how to use one hot encoding in multiple columns at once

1categorical_cols = ['a', 'b', 'c', 'd'] 
2
3from sklearn.preprocessing import LabelEncoder
4# instantiate labelencoder object
5le = LabelEncoder()
6
7# apply le on categorical feature columns
8data[categorical_cols] = data[categorical_cols].apply(lambda col: le.fit_transform(col))    
9from sklearn.preprocessing import OneHotEncoder
10ohe = OneHotEncoder()
11
12#One-hot-encode the categorical columns.
13#Unfortunately outputs an array instead of dataframe.
14array_hot_encoded = ohe.fit_transform(data[categorical_cols])
15
16#Convert it to df
17data_hot_encoded = pd.DataFrame(array_hot_encoded, index=data.index)
18
19#Extract only the columns that didnt need to be encoded
20data_other_cols = data.drop(columns=categorical_cols)
21
22#Concatenate the two dataframes : 
23data_out = pd.concat([data_hot_encoded, data_other_cols], axis=1)
24