Representations

Every built-in transformer exported from pretab.transformers. These are the standalone, scikit-learn compatible representations. For a capability-oriented view, see the comparison table.

Spline expansions

`BSplineTransformer`	Transform numerical features using a B-spline basis expansion.
`MSplineTransformer`	Transform numerical features using an M-spline basis expansion.
`ISplineTransformer`	Transform numerical features using an I-spline (integrated spline) basis.
`CubicRegressionSplineTransformer`	Cubic Spline Transformer for one-dimensional or multi-dimensional input features.
`NaturalCubicSplineTransformer`	Natural Cubic Spline Transformer for continuous features.
`PSplineTransformer`	P-spline Transformer for smooth spline basis expansion with penalization.

Canonical import: pretab.expansion.spline.

Multivariate splines

`TensorProductSplineTransformer`	Tensor Product Spline Transformer for multivariate smooth basis expansion.
`ThinPlateSplineTransformer`	Multivariate low-rank thin-plate regression spline basis.

Canonical import: pretab.expansion.spline.multivariate.

Functional expansions

`RBFExpansionTransformer`	Radial Basis Function (RBF) feature expansion for numerical tabular data.
`ReLUExpansionTransformer`	Applies ReLU basis expansion to input features using fixed or data-driven center placement.
`SigmoidExpansionTransformer`	Applies sigmoid basis expansion to input features using specified or data-driven center placement.
`TanhExpansionTransformer`	Applies hyperbolic tangent (tanh) basis expansion to input features using specified or learned center locations.
`FourierFeatureTransformer`	Deterministic Fourier (sine/cosine) feature expansion for numerical data.

Canonical import: pretab.expansion.functional.

Kernel approximation

`RandomFourierFeaturesTransformer`	Random Fourier features approximating an RBF kernel map (multivariate).
`NystroemFeaturesTransformer`	Nystroem kernel-map approximation over the full feature block (multivariate).

Canonical import: pretab.kernel_approximation.

Numerical encoding

`NumericBinningTransformer`	Stateful binning transformer for numerical features.
`PLETransformer`	Piecewise Linear Encoding (PLE) transformer for numerical features.
`PeriodicEncodingTransformer`	Encode a cyclical variable using sine and cosine harmonics.

Canonical import: pretab.encoding.numerical.

Categorical encoding

`ContinuousOrdinalTransformer`	Encode categorical features as continuous integer values.
`OneHotFromOrdinalTransformer`	Convert ordinal-encoded features into a one-hot encoded representation.

Canonical import: pretab.encoding.categorical.

Embeddings

LanguageEmbeddingTransformer

Encode categorical text features into embeddings using a pre-trained language model.

Canonical import: pretab.embedding.

Preprocessing utilities

`MissingStateIndicator`	Emit a binary `__missing` column marking where the input was missing.
`NoTransformer`	Pass-through transformer that returns the input unchanged.
`ToFloatTransformer`	Convert input data to floating-point type.

Canonical import: pretab.preprocessing.