loc和iloc的用法和区别

loc——通过行标签索引行数据 
iloc——通过行号索引行数据 
ix——通过行标签或者行号索引行数据（基于loc和iloc 的混合） 

• 标签切片，如’a’:‘c’，与序列切片如0:2不同，后者不包含index=2的元素，前者包含结束标签’c’所在的行。

• 布尔类型数组作为标签，例如[True, False]等价于[‘a’,‘c’]

1.loc

import numpy as np
import pandas as pd
from pandas import *
from numpy import *

data=DataFrame(np.arange(16).reshape(4,4),index=list("ABCD"),columns=list("wxyz"))
print(data)
#    w   x   y   z
#A   0   1   2   3
#B   4   5   6   7
#C   8   9  10  11
#D  12  13  14  15

#loc
#行的选取
print(data.loc["A"])
print(type(data.loc["A"]))
#w    0
#x    1
#y    2
#z    3
#Name: A, dtype: int32
#<class 'pandas.core.series.Series'>

print(data.loc[["A"]])
print(type(data.loc[["A"]]))
#   w  x  y  z
#A  0  1  2  3
#<class 'pandas.core.frame.DataFrame'>
#综上，[]返回Series,[[]]返回DataFrame

print(data.loc["A","w"])
print(type(data.loc["A","w"]))
#0
#<class 'numpy.int32'>

print(data.loc[:,"w"])
print(type(data.loc[:,"w"]))
#A     0
#B     4
#C     8
#D    12
#Name: w, dtype: int32
#<class 'pandas.core.series.Series'>

print(data.loc["A":"C"])
print(type(data.loc["A":"C"]))
#   w  x   y   z
#A  0  1   2   3
#B  4  5   6   7
#C  8  9  10  11
#<class 'pandas.core.frame.DataFrame'>

print(data.loc["A":"C","w":"y"])
print(type(data.loc["A":"C","w":"y"]))
#   w  x   y
#A  0  1   2
#B  4  5   6
#C  8  9  10
#<class 'pandas.core.frame.DataFrame'>

print(data.loc[["A","C"],["w","y"]])
print(type(data.loc[["A","C"],["w","y"]]))
#   w   y
#A  0   2
#C  8  10
#<class 'pandas.core.frame.DataFrame'>

print(data.loc[:,["w","y"]])
print(type(data.loc[:,["w","y"]]))
#    w   y
#A   0   2
#B   4   6
#C   8  10
#D  12  14
#<class 'pandas.core.frame.DataFrame'>

#列的选取
print(data["w"])#等同于print(data.loc[:,"w"])
#A     0
#B     4
#C     8
#D    12
#Name: w, dtype: int32
print(data.loc[:,"w"])
#A     0
#B     4
#C     8
#D    12
#Name: w, dtype: int32
print(data["w"].equals(data.loc[:,"w"]))#True

#根据特殊条件选取行列
print(data["w"]>5)
#A    False
#B    False
#C     True
#D     True
#Name: w, dtype: bool

print(data.loc[data["w"]>5])
#    w   x   y   z
#C   8   9  10  11
#D  12  13  14  15
print(data.loc[data["w"]>5,"w"])
print(type(data.loc[data["w"]>5,"w"]))
#C     8
#D    12
#Name: w, dtype: int32
#<class 'pandas.core.series.Series'>
print(data.loc[data["w"]>5,["w"]])
print(type(data.loc[data["w"]>5,["w"]]))
#    w
#C   8
#D  12
#<class 'pandas.core.frame.DataFrame'>
print(data["w"]==0)
print(data.loc[lambda data:data["w"]==0])
print(type(data.loc[lambda data:data["w"]==0]))
#A     True
#B    False
#C    False
#D    False
#Name: w, dtype: bool
#   w  x  y  z
#A  0  1  2  3
#<class 'pandas.core.frame.DataFrame'>

#loc赋值
print(data)
#    w   x   y   z
#A   0   1   2   3
#B   4   5   6   7
#C   8   9  10  11
#D  12  13  14  15
data.loc[["A","C"],["w","x"]]=999
print(data)
#     w    x   y   z
#A  999  999   2   3
#B    4    5   6   7
#C  999  999  10  11
#D   12   13  14  15

2.iloc

data=DataFrame(np.arange(16).reshape(4,4),index=list("ABCD"),columns=list("wxyz"))
print(data)
#    w   x   y   z
#A   0   1   2   3
#B   4   5   6   7
#C   8   9  10  11
#D  12  13  14  15

print(data.iloc[0])
print(type(data.iloc[0]))
#w    0
#x    1
#y    2
#z    3
#Name: A, dtype: int32
#<class 'pandas.core.series.Series'>
#print(data.iloc["A"])报错

#print(data.loc[0])报错
print(data.loc[["A"]])
print(type(data.loc["A"]))
#   w  x  y  z
#A  0  1  2  3
#<class 'pandas.core.series.Series'>

3.iloc和loc差别

iloc是按照行数取值，而loc按着index名取值

data=DataFrame(np.arange(16).reshape(4,4),index=list("1234"),columns=list("wxyz"))
print(data)
#    w   x   y   z
#1   0   1   2   3
#2   4   5   6   7
#3   8   9  10  11
#4  12  13  14  15
print(data.iloc[0])
#w    0
#x    1
#y    2
#z    3
#Name: 1, dtype: int32
#print(data.loc[0])报错

参考：https://blog.csdn.net/boywaiter/article/details/86012620