熊猫从字符串中删除字符

Question

我想简化我的 3 行代码，将方括号去掉为一行：

df = pd.DataFrame(dict(words=['[hello]',
                            '[hello,[Name, World, Max]',
                            np.nan, 
                            '[Goodbye]',
                            np.nan,
                            '[hello, [goodbye], hello]']))
df['words'].fillna("N/A", inplace=True)
df['words']=df['words'].str.extract("\[", "")
df['words']=df['words'].str.replace("\]", "")

我尝试了这里提到的几种方法，但都不起作用（替换、扩展）：remove characters from pandas column

Answer 1

我认为您需要 Series.str.replace 的值在 [] 中：

df['words'] = df['words'].fillna("N/A").str.replace("[\]\[]", "")

或者将| 用于正则表达式or：

df['words'] = df['words'].fillna("N/A").str.replace("\]|\[", "")

---

print (df)
                    words
0                   hello
1  hello,Name, World, Max
2                     N/A
3                 Goodbye
4                     N/A
5   hello, goodbye, hello

Answer 2

import numpy as np

df = pd.DataFrame(dict(words=['[hello]',
                            '[hello,[Name, World, Max]',
                            np.nan, 
                            '[Goodbye]',
                            np.nan,
                            '[hello, [goodbye], hello]']))
df ['words'] = df['words'].fillna("N/A", inplace=False).replace('\[{0,1}\]{0,1}', '', regex=True)