plaid.pipelines.sklearn_block_wrappers
======================================

.. py:module:: plaid.pipelines.sklearn_block_wrappers

.. autoapi-nested-parse::

   Wrapped scikit-learn transformers and regressors for PLAID Dataset compatibility.

   Provides adapters to use scikit-learn estimators within the PLAID feature/block system:

   - WrappedPlaidSklearnTransformer: wraps a TransformerMixin

   - WrappedPlaidSklearnRegressor: wraps a RegressorMixin



Attributes
----------

.. autoapisummary::

   plaid.pipelines.sklearn_block_wrappers.Self


Classes
-------

.. autoapisummary::

   plaid.pipelines.sklearn_block_wrappers.WrappedSklearnTransformer
   plaid.pipelines.sklearn_block_wrappers.WrappedSklearnRegressor


Functions
---------

.. autoapisummary::

   plaid.pipelines.sklearn_block_wrappers.get_2Darray_from_homogeneous_identifiers


Module Contents
---------------

.. py:data:: Self

.. py:function:: get_2Darray_from_homogeneous_identifiers(dataset: plaid.Dataset, features_identifiers: list[plaid.containers.FeatureIdentifier]) -> plaid.types.Array

   Returns a 2D array from a Dataset and a feature id.

   The function calls `dataset.get_tabular_from_homogeneous_identifiers(...)`, then removes
   either the second or third dimension if it has size 1, so that the output is 2D.

   :param dataset: A Dataset object exposing `get_tabular_from_homogeneous_identifiers`.
   :type dataset: Dataset
   :param features_identifiers: a list of input feature identifiers.
   :type features_identifiers: list[FeatureIdentifier]

   :returns: A NumPy array of shape (n_samples, n_features).

   :raises AssertionError: If the number of features in the output does not match the identifiers.
   :raises ValueError: If both the second and third dimensions have size greater than 1.


.. py:class:: WrappedSklearnTransformer(sklearn_block: plaid.types.SklearnBlock, in_features_identifiers: list[plaid.containers.FeatureIdentifier], out_features_identifiers: Optional[list[plaid.containers.FeatureIdentifier]] = None)

   Bases: :py:obj:`sklearn.base.TransformerMixin`, :py:obj:`sklearn.base.BaseEstimator`


   Adapter for using a scikit-learn transformer on PLAID Datasets.

   Transforms tabular data extracted from homogeneous feature identifiers,
   and returns results as a `Dataset`. Supports forward and inverse transforms.

   :param sklearn_block: A scikit-learn Transformer implementing fit/transform APIs.
   :type sklearn_block: SklearnBlock
   :param in_features_identifiers: List of feature identifiers to extract input data from.
   :type in_features_identifiers: list[FeatureIdentifier]
   :param out_features_identifiers: List of feature identifiers used for outputs. If None,
                                    defaults to `in_features_identifiers`.
   :type out_features_identifiers: list[FeatureIdentifier], optional


   .. py:attribute:: sklearn_block


   .. py:attribute:: in_features_identifiers


   .. py:attribute:: out_features_identifiers
      :value: None



   .. py:method:: fit(dataset: plaid.Dataset, _y=None) -> Self

      Fits the underlying scikit-learn transformer on selected input features.

      :param dataset: A `Dataset` object containing the features to transform.
      :param _y: Ignored.

      :returns: The fitted transformer.
      :rtype: self



   .. py:method:: transform(dataset: plaid.Dataset) -> plaid.Dataset

      Applies the fitted transformer to the selected input features.

      :param dataset: A `Dataset` object to transform.

      :returns: Transformed features wrapped as a new `Dataset`.
      :rtype: Dataset



   .. py:method:: inverse_transform(dataset: plaid.Dataset) -> plaid.Dataset

      Applies inverse transformation to the output features.

      :param dataset: A `Dataset` object with transformed output features.

      :returns: Dataset with inverse-transformed features.
      :rtype: Dataset



.. py:class:: WrappedSklearnRegressor(sklearn_block: plaid.types.SklearnBlock, in_features_identifiers: list[plaid.containers.FeatureIdentifier], out_features_identifiers: list[plaid.containers.FeatureIdentifier])

   Bases: :py:obj:`sklearn.base.RegressorMixin`, :py:obj:`sklearn.base.BaseEstimator`


   Adapter for using a scikit-learn regressor with PLAID Dataset.

   Fits and predicts on tabular arrays extracted from stacked features,
   while preserving the feature/block structure expected by PLAID.

   :param sklearn_block: A scikit-learn regressor with fit/predict API.
   :param in_features_identifiers: List of feature identifiers for inputs.
   :param out_features_identifiers: List of feature identifiers for outputs.


   .. py:attribute:: sklearn_block


   .. py:attribute:: in_features_identifiers


   .. py:attribute:: out_features_identifiers


   .. py:method:: fit(dataset: plaid.Dataset, _y=None) -> Self

      Fits the wrapped scikit-learn regressor on the stacked input/output data.

      :param dataset: A `Dataset` containing both input and output features.
      :param _y: Ignored.

      :returns: The fitted regressor.
      :rtype: self



   .. py:method:: predict(dataset: plaid.Dataset) -> plaid.Dataset

      Predicts target values using the fitted regressor.

      :param dataset: A `Dataset` with input features.

      :returns: A new `Dataset` containing predicted target features.
      :rtype: Dataset