DataFrame.apply()

原文地址：https://github.com/wk2014/PandasLearnClass/blob/master/DataFrame.Apply.ipynb


from pandas import DataFrame, Series

import pandas as pd

import numpy as np

​

frame = DataFrame(np.random.randn(4,3),columns=list('bde'),index=['Utah','Ohio','Texas','Oregon'])

frame

	b	d	e
Utah	1.011072	-1.198988	-0.794338
Ohio	1.611431	0.149355	-1.316985
Texas	0.671086	-1.427144	-1.764187
Oregon	-0.496996	-0.279463	0.523360

 

#numpy 的 ufuncs(元素级数组方法) 可用于操作pandas 对象

np.abs(frame)

	b	d	e
Utah	1.011072	1.198988	0.794338
Ohio	1.611431	0.149355	1.316985
Texas	0.671086	1.427144	1.764187
Oregon	0.496996	0.279463	0.523360

 

np.round(frame,decimals=2)

	b	d	e
Utah	1.01	-1.20	-0.79
Ohio	1.61	0.15	-1.32
Texas	0.67	-1.43	-1.76
Oregon	-0.50	-0.28	0.52

 

#元素级python函数应用

format = lambda x: '%.2f' % x

frame.applymap(format)

	b	d	e
Utah	1.01	-1.20	-0.79
Ohio	1.61	0.15	-1.32
Texas	0.67	-1.43	-1.76
Oregon	-0.50	-0.28	0.52

format

frame['e'].map(format)

Utah      -0.79
Ohio      -1.32
Texas     -1.76
Oregon     0.52
Name: e, dtype: object

沿行广播，操作列

#将函数应用到由各行或列所形成的一维数组上。==>frame.apply(f)

#1.返回元素值

f = lambda x: x.max() - x.min()

#沿行广播，操作列

frame.apply(f)

b    2.108427
d    1.576499
e    2.287547
dtype: float64

 

#沿列广播，操作行对象

frame.apply(f,axis=1)

Utah      2.210060
Ohio      2.928416
Texas     2.435273
Oregon    1.020355
dtype: float64

 

#return 普通 Series 对象

def f2(x):

    return Series([x.min(),x.max()],index=['min','max'])

​

frame.apply(f2)

	b	d	e
min	-0.496996	-1.427144	-1.764187
max	1.611431	0.149355	0.523360

 

#return MultiIndex Series 对象

def f3(x):

    mIndex = pd.MultiIndex(levels=[['max','min'],['city','value']],labels=[[0,0,1,1],[0,1,0,1]])

    se= Series([x.argmax(),x.max(),x.argmin(),x.min()],index=mIndex)

    return se

​

frame.apply(f3)

		b	d	e
max	city	Ohio	Ohio	Oregon
	value	1.61143	0.149355	0.52336
min	city	Oregon	Texas	Texas
	value	-0.496996	-1.42714	-1.76419

 

#return MultiIndex Series 对象

def f3(x):

    mIndex = pd.MultiIndex(levels=[['max','min'],['city','value']],labels=[[0,0,1,1],[0,1,0,1]])

    se= Series([x.argmax(),x.max(),x.argmin(),x.min()],index=mIndex)

    return se

​

frame.apply(f3).unstack()

	b		d		e
	city	value	city	value	city	value
max	Ohio	1.61143	Ohio	0.149355	Oregon	0.52336
min	Oregon	-0.496996	Texas	-1.42714	Texas	-1.76419