Pandas入门之九：数据选择

已信任
Jupyter 服务器: 本地
Python 3: Not Started
[36]



import pandas as pd
import numpy as np
[37]



# .loc基于标签
# .iloc基于索引



[39]




df = pd.DataFrame(np.random.randn(8,4),index=['a','b','c','d','e','f','g','h'],columns=['A','B','C','D'])
df
A    B    C    D
a    -0.629391    -2.199736    0.547864    0.137235
b    0.346160    0.648808    0.290378    -0.598081
c    1.658532    1.114108    -1.595136    0.196397
d    -0.760715    -0.186863    1.043977    0.287979
e    0.220411    -1.352767    0.924732    -0.456502
f    -1.952481    1.228079    0.900851    0.597607
g    -1.683863    1.276186    -0.551237    -0.278381
h    0.083268    1.494524    0.162098    0.414894
[40]



# 选择A和B列所有内容
df.loc[:,['A','B']]
A    B
a    -0.629391    -2.199736
b    0.346160    0.648808
c    1.658532    1.114108
d    -0.760715    -0.186863
e    0.220411    -1.352767
f    -1.952481    1.228079
g    -1.683863    1.276186
h    0.083268    1.494524
[41]



# 选择标签选择a-h行
df.loc['a':'e']
A    B    C    D
a    -0.629391    -2.199736    0.547864    0.137235
b    0.346160    0.648808    0.290378    -0.598081
c    1.658532    1.114108    -1.595136    0.196397
d    -0.760715    -0.186863    1.043977    0.287979
e    0.220411    -1.352767    0.924732    -0.456502
[42]



# 选择任意行,任意列
df.loc['a':'e','B':]
B    C    D
a    -2.199736    0.547864    0.137235
b    0.648808    0.290378    -0.598081
c    1.114108    -1.595136    0.196397
d    -0.186863    1.043977    0.287979
e    -1.352767    0.924732    -0.456502
[43]



# 取出标签a中大于1的数据
df.loc['a']>1
A    False
B    False
C    False
D    False
Name: a, dtype: bool
[44]



df.loc[:,'A']>1
a    False
b    False
c     True
d    False
e    False
f    False
g    False
h    False
Name: A, dtype: bool
[49]



df.loc[:,df.loc['d']>1]
        C
a    0.547864
b    0.290378
c    -1.595136
d    1.043977
e    0.924732
f    0.900851
g    -0.551237
h    0.162098
[50]



# iloc基于位置的索引，第1行数据
df.iloc[0]
A   -0.629391
B   -2.199736
C    0.547864
D    0.137235
Name: a, dtype: float64
[51]



df.iloc[1]
A    0.346160
B    0.648808
C    0.290378
D   -0.598081
Name: b, dtype: float64
[55]



df.iloc[0:3,1:3]
B    C
a    -2.199736    0.547864
b    0.648808    0.290378
c    1.114108    -1.595136
[56]



# 取任意行和任意列
df.iloc[0:3,[1,3]]
          B           D
a    -2.199736    0.137235
b    0.648808    -0.598081
c    1.114108    0.196397
[58]



# 快速获取数据
df.A
a   -0.629391
b    0.346160
c    1.658532
d   -0.760715
e    0.220411
f   -1.952481
g   -1.683863
h    0.083268
Name: A, dtype: float64
[-]