pytorch_pfn_extras.handler.Logic

class pytorch_pfn_extras.handler.Logic(model_name='main', options=None)

Parameters

• model_name (str) –

• options (Optional[Dict[str, Any]]) –

Return type

None

__init__(model_name='main', options=None)

A set of methods that defines the training logic.

Parameters

• model_name (str) – Name of the model. Default is 'main'.

• options (dict, optional) –

  The configuration options.

  • 'backward_outputs' (list of str):

    A list of names of outputs that require compution of the gradient.

  • 'autocast' (bool):

    If True, torch.cuda.amp.autocast is enabled. Default is False.

  • 'grad_scaler' (torch.cuda.amp.GradScaler):

    A gradient scaler that outputs are applied to.

Return type

None

Methods

__init__([model_name, options])	A set of methods that defines the training logic.
consume_options(options)	A method to update options of Logic.
eval_step(models, batch_idx, batch)	A method for an evaluation step.
train_epoch_begin(models, epoch, loader)	A method called when starting a new epoch of training.
train_epoch_end(models, epoch)	A method called when completing an epoch of training.
train_step(models, optimizers, batch_idx, batch)	A method invokes the model forward and backward passes.
train_step_optimizers(models, optimizers, …)	A method in charge of stepping the provided optimizers.
train_validation_begin(models)	A method called when starting a validation.
train_validation_end(models)	A method called when the validation completes.

pytorch_pfn_extras.handler.BaseLogic pytorch_pfn_extras.handler.BaseHandler