**********************************
:py:mod:`wildboar.model_selection`
**********************************

.. py:module:: wildboar.model_selection

.. autoapi-nested-parse::

   
   Methods for model selection.
















   ..
       !! processed by numpydoc !!


Package Contents
----------------

Classes
-------

.. autoapisummary::

   wildboar.model_selection.RepeatedOutlierSplit



Functions
---------

.. autoapisummary::

   wildboar.model_selection.outlier_train_test_split



.. py:class:: RepeatedOutlierSplit(n_splits=None, *, test_size=0.2, n_outlier=0.05, shuffle=True, random_state=None)

   
   Repeated random outlier cross-validator.


   :Parameters:

       **n_splits** : int, optional
           The maximum number of splits.
           - if None, the number of splits is determined by the number of outliers
           as, `total_n_outliers/(n_inliers * n_outliers)`
           - if int, the number of splits is an upper-bound.

       **test_size** : float, optional
           The size of the test set.

       **n_outlier** : float, optional
           The fraction of outliers in the training and test sets.

       **shuffle** : bool, optional
           Shuffle the training indicies in each iteration.

       **random_state** : int or RandomState, optional
           The psudo-random number generator.









   .. rubric:: Notes

   Contrary to other cross-validation strategies, the random outlier
   cross-validator does not ensure that all folds will be different.
   Instead, the inlier samples are shuffled and new outlier samples
   are inserted in the training and test sets repeatedly.





   ..
       !! processed by numpydoc !!

   .. py:method:: get_n_splits(X, y, groups=None)

      
      Return the number of splitting iterations in the cross-validator.


      :Parameters:

          **X** : object
              The samples.

          **y** : object
              The labels.

          **groups** : object, optional
              Always ignored, exists for compatibility.

      :Returns:

          int
              Returns the number of splitting iterations in the cross-validator.













      ..
          !! processed by numpydoc !!


   .. py:method:: split(x, y, groups=None)

      
      Return training and test indicies.


      :Parameters:

          **x** : object
              Always ignored, exists for compatibility.

          **y** : object
              The labels.

          **groups** : object, optional
              Always ignored, exists for compatibility.


      :Yields:

          **train_idx, test_idx** : ndarray
              The training and test indicies












      ..
          !! processed by numpydoc !!


.. py:function:: outlier_train_test_split(x, y, normal_class, test_size=0.2, anomalies_train_size=0.05, random_state=None)

   
   Outlier training and testing split from classification dataset.


   :Parameters:

       **x** : array-like of shape (n_samples, n_timestep) or (n_samples, n_dim, n_timestep)
           Input data samples.

       **y** : array-like of shape (n_samples,)
           Input class label.

       **normal_class** : int
           Class label that should be considered as the normal class.

       **test_size** : float, optional
           Size of the test set.

       **anomalies_train_size** : float, optional
           Contamination of anomalies in the training dataset.

       **random_state** : int or RandomState, optional
           Psudo random state used for stable results.

   :Returns:

       **x_train** : array-like
           Training samples.

       **x_test** : array-like
           Test samples.

       **y_train** : array-like
           Training labels (either 1 or -1, where 1 denotes normal and -1 anomalous).

       **y_test** : array-like
           Test labels (either 1 or -1, where 1 denotes normal and -1 anomalous).










   .. rubric:: Examples

   >>> from wildboar.datasets import load_two_lead_ecg
   >>> x, y = load_two_lead_ecg()
   >>> x_train, x_test, y_train, y_test = train_test_split(
   ...     x, y, 1, test_size=0.2, anomalies_train_size=0.05
   ... )



   ..
       !! processed by numpydoc !!