Fix optimizer reloading from checkpoint #1329
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Outdated suggestions cannot be applied.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.
This is kind of a weird problem.
Basically if you have a model like:
It has a
parameters()
like{"l1": ..., "drop": {}, "l2": ...}
. The optimizer state is setup to match that. When serializing the optimizer state it gets flattened which removes the emptydrop
value from the tree. That in turn causes issues when deserializing the optimizer because its' state no longer matches that ofparameters()
.I think there are two possible fixes here, I implemented the first, but I'm open to trying the second:
Module.parameters()
andModule.trainable_parameters()
to not include empty values.This closes #1328