flip.covariance.emulators.nnmatrix
==================================

.. py:module:: flip.covariance.emulators.nnmatrix


Attributes
----------

.. autoapisummary::

   flip.covariance.emulators.nnmatrix.log
   flip.covariance.emulators.nnmatrix.torch_install
   flip.covariance.emulators.nnmatrix.default_regression_object
   flip.covariance.emulators.nnmatrix._emulator_type


Classes
-------

.. autoapisummary::

   flip.covariance.emulators.nnmatrix.RegressionNet


Functions
---------

.. autoapisummary::

   flip.covariance.emulators.nnmatrix.train_torch_model
   flip.covariance.emulators.nnmatrix.train
   flip.covariance.emulators.nnmatrix.evaluate


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

.. py:data:: log

.. py:data:: torch_install
   :value: True


.. py:data:: default_regression_object

.. py:data:: _emulator_type
   :value: 'matrix'


.. py:class:: RegressionNet(input_dimension, dimension_hidden_layers=64, number_hidden_layers=3, output_dimension=1, activation_function=default_activation_function)

   Bases: :py:obj:`default_regression_object`


   .. py:attribute:: model


   .. py:method:: forward(x)


.. py:function:: train_torch_model(number_epochs, model, normalized_input, normalized_output, optimizer, loss_function, verbose, model_name)

.. py:function:: train(square_covariance, output_variance, output_non_diagonal, parameter_values, verbose=False, dimension_hidden_layers=64, number_hidden_layers=3, number_epochs=3000, activation_function=default_activation_function, loss_function=default_loss_function, tolerance_optimizer=0.001)

.. py:function:: evaluate(model, evaluation_value, evaluation_dictionary)