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Outdated suggestions cannot be applied.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.
I discovered another bug with the GCS config.
It looks like the
local_save_file
isn't really necessary since PTL trainer already saves the model locally.Also,
GCSCheckpointHandler
complains (after completing an epoch, which is not ideal) that it can't find a saved file if it's not specified correctly.We can instead upload the file that the trainer is already using and avoid having the user to manually specify a save file.
The
GCSCheckpointHandler
now also properly saves to the user's desired folder instead of the full path to the modelSo previously it would be saved as some thing like
<bucket_name>/home/user/path/to/model.ckpt
Now it actually saves it to the bucket/folder correctly.