Lightning Wrapper¶
As explained in Training and Evaluation, Asteroid provides a thin wrapper on the top of PyTorchLightning for training your models.
-
class
asteroid.engine.system.
System
(model, optimizer, loss_func, train_loader, val_loader=None, scheduler=None, config=None)[source]¶ Bases:
pytorch_lightning.core.lightning.LightningModule
Base class for deep learning systems. Contains a model, an optimizer, a loss function, training and validation dataloaders and learning rate scheduler.
Note that by default, any PyTorch-Lightning hooks are not passed to the model. If you want to use Lightning hooks, add the hooks to a subclass:
class MySystem(System): def on_train_batch_start(self, batch, batch_idx, dataloader_idx): return self.model.on_train_batch_start(batch, batch_idx, dataloader_idx)
Parameters: - model (torch.nn.Module) – Instance of model.
- optimizer (torch.optim.Optimizer) – Instance or list of optimizers.
- loss_func (callable) – Loss function with signature (est_targets, targets).
- train_loader (torch.utils.data.DataLoader) – Training dataloader.
- val_loader (torch.utils.data.DataLoader) – Validation dataloader.
- scheduler (torch.optim.lr_scheduler._LRScheduler) – Instance, or list
of learning rate schedulers. Also supports dict or list of dict as
{"interval": "step", "scheduler": sched}
whereinterval=="step"
for step-wise schedulers andinterval=="epoch"
for classical ones. - config – Anything to be saved with the checkpoints during training. The config dictionary to re-instantiate the run for example.
Note
By default,
training_step
(used bypytorch-lightning
in the training loop) andvalidation_step
(used for the validation loop) sharecommon_step
. If you want different behavior for the training loop and the validation loop, overwrite bothtraining_step
andvalidation_step
instead.For more info on its methods, properties and hooks, have a look at lightning’s docs: https://pytorch-lightning.readthedocs.io/en/stable/lightning_module.html#lightningmodule-api
-
forward
(*args, **kwargs)[source]¶ Applies forward pass of the model.
Returns: torch.Tensor
-
common_step
(batch, batch_nb, train=True)[source]¶ Common forward step between training and validation.
The function of this method is to unpack the data given by the loader, forward the batch through the model and compute the loss. Pytorch-lightning handles all the rest.
Parameters: - batch – the object returned by the loader (a list of torch.Tensor in most cases) but can be something else.
- batch_nb (int) – The number of the batch in the epoch.
- train (bool) – Whether in training mode. Needed only if the training and validation steps are fundamentally different, otherwise, pytorch-lightning handles the usual differences.
Returns: torch.Tensor
– The loss value on this batch.Note
This is typically the method to overwrite when subclassing
System
. If the training and validation steps are somehow different (except forloss.backward()
andoptimzer.step()
), the argumenttrain
can be used to switch behavior. Otherwise,training_step
andvalidation_step
can be overwriten.
-
training_step
(batch, batch_nb)[source]¶ Pass data through the model and compute the loss.
Backprop is not performed (meaning PL will do it for you).
Parameters: - batch – the object returned by the loader (a list of torch.Tensor in most cases) but can be something else.
- batch_nb (int) – The number of the batch in the epoch.
Returns: torch.Tensor, the value of the loss.
-
validation_step
(batch, batch_nb)[source]¶ Need to overwrite PL validation_step to do validation.
Parameters: - batch – the object returned by the loader (a list of torch.Tensor in most cases) but can be something else.
- batch_nb (int) – The number of the batch in the epoch.
-
on_save_checkpoint
(checkpoint)[source]¶ Overwrite if you want to save more things in the checkpoint.
-
static
config_to_hparams
(dic)[source]¶ Sanitizes the config dict to be handled correctly by torch SummaryWriter. It flatten the config dict, converts
None
to"None"
and any list and tuple into torch.Tensors.Parameters: dic (dict) – Dictionary to be transformed. Returns: dict – Transformed dictionary.