Add flag skip_epoch_completed to Engine.terminate_epoch()

[bonassifabio]

🚀 Feature

It is proposed to add a flag skip_epoch_completed (which defaults to False) to the method Engine.terminate_epoch() to terminate the current epoch without firing Events.EPOCH_COMPLETED.

Motivation

Suppressing the firing of Events.EPOCH_COMPLETED might be desirable for example to avoid checkpointing, early stopping, validation, and calling the scheduler step at the end of a terminated epoch.

What's happening now

Right now, Events.EPOCH_COMPLETED is always fired after Events. TERMINATE_SINGLE_EPOCH.

What would happen after this change

By default, the behavior would be the same, leaving the user the possibility to prevent Events.EPOCH_COMPLETED from being fired.

Context

See issue #3308 and PR #3309

[github-actions]

Hey 👋, I've just created a thread for this issue on PyTorch-Ignite Discord where you can quickly talk to the community on the topic.

^{🤖 This comment was automatically posted by Discuss on Discord}