-
When I explained LSTM with captum, the following error occurred:
Traceback (most recent call last):
File "D:\论文机器学习模型\shiyan.py", line 210, in
attributions_ig, delta_ig = lig.att…
-
Our `IntegratedGradients` method is for TensorFlow models only, for feature parity it would be desirable to extend it to PyTorch models as well.
There are a few options and some design decisions to…
-
### 🐛 Describe the bug
![image](https://user-images.githubusercontent.com/56910741/227738299-a9311914-6452-4068-ab8a-12d74be45bb0.png)
Explainer.__call__() requires an input argument of batch
e…
-
Hi everyone, I encountered a problem while trying to run Captum Insights in a Jupyter Notebook. I installed the nbextensions mentioned in the Readme, but still receive a JS error. What am I missing?
…
-
## 🐛 Bug
Captum's layer attribution does attribution for the layer's output (actual input sample in input attribution = layer output in layer attribution). Captum supports multiple outputs but expe…
-
## 📚 Documentation
I would like to add an LRP tutorial/ notebook to help other people with LRP.
I suggest using this tutorial as a reference: [Python LRP tutorial](https://git.tu-berlin.de/gmont…
-
## 🐛 Bug
When LRP is used with a custom propagation rule, calling `attribute` more than a single time fails because the custom rule is removed in https://github.com/pytorch/captum/blob/5543b4a8d256…
-
## 🐛 Bug
I found a module that was listed in the code and on the API docs site, but there's not public path to use it and no tests were ever written for it:
```
captum.attr.InputBaselineXGradie…
-
## Issue
I am trying to use the Captum library to get attributions for my sequence-to-sequence T5 model. However the attributions are returned to the input token of the **decoder** rather than the i…
-
## 🐛 Bug
The `captum.attr.visualization` module is missing the link to its source code. I traced the issue with Sphinx's viewcode extension, and discovered that it gives the following warning.
`…