checkpoint as a model config parameter for warmup cosine learning rates #66
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Outdated suggestions cannot be applied.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 PR adds the an extra parameter for the CLI to provide the checkpoint path to be loaded such that is compatible when using warmup cosine learning rates. By default, the lightning CLI has a
ckpt_path
, but this loads the previous checkpoint's learning rate as well as the weights. When using the latter option, the re-training for next model starts as a 'sine' function rather than cosine.