why i got error when using SimpleImputer for impute Nan values?

Question

I have the following code, where sp_col is a sliced column of my dataframe df_1:

from sklearn.impute import SimpleImputer
import numpy as np

imputer = SimpleImputer(missing_values=np.NaN,strategy='mean')
sp_col =df_1.iloc[:-1,9:28].values.reshape(-1,1)

# print(sp_col.isnull().sum())

for items,j in sp_col.items():    
    im2 = sp_col
    im2 = imputer.fit(im2)
    im3 = imputer.transform(sp_col) 

I then get the following error:

AttributeError                            Traceback (most recent call last)
<ipython-input-207-28b34f875d63> in <module>
      8 # print(sp_col.isnull().sum())
      9 
---> 10 for items,j in sp_col.items():
     11         im2 = sp_col
     12         im2 = imputer.fit(im2)

AttributeError: 'numpy.ndarray' object has no attribute 'items'

   

Answer 1

Try this:

from sklearn.impute import SimpleImputer
import numpy as np

imputer = SimpleImputer(missing_values=np.NaN,strategy='mean')

df_1.iloc[:, 9:28] = si.fit_transform(df_1.iloc[:, 9:28])

Answer 2

As the error says, you are trying to use the .items() attribute which numpy arrays do not have. .items() is a method for pandas series, but since you converted it to a numpy array that doesn't work anymore. Either keep the data as a pandas series or change your loop to for value in sp_col:.