MultiSolverInterface#

class MultiSolverInterface(problem, models, optimizers=None, schedulers=None, weighting=None, use_lt=True)[source]#

Bases: SolverInterface

Base class for PINA solvers using multiple torch.nn.Module.

Initialization of the MultiSolverInterface class.

Parameters:

  • problem (AbstractProblem) – The problem to be solved.

  • models – The neural network models to be used.

  • optimizers (list[Optimizer]) – The optimizers to be used. If None, the torch.optim.Adam optimizer is used for all models. Default is None.

  • schedulers (list[Scheduler]) – The schedulers to be used. If None, the torch.optim.lr_scheduler.ConstantLR scheduler is used for all the models. Default is None.

  • weighting (WeightingInterface) – The weighting schema to be used. If None, no weighting schema is used. Default is None.

  • use_lt (bool) – If True, the solver uses LabelTensors as input.

Raises:

ValueError – If the models are not a list or tuple with length greater than one.

Warning

MultiSolverInterface uses manual optimization by setting automatic_optimization=False in LightningModule. For more information on manual optimization please see here.

on_train_batch_end(outputs, batch, batch_idx)[source]#

This method is called at the end of each training batch and overrides the PyTorch Lightning implementation to log checkpoints.

Parameters:

  • outputs (torch.Tensor) – The model’s output for the current batch.

  • batch (list[tuple[str, dict]]) – A batch of data. Each element is a tuple containing a condition name and a dictionary of points.

  • batch_idx (int) – The index of the current batch.

configure_optimizers()[source]#

Optimizer configuration for the solver.

Returns:

The optimizer and the scheduler

Return type:

tuple[list[Optimizer], list[Scheduler]]

property models#

The models used for training.

Returns:

The models used for training.

Return type:

torch.nn.ModuleList

property optimizers#

The optimizers used for training.

Returns:

The optimizers used for training.

Return type:

list[Optimizer]

property schedulers#

The schedulers used for training.

Returns:

The schedulers used for training.

Return type:

list[Scheduler]