feat: add longclip #20
Draft
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
koukandre
reviewed
Apr 24, 2024
src/open_clip/loss.py
Outdated
| @@ -16,6 +17,7 @@ | |||
| except ImportError: | |||
| hvd = None | |||
|
|
|||
| from utils import PCA | |||
koukandre
reviewed
Apr 24, 2024
| @@ -200,6 +205,9 @@ def train_one_epoch( | |||
|
|
|||
| losses['embedding_loss'] = args.emb_loss_weight * embedding_loss | |||
|
|
|||
| if args.longclip: | |||
| modelout_short = model(images_short, texts_short) | |||
| loss_short = loss(**modelout_short, output_dict=True, pca_dim=32) | |||
There was a problem hiding this comment.
losses['short_loss'] = loss(**modelout_short, output_dict=True, pca_dim=32)
this also, if we use only one loss for image-text pair
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Outdated suggestions cannot be applied.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.
support LongCLIP style training with an additional
longcliparg, during training apply a non-zero PCA onimage_featuresto obtain principal components ofimage_featuresand maintain ashort_losstogether with full-caption loss.