Python-pandas入门篇

https://pandas.pydata.org/
本篇介绍pandasd 入门应用，创建一个Series和DataFrame，以及简单的增删改查。

1.1、Series

导入工具包

import numpy as np
import pandas as pd 
from pandas import Series,DataFrame

简单创建 Series

# 简单创建
obj=Series([4,7,-5,3])

# 自己设置索引
obj2=Series([4,5,6,7],index=['a','b','c','d'])

--------------------------------------------------------------------------------------------------------------------------------
Out[2]:
0    4
1    7
2   -5
3    3
dtype: int64
--------------------------------------------------------------------------------------------------------------------------------
a    4
b    5
c    6
d    7
dtype: int64
--------------------------------------------------------------------------------------------------------------------------------

通过字典创建Series

sdata={'Ohio':35000,'Texas':7000,'Oregon':16000,'Utah':5000,'California':5300}
obj3=Series(sdata)
obj3
 --------------------------------------------------------------------------------------------------------------------------------
 California     5300
Ohio          35000
Oregon        16000
Texas          7000
Utah           5000
dtype: int64

 --------------------------------------------------------------------------------------------------------------------------------
# 转回字典
s1=obj3.to_dict()

 --------------------------------------------------------------------------------------------------------------------------------
# 设置索引，修改索引
states=['Ohio','Texas','Oregon','Utah','California']
obj4=Series(sdata,index=states)

obj4.index=['a','b','c','d','f']

查看值和索引

obj.values
-------------------------------------------------------------------
array([ 4,  7, -5,  3], dtype=int64)
-------------------------------------------------------------------

obj.index
-------------------------------------------------------------------
RangeIndex(start=0, stop=4, step=1)
-------------------------------------------------------------------

查找

• 通过索引值进行查找

    obj2[['a','c']]
    ---------------------------------------------------------------
    a    4
    c    6
    dtype: int64
    ---------------------------------------------------------------
    
    obj2[['a','c']]
    ---------------------------------------------------------------
    a    4
    b    5
    c    6
    dtype: int64
    ---------------------------------------------------------------

更改

• 通过查找到某个值，直接重新赋值

obj2['a']=9
obj2

a    9
b    5
c    6
d    7
dtype: int64

1.2、DataFrame

简单创建DataFrame

# 通过字典创建三列数据

data={'state':['Ohio','Texas','Oregon','Utah'],'year':[2000,2001,2002,2003],'pop':[1.5,1.6,0.8,1.7]}
df1=DataFrame(data)
df1
  ---------------------------------------------------------------
	state	year	pop
0	Ohio	2000	1.5
1	Texas	2001	1.6
2	Oregon	2002	0.8
3	Utah	2003	1.7
 ---------------------------------------------------------------
 # 重新排列顺序
 
df2=DataFrame(data,columns=['year','state','pop'])
----------------------------------------------------------------
	year	state	pop
0	2000	Ohio	1.5
1	2001	Texas	1.6
2	2002	Oregon	0.8
3	2003	Utah	1.7

字典创建DataFrame

pop={'Nevada':{2001:2.4,2002:3.1},'Ohio':{2000:1.2,2001:2.4,2002:1.9}}
df3=DataFrame(pop)
df3

---------------------------------------------------------------------
        Nevada	Ohio
2000	NaN	    1.2
2001	2.4	    2.4
2002	3.1	    1.9

添加索引跟列值的名称

df3.index.name='year'
df3.columns.name='state'
df3
--------------------------------------------
state	Nevada	   Ohio
year		
2000	NaN	        1.2
2001	2.4	        2.4
2002	3.1	        1.9

索引，字段名重命名

df1=DataFrame(np.arange(9).reshape(3,3),index=['bj','sh','gz'],columns=['c1','c2','c3'])
df1
----------------------------------------------------------------------------------------------------------------
	c1	c2	c3
bj	0	1	2
sh	3	4	5
gz	6	7	8
************************************************************************************************
# map
df1.index=Series(['北京','上海','广州'])

df1.index=df1.index.map(str.upper)

# rename
df1.rename(index={'BJ':'beijing'})

df1.rename(index=str.lower,columns=str.upper)


-----------------------------------------------------------------------------------------------------------------
C1	C2	C3
bj	0	1	2
sh	3	4	5
gz	6	7	8

 # 自定义一个函数进行改名
     
def test_map(x):
    return x+"_abc"
            
df1.index=df1.index.map(test_map)
df1.rename(index=test_map)

查找

# 查找某一列值
df2['year']
df2.year
--------------------------------------------------------
0    2000
1    2001
2    2002
3    2003
Name: year, dtype: int64
--------------------------------------------------------
# 查找某一个值
df2['year'][1]
df2.year[1]
-------------------------------------------------------
2001
-------------------------------------------------------

赋值

# 新增一列（debt）

df2['debt']=np.arange(4.)
df2['debt']=16.5
------------------------------------------------------------------

year	state	pop	debt
0	2000	Ohio	1.5	16.5
1	2001	Texas	1.6	16.5
2	2002	Oregon	0.8	16.5
3	2003	Utah	1.7	16.5

修改

# 指定一个值赋值

df2['pop'][0]=2
---------------------------------------------------------------------
# 根据索引批量修改（未指定的值默认为NaN）

val=Series([-1,-2,-3],index=[0,2,3])
df2['debt']=val
df2
---------------------------------------------------------------------
	year	state	pop	debt
0	2000	Ohio	2.0	-1.0
1	2001	Texas	1.6	NaN
2	2002	Oregon	0.8	-2.0
3	2003	Utah	1.7	-3.0

删除

# 判断是否为Ohio，并新增一列
df2['eastern']=df2.state=='Ohio'

-----------------------------------------------------------
year	state	pop	debt	eastern
0	2000	Ohio	2.0	-1.0	True
1	2001	Texas	1.6	NaN	False
2	2002	Oregon	0.8	-2.0	False
3	2003	Utah	1.7	-3.0	False
-----------------------------------------------------------
# 删除eastern

del df2['eastern']

----------------------------------------------------------
year	state	pop	debt
0	2000	Ohio	2.0	-1.0
1	2001	Texas	1.6	NaN
2	2002	Oregon	0.8	-2.0
3	2003	Utah	1.7	-3.0
----------------------------------------------------------

# 删除多列
df=a.drop(['R10D','R20D'],axis=1)