TinyCLIP integration for ActionCLIP#47
Open
FransHk wants to merge 4 commits into
Open
Conversation
Relic from earlier integration.
Updated all .yaml files to include novel architectures. Skip checksum only on TinyCLIP
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 integrates two TinyCLIP ViT models to the existing model framework with minimal changes. This is possible because TinyCLIP provides a pure ViT-based model, like CLIP. The TinyCLIP model is a CLIP distillation that provides significant speed-ups to the CLIP model while retaining and in some cases improving its zero-shot IN1K accuracy. A small state_dict conversion helper method and optional sha256 ignore flag are added to accommodate for this integration.
The TinyCLIP paper
The TinyCLIP models (Git)
Graphs below show rough indication of ActionCLIP during train time on HMDB51 (no pre-train). Train step indicates the batches processed per minute (wall clock) time. TinyCLIP-based ActionCLIP model trains much faster while performance is almost similar to vanilla CLIP.
