jacobgil/pytorch-grad-cam
A PyTorch library for visualizing and evaluating explainability in CNN and Vision Transformer models.
Not currently ranked — collecting fresh signals.
star history
This repository provides state-of-the-art explainability methods for computer vision models in PyTorch. It implements various pixel attribution techniques like Grad-CAM and Score-CAM for diagnosing model predictions. The library supports CNNs and Vision Transformers across tasks including classification, object detection, semantic segmentation, and embedding similarity. It also includes smoothing methods and trust metrics for evaluating explanation quality.
Frequently asked
- What is jacobgil/pytorch-grad-cam?
- A PyTorch library for visualizing and evaluating explainability in CNN and Vision Transformer models.
- Is pytorch-grad-cam open source?
- Yes — jacobgil/pytorch-grad-cam is open source, released under the MIT license.
- What language is pytorch-grad-cam written in?
- jacobgil/pytorch-grad-cam is primarily written in Python.
- How popular is pytorch-grad-cam?
- jacobgil/pytorch-grad-cam has 12.9k stars on GitHub.
- Where can I find pytorch-grad-cam?
- jacobgil/pytorch-grad-cam is on GitHub at https://github.com/jacobgil/pytorch-grad-cam.