API Reference / preprocessing /

encoding/

Encoding transformers for categorical feature encoding.

This module provides various encoding techniques for converting between categorical (nominal/string) and numeric representations.

Available transformers:
  • OneHotEncoder: One column per category, indicator-coded.
  • OrdinalEncoder: Categories mapped to integer codes.
  • LabelEncoder: String labels mapped to a categorical index.
  • RareCategoryEncoder: Infrequent categories merged into a single bucket.
python 
>>> from tuiml.preprocessing.encoding import OneHotEncoder
>>> import numpy as np
python 
>>> # One-hot encode categorical features
>>> X = np.array([[0, 1], [1, 2], [2, 0]])  # Categorical indices
>>> encoder = OneHotEncoder(categories=[[0, 1, 2], [0, 1, 2]])
>>> X_encoded = encoder.fit_transform(X)

Modules

label

LabelEncoder transformer.

1 class · 0 functions
one_hot

OneHotEncoder transformer.

1 class · 0 functions
ordinal

OrdinalEncoder transformer.

1 class · 0 functions
rare_category

RareCategoryEncoder filter.

1 class · 0 functions