ml_grid.pipeline.hyperparameter_search
======================================

.. py:module:: ml_grid.pipeline.hyperparameter_search


Classes
-------

.. autoapisummary::

   ml_grid.pipeline.hyperparameter_search.HyperparameterSearch


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

.. py:class:: HyperparameterSearch(algorithm: sklearn.base.BaseEstimator, parameter_space: Union[Dict, List[Dict]], method_name: str, global_params: Any, sub_sample_pct: int = 100, max_iter: int = 100, ml_grid_object: Any = None)

   Initializes the HyperparameterSearch class.

   :param algorithm: The scikit-learn compatible estimator instance.
   :type algorithm: BaseEstimator
   :param parameter_space: The hyperparameter search space.
   :type parameter_space: Union[Dict, List[Dict]]
   :param method_name: The name of the algorithm.
   :type method_name: str
   :param global_params: The global parameters object.
   :type global_params: Any
   :param sub_sample_pct: Percentage of the parameter space to sample for
                          randomized search. Defaults to 100.
   :type sub_sample_pct: int, optional
   :param max_iter: The maximum number of iterations for randomized or
                    Bayesian search. Defaults to 100.
   :type max_iter: int, optional
   :param ml_grid_object: The main pipeline object containing data and
                          other parameters. Defaults to None.
   :type ml_grid_object: Any, optional


   .. py:attribute:: algorithm
      :type:  sklearn.base.BaseEstimator

      The scikit-learn compatible estimator instance.


   .. py:attribute:: parameter_space
      :type:  Union[Dict, List[Dict]]

      The hyperparameter search space.


   .. py:attribute:: method_name
      :type:  str

      The name of the algorithm.


   .. py:attribute:: global_params
      :type:  ml_grid.util.global_params.global_parameters

      A reference to the global parameters singleton instance.


   .. py:attribute:: sub_sample_pct
      :type:  int

      Percentage of the parameter space to sample for randomized search.
      Defaults to 100.


   .. py:attribute:: max_iter
      :type:  int

      The maximum number of iterations for randomized or Bayesian search.
      Defaults to 100.


   .. py:attribute:: ml_grid_object
      :type:  Any

      The main pipeline object containing data and other parameters.


   .. py:method:: run_search(X_train: pandas.DataFrame, y_train: pandas.Series) -> sklearn.base.BaseEstimator

      Executes the hyperparameter search.

      This method selects the search strategy (Grid, Random, or Bayesian) based
      on global parameters and runs the search on the provided training data.

      :param X_train: Training features.
      :type X_train: pd.DataFrame
      :param y_train: Training labels.
      :type y_train: pd.Series

      :returns: The best estimator found during the search.
      :rtype: BaseEstimator