lasdi.gplasdi
=============

.. py:module:: lasdi.gplasdi


Classes
-------

.. autoapisummary::

   lasdi.gplasdi.BayesianGLaSDI


Functions
---------

.. autoapisummary::

   lasdi.gplasdi.average_rom
   lasdi.gplasdi.sample_roms
   lasdi.gplasdi.get_fom_max_std
   lasdi.gplasdi.optimizer_to


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

.. py:function:: average_rom(autoencoder, physics, latent_dynamics, gp_dictionary, param_grid)

.. py:function:: sample_roms(autoencoder, physics, latent_dynamics, gp_dictionary, param_grid, n_samples)

   Collect n_samples of ROM trajectories on param_grid.
   gp_dictionary: list of Gaussian process regressors (size of n_test)
   param_grid: numpy 2d array
   n_samples: integer
   assert(len(gp_dictionnary) == param_grid.shape[0])

   output: np.array of size [n_test, n_samples, physics.nt, autoencoder.n_z]


.. py:function:: get_fom_max_std(autoencoder, Zis)

   Computes the maximum standard deviation accross the parameter space grid and finds the corresponding parameter location



.. py:function:: optimizer_to(optim, device)

.. py:class:: BayesianGLaSDI(physics, autoencoder, latent_dynamics, param_space, config)

   .. py:attribute:: X_train


   .. py:attribute:: X_test


   .. py:attribute:: autoencoder


   .. py:attribute:: latent_dynamics


   .. py:attribute:: physics


   .. py:attribute:: param_space


   .. py:attribute:: timer


   .. py:attribute:: n_samples


   .. py:attribute:: lr


   .. py:attribute:: n_iter


   .. py:attribute:: max_iter


   .. py:attribute:: max_greedy_iter


   .. py:attribute:: ld_weight


   .. py:attribute:: coef_weight


   .. py:attribute:: optimizer


   .. py:attribute:: MSE


   .. py:attribute:: path_checkpoint


   .. py:attribute:: path_results


   .. py:attribute:: best_loss


   .. py:attribute:: best_coefs
      :value: None



   .. py:attribute:: restart_iter
      :value: 0



   .. py:method:: train()


   .. py:method:: get_new_sample_point()


   .. py:method:: export()


   .. py:method:: load(dict_)