Encoding categorical variables
Preprocessing for Machine Learning in Python
James Chapman
Curriculum Manager, DataCamp
Categorical variables
user subscribed fav_color
0 1 y blue
1 2 n green
2 3 n orange
3 4 y green
Encoding binary variables - pandas
print(users["subscribed"])
0 y
1 n
2 n
3 y
Name: subscribed, dtype: object
print(users[["subscribed", "sub_enc"]])
subscribed sub_enc
0 y 1
1 n 0
2 n 0
3 y 1
users["sub_enc"] = users["subscribed"].apply(lambda val: 1 if val == "y" else 0)
Encoding binary variables - scikit-learn
from sklearn.preprocessing import LabelEncoderle = LabelEncoder() users["sub_enc_le"] = le.fit_transform(users["subscribed"])print(users[["subscribed", "sub_enc_le"]])
subscribed sub_enc_le
0 y 1
1 n 0
2 n 0
3 y 1
One-hot encoding
| fav_color |
|---|
| blue |
| green |
| orange |
| green |
Values: [blue, green, orange]
- blue: [1, 0, 0]
- green: [0, 1, 0]
- orange: [0, 0, 1]
| fav_color_enc |
|---|
| [1, 0, 0] |
| [0, 1, 0] |
| [0, 0, 1] |
| [0, 1, 0] |
print(users["fav_color"])
0 blue
1 green
2 orange
3 green
Name: fav_color, dtype: object
print(pd.get_dummies(users["fav_color"]))
blue green orange
0 1 0 0
1 0 1 0
2 0 0 1
3 0 1 0
Let's practice!
Preprocessing for Machine Learning in Python
