MLP Classifier (MLPC)
=====================

.. class:: MLPC(self, model_name: str = 'MLP Classifier', random_state: int = 42, **kwargs)

MLP Classifier Wrapper class  - parent class :class:`Classifier`

.. list-table::
   :widths: 25 75
   :header-rows: 0

   * - Parameters
     - 
        hidden_layer_sizes : tuple[int],
            the ith element represents the number of neurons in the ith hidden layer
        activation : str,
            activation function for the hidden layer
        solver : str,
            solver for weight optimization
        alpha : float,
            l2 penalty (regularization term) parameter
        learning_rate : str,
            learning rate schedule for weight updates
        tol : float,
            tolerance for stopping criteria
        max_iter : int,
            maximum number of iterations taken for the solvers to converge
        batch_size : int or str,
            size of minibatches for stochastic optimizers
        early_stopping : bool,
            True: tests on 10% of train data and stops if there is for 'n_iter_no_change' no improvement in the metrics
        random_state : int,                 default=42
            random_state for model
        


   * - Attributes
     - 
         
        cv_scores : dict[str, float]
            dictionary with cross validation results
        
       
         
        feature_names : list[str]
            names of all the features that the model saw during training. Is empty if model was not fitted yet.
        
       
         
        grid : ConfigurationSpace
            hyperparameter tuning grid of the model
        
       
         
        model : model object
            model with 'fit', 'predict', 'set_params', and 'get_params' method (see sklearn API)
        
       
         
        model_name : str
            name of the model. Used in loading bars and dictionaries as identifier of the model
        
       
         
        model_type : str
            kind of estimator (e.g. 'RFC' for RandomForestClassifier)
        
       
         
        rCVsearch_results : pd.DataFrame or None
            results from randomCV hyperparameter tuning. Is ``None`` if randomCVsearch was not used yet.
        
       
         
        train_score : float
            train score value
        
       
         
        train_time : str
            train time in format: "0:00:00" (hours:minutes:seconds)
        
       

.. note::

   You can use all parameters of the wrapped model when initialising the wrapper class.

.. raw:: html

   <h2>Example</h2>

>>> from sam_ml.models.classifier import MLPC
>>>
>>> model = MLPC()
>>> print(model)
MLPC(model_name='MLP Classifier')
        

.. raw:: html

   <h2>Methods</h2>

.. list-table::
   :widths: 25 75
   :header-rows: 1

   * - Method
     - Description
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.cross_validation`
     - Random split crossvalidation
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.cross_validation_small_data`
     - One-vs-all cross validation for small datasets
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.evaluate`
     - Function to create multiple scores with predict function of model
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.evaluate_proba`
     - Function to create multiple scores for binary classification with predict_proba function of model
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.evaluate_score`
     - Function to create a score with predict function of model
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.evaluate_score_proba`
     - Function to create a score for binary classification with predict_proba function of model
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.feature_importance`
     - Function to generate a matplotlib plot of the top45 feature importance from the model.
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.fit`
     - Function to fit the model
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.fit_warm_start`
     - Function to warm_start fit the model
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.get_deepcopy`
     - Function to create a deepcopy of object
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.get_params`
     - Function to get the parameter from the model object
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.get_random_config`
     - Function to generate one grid configuration
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.get_random_configs`
     - Function to generate grid configurations
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.load_model`
     - Function to load a pickled model class object
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.predict`
     - Function to predict with predict-method from model object
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.predict_proba`
     - Function to predict with predict_proba-method from model object
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.randomCVsearch`
     - Hyperparametertuning with randomCVsearch
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.replace_grid`
     - Function to replace self.grid
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.save_model`
     - Function to pickle and save the class object
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.set_params`
     - Function to set the parameter of the model object
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.smac_search`
     - Hyperparametertuning with SMAC library HyperparameterOptimizationFacade [can only be used in the sam_ml version with swig]
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.train`
     - Function to train the model
   
   * - :meth:`~sam_ml.models.classifier.MLPClassifier.MLPC.train_warm_start`
     - Function to warm_start train the model
   

.. note::

   A lot of methods use parameters for advanced scoring. For additional information on advanced scoring, see :ref:`scoring documentation <classifier-scoring-section>`


.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.cross_validation

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.cross_validation_small_data

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.evaluate

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.evaluate_proba

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.evaluate_score

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.evaluate_score_proba

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.feature_importance

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.fit

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.fit_warm_start

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.get_deepcopy

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.get_params

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.get_random_config

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.get_random_configs

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.load_model

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.predict

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.predict_proba

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.randomCVsearch

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.replace_grid

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.save_model

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.set_params

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.smac_search

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.train

.. automethod:: sam_ml.models.classifier.MLPClassifier.MLPC.train_warm_start