python - 用np.nan替换问号(Replacing question marks with np.nan)

Question

I am trying to replace question marks in my data set with np.nan:

(我试图用np.nan替换数据集中的问号：)

I tried using the following code:

(我尝试使用以下代码：)

df['Workclass'] = [row if row!='?' else np.nan for row in df['Workclass']]

And this:

(和这个：)

df['Workclass'] = df['Workclass'].map(lambda x: np.nan if x=="?"  else x)

And this:

(和这个：)

df['Workclass'] = df['Workclass'].replace(to_replace =['?'], value = np.nan, regex = True)

But none of these solutions seem to change the frequency of question marks in the column.

(但是这些解决方案似乎都不能改变列中问号的频率。)

  ask by Murtaza Kamal translate from so

Answer 1

Try this:

(尝试这个：)

df['Workclass'].apply(lambda x: np.nan if x == '?' else x) 

and if that works:

(如果可行：)

df['Workclass'] = df['Workclass'].apply(lambda x: np.nan if x == '?' else x) 

if your looking for '?'

(如果您要查找“？”)

anywhere in a string you can use this:

(您可以在字符串中的任何位置使用此命令：)

df['Workclass'].apply(lambda x: np.nan if str(x).find('?')>-1 else x)