Python – How pandas uses groupby to replace NaN values with averages

I tried using it to

replace the NaN value in the column feature count (it is an integer ranging from 1 to 10), using groupby(client_id or client_name),
However, the NaN value does not seem to disappear.

df['feature_count'].isnull().sum()

The output is:

2254

Now I use:

df['feature_count'].fillna(df.groupby('client_name')['feature_count'].mean(), inplace=True)

But the output remains the same:

df['feature_count'].isnull().sum()

2254

Is there another way to replace NaN values with other non-NaN values for columns grouped by ID?