pandas学习（二）

pandas学习二

import numpy as np
import pandas as pd

#定义初始数据
date=pd.date_range('20200408',periods=6)
df=pd.DataFrame(np.arange(24).reshape((6,4)),index=date,columns=['a','b','c','d'])

date

DatetimeIndex(['2020-04-08', '2020-04-09', '2020-04-10', '2020-04-11',
               '2020-04-12', '2020-04-13'],
              dtype='datetime64[ns]', freq='D')

df

	a	b	c	d
2020-04-08	0	1	2	3
2020-04-09	4	5	6	7
2020-04-10	8	9	10	11
2020-04-11	12	13	14	15
2020-04-12	16	17	18	19
2020-04-13	20	21	22	23

#输出某列数据
print(df['a'],df.a)

2020-04-08     0
2020-04-09     4
2020-04-10     8
2020-04-11    12
2020-04-12    16
2020-04-13    20
Freq: D, Name: a, dtype: int32 2020-04-08     0
2020-04-09     4
2020-04-10     8
2020-04-11    12
2020-04-12    16
2020-04-13    20
Freq: D, Name: a, dtype: int32

#输出某些行数据
df[0:3]

	a	b	c	d
2020-04-08	0	1	2	3
2020-04-09	4	5	6	7
2020-04-10	8	9	10	11

用loc进行纯标签名输出

#选取某行 
df.loc['20200408']

a    0
b    1
c    2
d    3
Name: 2020-04-08 00:00:00, dtype: int32

#输出ab列的所有数据
df.loc[:,['a','b']]

	a	b
2020-04-08	0	1
2020-04-09	4	5
2020-04-10	8	9
2020-04-11	12	13
2020-04-12	16	17
2020-04-13	20	21

df.loc['20200408',['a','b']]

a    0
b    1
Name: 2020-04-08 00:00:00, dtype: int32

用iloc进行纯索引输出

#输出第三行第一位的数据
df.iloc[3,1]

13

#输出多行多列数据
df.iloc[3:5,1:3]

	b	c
2020-04-11	13	14
2020-04-12	17	18

#指定具体索引输出
df.iloc[[1,3,5],[0,3]]

	a	d
2020-04-09	4	7
2020-04-11	12	15
2020-04-13	20	23

用ix进行标签名+索引混合输出 (ix以及被放弃了）

df.ix[:3,['a','c']]

D:\Anaconda3\lib\site-packages\ipykernel_launcher.py:1: DeprecationWarning: 
.ix is deprecated. Please use
.loc for label based indexing or
.iloc for positional indexing

See the documentation here:
http://pandas.pydata.org/pandas-docs/stable/indexing.html#ix-indexer-is-deprecated
  """Entry point for launching an IPython kernel.

	a	c
2020-04-08	0	2
2020-04-09	4	6
2020-04-10	8	10

数据筛选

#筛选出小于8的数据
df[df<8]

	a	b	c	d
2020-04-08	0.0	1.0	2.0	3.0
2020-04-09	4.0	5.0	6.0	7.0
2020-04-10	NaN	NaN	NaN	NaN
2020-04-11	NaN	NaN	NaN	NaN
2020-04-12	NaN	NaN	NaN	NaN
2020-04-13	NaN	NaN	NaN	NaN

df

	a	b	c	d
2020-04-08	0	0	0	0
2020-04-09	4	5	6	7
2020-04-10	8	9	10	11
2020-04-11	12	13	14	15
2020-04-12	16	17	18	19
2020-04-13	20	21	22	23

#筛选出a里面小于3的值，该值所在的行全部设置为0
df[df.a<3]=0
df

	a	b	c	d
2020-04-08	0	0	0	0
2020-04-09	4	5	6	7
2020-04-10	8	9	10	11
2020-04-11	12	13	14	15
2020-04-12	16	17	18	19
2020-04-13	20	21	22	23

#筛选出a里面小于5的值，该值所在的b列全部设置为1
df.b[df.a<5]=1
df

	a	b	c	d
2020-04-08	0	1	0	0
2020-04-09	4	1	6	7
2020-04-10	8	9	10	11
2020-04-11	12	13	14	15
2020-04-12	16	17	18	19
2020-04-13	20	21	22	23

#新加一列空值
df['e']=np.nan
df

	a	b	c	d	e
2020-04-08	0	1	0	0	NaN
2020-04-09	4	1	6	7	NaN
2020-04-10	8	9	10	11	NaN
2020-04-11	12	13	14	15	NaN
2020-04-12	16	17	18	19	NaN
2020-04-13	20	21	22	23	NaN