Pipelines

class ignite_framework.pipelines.supervised_engines.SupervisedEvaluator(name, model, loss_fn, dataloader=None, engine_run_started_ref=<ignite_framework.framework.framework.StateObjectsReference object>, n_epochs=1, n_samples=None, device='', non_blocking=False, prepare_batch_fn=<function _prepare_batch>)

Supervised Evaluator for standardized evaluation.

Parameters:

• name –
• model –
• loss_fn –
• dataloader –
• engine_run_started_ref –
• n_epochs –
• n_samples –
• device –
• non_blocking –
• prepare_batch_fn –

device

StateDevice is the device parameter defining the device the processes of the state object container should run on.

get(*name_or_parts)

Get method for parameterized attribute names, especially for state object containers, but not only.

Main reason is to simply avoid unhandy commands, e.g. getattr(state.transitions, ‘{}_{}’.format(name, suffix) commands and simply write state.transitions.get(name, suffix).

get_bso_names()

name = 'SupervisedEvaluator'

run(dataloader=None)

set(*name_or_parts_and_value)

class ignite_framework.pipelines.supervised_engines.SupervisedTrainer(name, model, optimizer, loss_fn, dataloader=None, engine_run_started_ref=<ignite_framework.framework.framework.StateObjectsReference object>, n_epochs=3, n_samples=None, device='', non_blocking=False, prepare_batch_fn=<function _prepare_batch>)

Supervised trainer for standard training loops.

Parameters:

• name – name of supervised trainer instance and name part of all composed state object containers
• model – model that will be assigned to state.modules if not already
• optimizer – optimizer that will be assigned to state.optimizers if not already
• loss_fn –
• dataloader – dataloader that will be assigned to state.dataloaders if not already
• engine_run_started_ref – state object to which’s callbacks the engine’s run’method will be added.
• n_epochs – total number of training epochs
• n_samples – total number of training samples, overrules n_epochs
• device – device on which the model will run
• non_blocking –
• prepare_batch_fn –

device

StateDevice is the device parameter defining the device the processes of the state object container should run on.

get(*name_or_parts)

Get method for parameterized attribute names, especially for state object containers, but not only.

Main reason is to simply avoid unhandy commands, e.g. getattr(state.transitions, ‘{}_{}’.format(name, suffix) commands and simply write state.transitions.get(name, suffix).

get_bso_names()

name = 'SupervisedTrainer'

run(dataloader=None)

set(*name_or_parts_and_value)

ignite_framework.pipelines.supervised_engines.get_labels(value)