:py:mod:`ForeTiS.model._torch_model`
====================================

.. py:module:: ForeTiS.model._torch_model


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

Classes
~~~~~~~

.. autoapisummary::

   ForeTiS.model._torch_model.TorchModel




.. py:class:: TorchModel(optuna_trial, datasets, featureset_name, optimize_featureset, pca_transform = None, current_model_name = None, batch_size = None, n_epochs = None, target_column = None)

   Bases: :py:obj:`ForeTiS.model._base_model.BaseModel`, :py:obj:`abc.ABC`

   Parent class based on :obj:`~ForeTiS.model._base_model.BaseModel` for all PyTorch models to share functionalities.
   See :obj:`~ForeTiS.model._base_model.BaseModel` for more information.

   *Attributes*

       *Inherited attributes*

       See :obj:`~ForeTiS.model._base_model.BaseModel`.

       *Additional attributes*

       - batch_size (*int*): Batch size for batch-based training
       - n_epochs (*int*): Number of epochs for optimization
       - num_monte_carlo (*int*): Number of monte carlo iteration for the bayesian neural networks
       - optimizer (*torch.optim.optimizer.Optimizer*): optimizer for model fitting
       - loss_fn: loss function for model fitting
       - early_stopping_patience (*int*): epochs without improvement before early stopping
       - early_stopping_point (*int*): epoch at which early stopping occured
       - device (*torch.device*): device to use, e.g. GPU
       - X_scaler (*sklearn.preprocessing.StandardScaler*): Standard scaler for the X data

   :param optuna_trial: Trial of optuna for optimization
   :param datasets: all datasets that are available
   :param current_model_name: name of the current model according to naming of .py file in package model
   :param batch_size: batch size for neural network models
   :param n_epochs: number of epochs for neural network models
   :param target_column: the target column for the prediction

   .. py:method:: train_val_loop(train, val)

      Implementation of a train and validation loop for  PyTorch models.
      See :obj:`~ForeTiS.model._base_model.BaseModel` for more information


   .. py:method:: train_val_loader(train, val)

      Get the Dataloader with training and validation data

      :poram train: training data
      :param val: validation data

      :return: train_loader, val_loader, val


   .. py:method:: train_one_epoch(train_loader, scaler)

      Train one epoch

      :param train_loader: DataLoader with training data


   .. py:method:: validate_one_epoch(val_loader)

      Validate one epoch

      :param val_loader: DataLoader with validation data

      :return: loss based on loss-criterion


   .. py:method:: retrain(retrain)

      Implementation of the retraining for PyTorch models.
      See :obj:`~ForeTiS.model._base_model.BaseModel` for more information


   .. py:method:: update(update, period)

      Implementation of the retraining for PyTorch models.
      See :obj:`~ForeTiS.model._base_model.BaseModel` for more information


   .. py:method:: predict(X_in)

      Implementation of a prediction based on input features for PyTorch models.
      See :obj:`~ForeTiS.model._base_model.BaseModel` for more information


   .. py:method:: get_loss(outputs, targets)

      Calculate the loss based on the outputs and targets

      :param outputs: outputs of the model
      :param targets: targets of the dataset

      :return: loss


   .. py:method:: get_dataloader(X, y = None, only_transform = None, predict = False, shuffle = False)

      Get a Pytorch DataLoader using the specified data and batch size

      :param X: feature matrix to use
      :param y: optional target vector to use
      :param only_transform: whether to only transform or not
      :param predict: weather to use the data for predictions or not
      :param shuffle: shuffle parameter for DataLoader

      :return: Pytorch DataLoader


   .. py:method:: common_hyperparams()
      :staticmethod:

      Add hyperparameters that are common for PyTorch models.
      Do not need to be included in optimization for every child model.
      Also See :obj:`~ForeTiS.model._base_model.BaseModel` for more information


   .. py:method:: get_torch_object_for_string(string_to_get)
      :staticmethod:

      Get the torch object for a specific string, e.g. when suggesting to optuna as hyperparameter

      :param string_to_get: string to retrieve the torch object

      :return: torch object