[DLSP25] Add Attention Visualization with ViT tutorial#854
Open
lucas-oma wants to merge 1 commit intoAtcold:masterfrom
Open
[DLSP25] Add Attention Visualization with ViT tutorial#854lucas-oma wants to merge 1 commit intoAtcold:masterfrom
lucas-oma wants to merge 1 commit intoAtcold:masterfrom
Conversation
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
ViT Attention Visualization – No Spoilers Edition
This PR introduces a new tutorial: Vision Transformer (ViT) attention visualization, inspired by the last recitation class where PhD student Alex Wang showed us how to visualize attention in ViTs. However, his notebook contained a "from-scratch" ViT implementation, which, if uploaded to the repo, would basically be like handing over Homework 3 to future students on a silver platter. 😅
So instead, I took inspiration from it but recreated the visualization using Hugging Face's pre-trained ViT model. Same cool insights, no spoilers!
🚀 What’s New?
transformers.ViTModel📸 Why This Image?
Tried a couple of images, and this one stood out. The attention heads clearly focus on different elements, like tracking people walking, locking onto the NYU banner, and highlighting the building's structure. I think it’s a great example of how ViTs "see" an image!
Follow-ups:
If you spot any inconsistencies or mistakes in the descriptions, comments, etc, don’t hesitate to let me know! I did proofread it a couple of times though.