使用Pandas中的fillna（）方法替换列中的特定字符串值

Question

I have a particular column(the column is called 'numbers') which outputs unique values like this:

df.numbers.unique()

Output:

([nan, '50', '22', '11', '46', '58', '22', '14', '18', '15', '33', 'XX'], dtype=object)

As seen above there are unidentified characters such as 'XX'. I wish to convert them to 'NaN' values instead. I tried replacing these with using this code:

 df.numbers.replace('XX',np.NaN)

However, when I print the unique values in the column again. The 'XX' is still there. I wish to get rid of the 'XX' and replace them with 'NaN' instead. I am just curious to know why it is not working. Assistance would be appreciated. Thanks in advance!

Answer 1

我想你只是忘了把它分配回来

df.numbers=df.numbers.replace('XX',np.NaN)