sportslabkit.vector_model.sklearn

Overview

Classes

SklearnVectorModel

A specialized subclass of BaseVectorModel for scikit-learn pipelines.

Classes

class sportslabkit.vector_model.sklearn.SklearnVectorModel(model_path: str = '', input_vector_size: int | None = None, output_vector_size: int | None = None)

Bases: sportslabkit.vector_model.base.BaseVectorModel

A specialized subclass of BaseVectorModel for scikit-learn pipelines.

This class is designed to facilitate the use of scikit-learn pipelines as vector-based models within the SportsLabKit ecosystem. It overrides the abstract methods from BaseVectorModel to provide implementations tailored for scikit-learn pipelines.

model

The loaded scikit-learn pipeline model. None if the model is not loaded.

Type:

Pipeline | None

Overview

Methods

forward(inputs, **kwargs)

Implement the forward pass specific to scikit-learn pipelines.

Members

forward(inputs: sportslabkit.types.Vector, **kwargs: Any) → sportslabkit.types.Vector

Implement the forward pass specific to scikit-learn pipelines.

This method takes a vector input and passes it through the scikit-learn pipeline’s predict method. Additional keyword arguments can be passed to the predict method via **kwargs.

Parameters:

• inputs (Vector) – The input vector, which should match the expected input shape of the pipeline.

• **kwargs (Any) – Additional keyword arguments to pass to the pipeline’s predict method.

Returns:

The output vector from the pipeline’s predict method.

Return type:

Vector

Raises:

ValueError – If the model attribute is None, indicating that the model has not been loaded.