如何用 pandas 中的字符串值替换 NaN

Question

I am working on the Kaggle Housing Prices project. I have tried for several hours to replace the NaN values in a string column 'BsmtQual'. For all houses with a SalePrice of less than 120000 that have a NaN value in the BsmtQual column, I want to replace it with 'Fa'.

df is my dataframe.

Fa_rng = df['SalePrice'] < 120000

I have attempted all the below and they did not change anything.

df.loc[Fa_rng,'BsmtQual'].fillna('Fa',inplace=True)
df.loc[Fa_rng,'BsmtQual'].replace('NaN','Fa',inplace=True)
df.loc[Fa_rng,'BsmtQual'].str.replace('NaN','Fa')

This one gets a warning saying "A value is trying to be set on a copy of a slice from a DataFrame" and does nothing.

df.loc[(df['BsmtQual'].isna()) & (df['SalePrice'] < 120000)].fillna('Fa',inplace=True)

How do I replace a NaN value in pandas for those houses to a 'Fa' string?

Answer 1

Don't use inplace :

df.loc[Fa_rng & df['BsmtQual'].isna(),'BsmtQual'] = 'Fa'