pandas选取特定索引的行

[python] view plain copy print?



>>> import numpy as np
>>> import pandas as pd
>>> index=np.array([2,4,6,8,10])
>>> data=np.array([3,5,7,9,11])
>>> data=pd.DataFrame({’num’:data},index=index)
>>> print(data)
num
2 3
4 5
6 7
8 9
10 11
>>> select_index=index[index>5]
>>> print(select_index)
[ 6 8 10]
>>> data[’num’].loc[select_index]
6<
4000
/span> 7
8 9
10 11
Name: num, dtype: int32
>>>

>>> import numpy as np
>>> import pandas as pd
>>> index=np.array([2,4,6,8,10])
>>> data=np.array([3,5,7,9,11])
>>> data=pd.DataFrame({'num':data},index=index)
>>> print(data)
num
2     3
4     5
6     7
8     9
10   11
>>> select_index=index[index>5]
>>> print(select_index)
[ 6  8 10]
>>> data['num'].loc[select_index]
6      7
8      9
10    11
Name: num, dtype: int32
>>>

注意，不能用iloc，iloc是将序列当作数组来访问，下标又会从0开始：

[python] view plain copy print?



>>> data[‘num’].iloc[2:5]
6 7
8 9
10 11
Name: num, dtype: int32
>>> data[’num’].iloc[[2,3,4]]
6 7
8 9
10 11
Name: num, dtype: int32
>>>

>>> data['num'].iloc[2:5]
6      7
8      9
10    11
Name: num, dtype: int32
>>> data['num'].iloc[[2,3,4]]
6      7
8      9
10    11
Name: num, dtype: int32
>>>

转自博客：

http://blog.csdn.net/o1101574955/article/details/51638128