fix: correct clip_txt_pool variable name typo#146
Open
Mr-Neutr0n wants to merge 1 commit intoStability-AI:masterfrom
Open
fix: correct clip_txt_pool variable name typo#146Mr-Neutr0n wants to merge 1 commit intoStability-AI:masterfrom
Mr-Neutr0n wants to merge 1 commit intoStability-AI:masterfrom
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Bug
In
modules/stage_c.py, thegen_c_embeddingsmethod unsqueezesclip_txt_pooledinto a new variableclip_txt_poolwhen the tensor is 2D (line 158). However, the subsequent mapper call on line 161 still uses the originalclip_txt_pooled— meaning the unsqueeze result is silently discarded.This is inconsistent with how
clip_imgis handled two lines below, whereclip_img = clip_img.unsqueeze(1)correctly updates the same variable that is then passed toclip_img_mapper.Fix
Changed
clip_txt_pool = clip_txt_pooled.unsqueeze(1)toclip_txt_pooled = clip_txt_pooled.unsqueeze(1)so the unsqueezed tensor is actually used byclip_txt_pooled_mapper, matching theclip_imgpattern.