skip vision parts of the model for test_train_vlm_multi_image as well#5774
Open
kaixuanliu wants to merge 1 commit into
Open
skip vision parts of the model for test_train_vlm_multi_image as well#5774kaixuanliu wants to merge 1 commit into
kaixuanliu wants to merge 1 commit into
Conversation
Signed-off-by: Liu, Kaixuan <kaixuan.liu@intel.com>
Member
|
|
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
We need to skip vision part of the model in multi-image test case, just like what we did in test_train_vlm, pls help review, @qgallouedec @kashif
Note
Low Risk
Low risk: test-only change that relaxes parameter-update assertions for known non-updating vision components in tiny VLMs.
Overview
Aligns
test_train_vlm_multi_imagewithtest_train_vlmby skipping parameter-change assertions for vision-related modules (e.g.,vision_tower, projector,visual,image_newline) where tiny model initialization can prevent gradient flow.This reduces flaky/false-negative failures in the multi-image VLM training test while still asserting that the rest of the model updates during
trainer.train().Reviewed by Cursor Bugbot for commit 88856bd. Bugbot is set up for automated code reviews on this repo. Configure here.