use trained model if no custom checkpoint call back is specified. #3199
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Peng Zhang <pengz@uber.com>
Signed-off-by: Peng Zhang <pengz@uber.com>
chongxiaoc
reviewed
Oct 5, 2021
| @@ -128,11 +128,16 @@ def train(serialized_model): | |||
| # Lightning requires to add checkpoint callbacks for all ranks. | |||
| # Otherwise we are seeing hanging in training. | |||
| _checkpoint_callback = checkpoint_callback | |||
There was a problem hiding this comment.
For simplicity, just relying on this _checkpoint_callback variable != None check is enough.
If _checkpoint_callback is not None: modify dir_path and filename, and callbacks.append(_checkpoint_callback)
There was a problem hiding this comment.
if _checkpoint_callback is None, it will set to default ModelCheckpoint in line 141.
If user do not pass in a checkpoint callback, we still want to add the default ModelCheckpoint in to callback list, right?
chongxiaoc
reviewed
Oct 5, 2021
| # rank 0 overwrites model with best checkpoint and returns. | ||
| best_model = model.load_from_checkpoint(_checkpoint_callback.best_model_path) | ||
| # Rank 0 overwrites model with best checkpoint and returns. | ||
| best_model = model if use_last_model_weights else model.load_from_checkpoint(_checkpoint_callback.best_model_path, strict=False) |
There was a problem hiding this comment.
best_model = model if not _checkpoint_callback else ...
There was a problem hiding this comment.
_checkpoint_callback is always not None because of the line 141
chongxiaoc
requested changes
Oct 5, 2021
|
duplicated with #3201 |
Merged
4 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Checklist before submitting
Description
In lightning estimator, we do not need to load from check point if only want to return the last weights. The model after trianer.fit already have them.