hila-chefer/Transformer-Explainability
[CVPR 2021] Official PyTorch implementation for Transformer Interpretability Beyond Attention Visualization, a novel method to visualize classifications by Transformer based networks.
Jupyter NotebookMIT
Issues
- 0
- 0
- 2
- 1
The relprop method of the Linear layer
#64 opened by Hongbo-Z - 0
Seeking Assistance: Adapting Explainability Technique for Longformer Model in Text Classification
#69 opened by ibrahimAlaaeddine01 - 1
gtsegs_ijcv.mat NA
#68 opened by Comodore125 - 2
- 0
Does def safe_divide really work?
#67 opened by 686llhhhhhh - 4
- 0
LRP for mean
#65 opened by LokeshBadisa - 1
bert colab does not work
#61 opened by leonorsilva - 5
name 'SequenceClassifierOutput' is not defined
#51 opened by songhuadan - 0
how to use
#62 opened by Flylee864 - 1
How can I adapt this to my custom model?
#58 opened by sarah-muller - 1
Question about Conv2D relprop
#55 opened by rtyu-1 - 0
The relprop method of the Linear layer seems to be unstable for small number of input features
#59 opened by josh-oo - 0
Training process
#57 opened by ascdqz - 1
Extension to other huggingface transformer models?
#54 opened by mollie9523 - 0
Could you please provide the code for reproducing the perturbation result in Table 1?
#52 opened by jwu125 - 2
Threshold for Attention Vector during comparison
#50 opened by jaydebsarker - 3
- 2
Questions regarding the evaluation process
#44 opened by zmy1116 - 2
About relprop in BERT.py
#46 opened by szubing - 1
- 1
Can you expand the lrp.py to cover BigBird?
#43 opened by ngoel17 - 2
Question regarding fine tuning
#40 opened by MichaelMMeskhi - 1
- 1
- 1
- 1
How to evaluate transformer explainability methods & Compare with other XAI methods
#38 opened by jaiswati - 1
Different Bert Models
#37 opened by Christoforos00 - 6
confused rgb/bgr?
#35 opened by Tsingularity - 4
Implementing with my own model
#34 opened by LanariLorenzo - 10
Load locally fine-tuned models
#25 opened by lucasresck - 6
How to accelerate when inferring CAM in parallel
#32 opened by Muyun99 - 1
gtsegs_ijcv.mat
#33 opened by torchmyheart - 2
Question about perturbation results
#31 opened by scott870430 - 2
Question about layers_ours.py
#29 opened by Mfun82 - 2
On DEIT distilled network
#30 opened by kspruthviraj - 4
Token Classification
#28 opened by zooeygirl - 2
Couldn't find corresponding code with 《Transformer Interpretability Beyond Attention Visualization》
#27 opened by Yung-zi - 2
Question about Conv2D relprop
#26 opened by jhb115 - 3
visualize TimeSformer
#24 opened by Holmes-Alan - 5
- 2
ImageNet-Segmentation dataset
#23 opened by JindongGu - 1
- 2
- 2
- 7
Reproducibility in another text dataset
#17 opened by carlosabcs - 8