# DSE-MSU/DeepRobust

A pytorch adversarial library for attack and defense methods on images and graphs

Repository: https://github.com/DSE-MSU/DeepRobust
Canonical: https://ross.abutalabs.com/products/deeprobust
Language: Python
License: MIT
License Family: permissive
Topics: adversarial-attacks, adversarial-examples, deep-neural-networks, defense, graph-neural-networks, machine-learning, deep-learning, graph-convolutional-networks, graph-mining
Last push: 2025-06-26T02:36:41+00:00

## Health v2 (maintenance only)
Score: 45/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 28, release rhythm 35, longevity 100
- inputs: {"age_days": 2538, "days_push": 434, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1085, forks 192 (observed 2026-08-28T04:03:32.021659+00:00)

## What it is
DeepRobust is a PyTorch library for adversarial robustness research, providing implementations of attack and defense methods for both image classifiers and graph neural networks. It is aimed at researchers benchmarking adversarial examples and defenses in machine learning.

## Use cases
- generate adversarial examples against image classifiers
- evaluate robustness of deep neural networks
- run adversarial attacks on graph neural networks
- test graph poisoning and evasion defenses
- benchmark attack and defense algorithms in PyTorch
- research adversarial machine learning on images and graphs

## When to choose
- you need ready-made adversarial attack and defense implementations in PyTorch
- you study robustness of both image models and graph neural networks
- you want a benchmarking baseline library for adversarial ML research

## When to avoid
- you need production-grade model security rather than research tooling
- you work in TensorFlow or another framework instead of PyTorch
- you only need general deep learning utilities without adversarial focus

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, security, benchmarking
- domain: machine-learning, deep-learning, security, computer-vision
- platform: python
- tags: adversarial-attacks, adversarial-defense, adversarial-robustness, graph-neural-networks, pytorch, adversarial-examples

## Member repositories
- DSE-MSU/DeepRobust (main) score 45

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:32.021659+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 2026-08-30T06:49:40.708029+00:00, confidence not recorded.
  - readme: https://github.com/DSE-MSU/DeepRobust (fetched 2026-08-28T04:03:32.021659+00:00, sha 1609b6911108)
  - registry_pypi: https://pypi.org/pypi/deeprobust/json (fetched 2026-08-29T12:52:36.740622+00:00, sha 601e37189ec8)
- Data as of 2026-08-30T08:39:29.467469+00:00.