby(df.index.month)['Humidity(%)(avg)'].max()

Date
1     72
2     76
3     60
4     66
5     85
6     80
7     85
8     86
9     93
10    92
11    88
12    64
Name: Humidity(%)(avg), dtype: int64

#按照每个季度进行统计
df.groupby(df.index.quarter)['Humidity(%)(avg)'].max()

Date
1    76
2    85
3    93
4    92
Name: Humidity(%)(avg), dtype: int64

%matplotlib inline
df.groupby(df.index.quarter)['Humidity(%)(avg)'].max().plot()

<matplotlib.axes._subplots.AxesSubplot at 0x117acb4b3c8>

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-0Ks2DJcg-1601386291153)(output_10_1.png)]

df.groupby(df.index.month)['Humidity(%)(avg)'].max().plot()

<matplotlib.axes._subplots.AxesSubplot at 0x117acc580f0>

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-S9wTY919-1601386291157)(output_11_1.png)]