captum
captum copied to clipboard
Model interpretability and understanding for PyTorch
## 🚀 Feature Interpreting LiDAR/PC-based model ## Motivation I'm frustrated and tired of integrating [PointNet](https://github.com/yanx27/Pointnet_Pointnet2_pytorch) and [LIME-3D](https://github.com/Explain3D/LIME-3D). ## Pitch Leads XAI from 2D to 3D. ## Alternatives Represents 3D point...
## ❓ Questions and Help the classification model is classification_pred = model([input_x[0], input_x[1],input_x[2], input_x[3]], os_event.size(0), {'gex': 0, 'methy': 1, 'mut':2, 'cna': 3}),the input_x of one sample is as the figure...
## 🚀 Feature It seems to me there is no reason why LayerLRP is restricted to lists only. It seems any ordered iterable should do the trick here. I would...
## 🐛 Bug ## To Reproduce Steps to reproduce the behavior: Run the code below, which is trying to use layer conductance attribution for llama2. ```python import torch from transformers...
I'm unable to find any guidelines for TCAV in medical applications. The only information I found was https://github.com/tensorflow/tcav/issues/115, where Kim Been confirms that we center-crop on the medical concepts. So...
I posted this originally on StackOverflow, but now I thought it may fit better here. After all it is related to this setup. I want to change the default classifier...
I posted this originally on StackOverflow, but now I thought it may fit better here. After all it is related to this setup. I am using the captum setup https://captum.ai/api/concept.html...
Summary: As title. Looks like `Any` is OK now...thank goodness Differential Revision: D77395444
So this project got left on the back burner for a while but I've brought it back. The original plan was to either merge it into Captum or spin it...