IterBasedTrainLoop

class mmengine.runner.IterBasedTrainLoop(runner, dataloader, max_iters, val_begin=1, val_interval=1000, dynamic_intervals=None)

Loop for iter-based training.

Parameters

• runner (Runner) – A reference of runner.

• dataloader (Dataloader or dict) – A dataloader object or a dict to build a dataloader.

• max_iters (int) – Total training iterations.

• val_begin (int) – The iteration that begins validating. Defaults to 1.

• val_interval (int) – Validation interval. Defaults to 1000.

• dynamic_intervals (List[Tuple[int, int]], optional) – The first element in the tuple is a milestone and the second element is a interval. The interval is used after the corresponding milestone. Defaults to None.

Return type

None

property epoch

Current epoch.

Type

int

property iter

Current iteration.

Type

int

property max_epochs

Total epochs to train model.

Type

int

property max_iters

Total iterations to train model.

Type

int

run()

Launch training.

Return type

None

run_iter(data_batch)

Iterate one mini-batch.

Parameters

data_batch (Sequence[dict]) – Batch of data from dataloader.

Return type

None