时间戳

首页 > 解决方案 > 根据索引值在“A”列中插入 Null

python - 根据索引值在“A”列中插入 Null

问题描述

我有一个索引号为“l1”的列表，我想根据这些索引号将“A”列中的值替换为 NaN

当前数据 

Index                A
0        Reviewer: Newbie | 35-44 on Treatment for 1 
1        Reviewer: 45-54 on Treatment for less than 1 
2        Reviewer: Ocetech| 65-74 Male on Treatment 
3        Reviewer: virleo| 55-64 Female on Treatment 
4        Reviewer: Diane perrin| 65-74 on Treatment for

l1 = [ 1 , 3, 4]

预期产出

Index                A
0        Reviewer: Newbie | 35-44 on Treatment for 1 
1        NaN 
2        Reviewer: Ocetech| 65-74 Male on Treatment 
3        NaN
4        NaN

标签： pythonpandas

解决方案

只是在复制你的 DataFrame 之后的细节..

导入 numpy 以将Nan值设置为所需的索引。

import pandas as pd
import numpy as np

你的数据框：

$ df
                                                A
0     Reviewer: Newbie | 35-44 on Treatment for 1
1    Reviewer: 45-54 on Treatment for less than 1
2      Reviewer: Ocetech| 65-74 Male on Treatment
3     Reviewer: virleo| 55-64 Female on Treatment
4  Reviewer: Diane perrin| 65-74 on Treatment for

您的索引列表：

$ l1
[1, 3, 4]

基于loc对于A将索引设置为Nan 使用 Numpy 的列..

df.loc[l1,'A'] = np.nan

结果：

print(df)
                                             A
0  Reviewer: Newbie | 35-44 on Treatment for 1
1                                          NaN
2   Reviewer: Ocetech| 65-74 Male on Treatment
3                                          NaN
4                                          NaN

如果您没有要替换为的长索引列表，请注意NaN，您可以直接指定它们，而不是传递一个列表索引。

$ df.loc[[1,3,4],'A'] = np.nan
$ print(df)
                                             A
0  Reviewer: Newbie | 35-44 on Treatment for 1
1                                          NaN
2   Reviewer: Ocetech| 65-74 Male on Treatment
3                                          NaN
4                                          NaN

另一种方法：

$ df.rename(index={1:np.nan, 3:np.nan, 4:np.nan}, inplace=True)

推荐阅读