kumoai.encoder

While Kumo automatically infers encoders based on each column’s dtype and stype, you can override the encoder for individual columns via ColumnProcessingPlan. The encoder you specify must be compatible with the column’s semantic type.

Enums

NAStrategy

Strategy for imputing missing values.

Scaler

Scaling strategy for numerical features.

Encoders

Null

Skips encoding for the column entirely. Supported on all semantic types.

1
from kumoai.encoder import Null
2
3
encoder = Null()

Numerical

Encodes a numerical column with optional scaling and missing value imputation.

1
from kumoai.encoder import Numerical, Scaler, NAStrategy
2
3
encoder = Numerical(scaler=Scaler.STANDARD, na_strategy=NAStrategy.MEAN)

MaxLogNumerical

Applies a log transformation after clipping values at the column maximum. Useful for heavy-tailed distributions.

1
from kumoai.encoder import MaxLogNumerical, NAStrategy
2
3
encoder = MaxLogNumerical(na_strategy=NAStrategy.MEAN)

MinLogNumerical

Applies a log transformation after clipping values at the column minimum.

1
from kumoai.encoder import MinLogNumerical, NAStrategy
2
3
encoder = MinLogNumerical(na_strategy=NAStrategy.MEAN)

Index

Encodes a categorical or ID column as a learned embedding index. Rare values (below min_occ) are mapped to a shared out-of-vocabulary embedding.

1
from kumoai.encoder import Index, NAStrategy
2
3
encoder = Index(min_occ=2, na_strategy=NAStrategy.SEPARATE)

Hash

Encodes a categorical column via feature hashing. Suitable for very high-cardinality columns.

1
from kumoai.encoder import Hash
2
3
encoder = Hash(num_components=128)

MultiCategorical

Encodes a multi-categorical column (a string containing multiple space- or comma-separated categories) as a bag-of-categories embedding.

1
from kumoai.encoder import MultiCategorical
2
3
encoder = MultiCategorical(min_occ=1)

GloVe

Encodes a text column using pre-trained GloVe word embeddings.

1
from kumoai.encoder import GloVe
2
3
encoder = GloVe(model_name="glove.6B", embedding_dim=50)

NumericalList

Encodes a sequence or embedding column (a list of floats) with optional scaling.

1
from kumoai.encoder import NumericalList
2
3
encoder = NumericalList()

Datetime

Encodes a timestamp column by decomposing it into cyclical calendar features. Each component is optional and enabled by default.

1
from kumoai.encoder import Datetime
2
3
encoder = Datetime(include_year=True, include_day_of_week=True)