test: add encode_vision tests for Embedder module#589
Open
stanley1208 wants to merge 1 commit intogoogle-deepmind:mainfrom
Open
test: add encode_vision tests for Embedder module#589stanley1208 wants to merge 1 commit intogoogle-deepmind:mainfrom
stanley1208 wants to merge 1 commit intogoogle-deepmind:mainfrom
Conversation
Resolves TODO(mblondel) requesting tests for encode_vision. Adds 2 tests verifying the vision projection from vision_proj_dim to embed_dim with various batch and patch dimensions.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Resolves
TODO(mblondel): Add tests for encode_vision herein_modules_test.py.The
Embedder.encode_vision()method projects SigLiP vision embeddings into the text embedding space via RMSNorm + Einsum projection. This was previously untested.Tests added (2 tests in
_modules_test.py)test_encode_vision_output_shape: verifies projection from[B, num_patches, vision_proj_dim]to[B, num_patches, embed_dim]test_encode_vision_different_batch_shapes: verifies correct output with single and multiple patchesTest plan
_modules_test.pytests unaffected