Imputing categorical values

Dealing with Missing Data in Python

Suraj Donthi

Deep Learning & Computer Vision Consultant

Complexity with categorical values

Most categorical values are strings
Cannot perform operations on strings
Necessity to convert/encode strings to numeric values and impute

Conversion techniques

One-hot encoder

Color	Color_Red	Color_Green	Color_Blue
Red	1	0	0
Green	0	1	0
Blue	0	0	1
Red	1	0	0
Blue	0	0	1
Blue	0	0	1

Ordinal encoder

Color	Value
Red	0
Green	1
Blue	2
Red	0
Blue	2
Blue	2

Imputation techniques

Fill with most frequent category
Impute using statistical models like KNN

Users profile data

users = pd.read_csv('userprofile.csv')
users.head()

    smoker    drink_level    dress_preference ambience    hijos        activity     budget
0    False    abstemious        informal      family    independent    student      medium
1    False    abstemious        informal      family    independent    student      low
2    False    social drinker    formal        family    independent    student      low
3    False    abstemious        informal      family    independent    professional medium
4    False    abstemious     no preference    family    independent    student      medium

Ordinal Encoding

from sklearn.preprocessing import OrdinalEncoder


# Create Ordinal Encoder
ambience_ord_enc = OrdinalEncoder()

# Select non-null values in ambience
ambience = users['ambience']
ambience_not_null = ambience[ambience.notnull()]
reshaped_vals = ambience_not_null.values.reshape(-1, 1)

# Encode the non-null values of ambience
encoded_vals = ambience_ord_enc.fit_transform(reshaped_vals)

# Replace the ambience column with ordinal values
users.loc[ambience.notnull(), 'ambience'] = np.squeeze(encoded_vals)

Ordinal Encoding

# Create dictionary for Ordinal encoders
ordinal_enc_dict = {}

# Loop over columns to encode
for col_name in users:
    # Create ordinal encoder for the column
    ordinal_enc_dict[col_name] = OrdinalEncoder()
    col = users[col_name]

    # Select the non-null values in the column
    col_not_null = col[col.notnull()]
    reshaped_vals = col_not_null.values.reshape(-1, 1)

    # Encode the non-null values of the column
    encoded_vals = ordinal_enc_dict[col_name].fit_transform(reshaped_vals)

    # Replace the values in the column with ordinal values
    users.loc[col.notnull(), col_name] = np.squeeze(encoded_vals)

Imputing with KNN

users_KNN_imputed = users.copy(deep=True)


# Create KNN imputer
KNN_imputer = KNN()


users_KNN_imputed.iloc[:, :] = np.round(KNN_imputer.fit_transform(users))


for col_name in users_KNN_imputed:
    # Reshape the values to 2-dimensions to 
    # avoid errors while storing in the DataFrame
    reshaped = users_KNN_imputed[col_name].values.reshape(-1, 1)
    users_KNN_imputed[col_name] = \
    ordinal_enc_dict[col_name].inverse_transform(reshaped)

Summary

Steps to impute categorical values

Convert non-missing categorical columns to ordinal values
Impute the missing values in the ordinal DataFrame
Convert back from ordinal values to categorical values

Let's practice!

Dealing with Missing Data in Python