encoders#

Encoders for encoding categorical variables and scaling continuous data.

Functions

_clipped_log(y)

_clipped_logit(x)

_identity(x)

_minus_one(x)

_plus_one(x)

_square(y)

softplus_inv(y)

Classes

EncoderNormalizer([method, center, ...])

Special Normalizer that is fit on each encoding sequence.

Expm1Transform([cache_size])

GroupNormalizer([method, groups, center, ...])

Normalizer that scales by groups.

MinusOneTransform([cache_size])

Transform x -> x - 1.

MultiNormalizer(normalizers)

Normalizer for multiple targets.

NaNLabelEncoder([add_nan, warn])

Labelencoder that can optionally always encode nan and unknown classes (in transform) as class 0

ReLuTransform([cache_size])

Transform x -> max(0, x).

SoftplusTransform([cache_size])

Transform via the mapping \(\text{Softplus}(x) = \log(1 + \exp(x))\).

TorchNormalizer([method, center, ...])

Basic target transformer that can be fit also on torch tensors.

TransformMixIn()

Mixin for providing pre- and post-processing capabilities to encoders.