# frgfm/torch-cam

Class activation maps for your PyTorch models (CAM, Grad-CAM, Grad-CAM++, Smooth Grad-CAM++, Score-CAM, SS-CAM, IS-CAM, XGrad-CAM, Layer-CAM, Finer-CAM, LeGrad, RefineCAM)

Repository: https://github.com/frgfm/torch-cam
Canonical: https://ross.abutalabs.com/products/torch-cam
Homepage: https://frgfm.github.io/torch-cam/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: pytorch, python, deep-learning, cnn, activation-maps, gradcam-plus-plus, gradcam, saliency-map, interpretability, interpretable-deep-learning, smoothgrad, score-cam, class-activation-map, grad-cam
Last push: 2026-08-26T08:03:10+00:00

## Health v2 (maintenance only)
Score: 72/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 22, longevity 100
- inputs: {"age_days": 2354, "days_push": 7, "days_rel": 310, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2304, forks 227 (observed 2026-08-28T04:06:35.699930+00:00)

## What it is
TorchCAM is a Python library that extracts class activation maps (CAMs) from PyTorch CNN classifiers, supporting many CAM variants such as Grad-CAM, Score-CAM, Layer-CAM, and more. It offers a minimal hook-based API for visualizing which spatial features influence a model's classification outputs.

## Use cases
- visualize which image regions a CNN classifier focuses on
- debug surprising model predictions with saliency maps
- compare Grad-CAM, Score-CAM, and other CAM methods on a PyTorch model
- explain classification outputs for a pretrained ResNet
- research and prototype new class activation map techniques
- generate activation map overlays for model interpretability reports

## When to avoid
- you need interpretability for transformers or non-CNN architectures
- you work in TensorFlow, JAX, or another non-PyTorch framework
- you need production-grade model explanation pipelines rather than exploratory visualization

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, data-visualization
- domain: deep-learning, computer-vision, machine-learning
- platform: python
- tags: pytorch, grad-cam, class-activation-map, saliency-map, explainability, cnn, interpretability

## Member repositories
- frgfm/torch-cam (main) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:35.699930+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:39:43.435328+00:00, confidence not recorded.
  - readme: https://github.com/frgfm/torch-cam (fetched 2026-08-28T04:06:35.699930+00:00, sha 38cffcf2c41f)
  - homepage: https://frgfm.github.io/torch-cam/ (fetched 2026-08-29T10:20:07.256895+00:00, sha cdc38b08ad40)
- Data as of 2026-08-30T08:39:29.467469+00:00.