hila-chefer Transformer-MM-Explainability issues

hila-chefer / Transformer-MM-Explainability

[ICCV 2021- Oral] Official PyTorch implementation for Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers, a novel method to visualize any Transformer-based network. Including examples for DETR, VQA.

MIT License

801 stars 107 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

How to use these examples to visualize the Grounding Dino or GLIP?

#41 meaquanana opened 1 week ago
0
No negative word importance

#40 Faiail opened 7 months ago
0
torch.nn.modules.module.ModuleAttributeError: 'ResidualAttentionBlock' object has no attribute 'attn_probs'

#39 TongLi97 opened 8 months ago
7
ssl error

#38 fzb408 opened 10 months ago
0
How to apply this work on google/vit model from hugging face ?

#37 MohieEldinMuhammad opened 1 year ago
1
save_visual_results in visualBERT

#36 guanhdrmq opened 1 year ago
0
fix memory leak

#35 lenbrocki opened 1 year ago
0
Applicability to decoder transformers with causal mask

#34 Aamer98 closed 1 year ago
0
when trying to use the colab notebook for RN50 im getting AttributeError: 'ModifiedResNet' object has no attribute 'transformer'

#33 jiheddachraoui opened 1 year ago
0
self.attn_probs in ResidualAttentionBlock() causes problems - how to make explainability work with mlfoundations / open_clip model

#32 tahirmiri opened 1 year ago
0
Is this really using the technique from the publication?

#31 entrity closed 1 year ago
1
Checking how well this works with Segment Anything?

#30 nahidalam closed 1 year ago
1
Use non hacked models

#29 josh-freeman closed 1 year ago
1
Readability of CLIP notebook

#28 josh-freeman opened 1 year ago
4
Update clip.py

#27 josh-freeman closed 1 year ago
1
Application to Sparse/Low-Rank Attention Matrices

#26 FarzanT closed 1 year ago
1
CVE-2007-4559 Patch

#25 TrellixVulnTeam opened 2 years ago
0
Question about the CLIP Demo

#24 Hoyyyaard closed 2 years ago
1
ImportError: No module named lxmert.lxmert.src.tasks

#23 CaffreyR closed 2 years ago
2
Using the methods for a custom architecture

#22 nelaturuharsha closed 2 years ago
4
Problems with running it in Google colab

#21 songhuadan closed 2 years ago
2
Adds link to the Huggingface demo for CLIP explainability

#20 bpiyush closed 2 years ago
0
Object detection/Segmentation Explainability

#19 jaiswati closed 2 years ago
1
Question about the visualization of CLIP‘s text token

#18 Kihensarn closed 2 years ago
2
Details about the changes in the code of base models

#17 NikhilM98 opened 2 years ago
1
In 6.2 LXMERT, report an error"requests.exceptions.MissingSchema: Invalid URL 'val2014COCO_val2014_000000092107.jpg': No schema supplied. "

#16 Shuai-Lv closed 2 years ago
1
The data set downloaded automatically is too large

#15 Shuai-Lv closed 2 years ago
1
How can I choose the method when I run the script ？

#14 Shuai-Lv closed 2 years ago
2
Generate relevance matrix in ViT of Hugging Face

#13 SketchX-QZY closed 2 years ago
2
Questions about the Relavance Matrix

#12 Paandaman closed 2 years ago
2
Batching for CLIP Explainability

#11 Alacarter closed 2 years ago
2
Questions about CLIP visualization.

#10 tingxueronghua closed 2 years ago
4
Question about the visualization for VIT?

#9 zhaoxin94 closed 2 years ago
2
CLIP ViT-B/16

#8 sanjayss34 closed 2 years ago
11
COCO 2014 or 2017?

#7 anguyen8 closed 3 years ago
4
Question about Vit

#6 scott870430 closed 2 years ago
6
when trying to use the colab notebook for clip-vit/B16 im getting ModuleAttributeError: 'ResidualAttentionBlock' object has no attribute 'attn_probs'

#5 DanJbk closed 3 years ago
2
Swin Transformer

#4 KP-Zhang closed 3 years ago
8
attn_grad

#3 betterze closed 3 years ago
2
Request for vanilla example notebook

#2 oliverdutton closed 3 years ago
7
Question about the CLIP Demo

#1 g-luo closed 3 years ago
2