PT: keep the same checkpoint behavior as TF #3191
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Set the default
save_ckpttomodel.ckptas the prefix. When saving checkpoints,model.ckpt-100.ptwill be saved, andmodel.ckpt.ptwill be symlinked tomodel.ckpt-100.pt. Acheckpointfile will be dedicated to recordmodel.ckpt-100.pt.This keeps the same behavior as the TF backend. One can do the below using the PT backend just like the TF backend:
dp --pt train input.json # one can cancel the training before it finishes dp --pt freeze