Back to packageRELEASE DETAILS

2.1.1

PyPI

advtorchattacks

A PyTorch library for adversarial attacks, inspired by torchattacks.

Published4 Mar 2025

Direct dependencies4

Transitive dependenciesNot calculated

README

Version 2.1.1 documentation

AdvTorchAttacks

AdvTorchAttacks is a PyTorch-based library for generating adversarial attacks on deep learning models. It provides implementations of widely used attack methods, enabling user to evaluate model robustness against adversarial perturbations.

Features

Implements FGSM, PGD, CW, MIFGSM, and AutoAttack adversarial attack methods.
Works with PyTorch models.
Returns adversarial images, perturbations images.

Installation

You can install advtorchattacks in two ways:

1️ Install from PYPI


pip install advtorchattacks

2 Install from GitHub


https://github.com/santhosh1705kumar/advtorchattacks

Downloads & popularity

Awaiting observations

No metric observations yet

Download counts and popularity will appear here as the collector receives them.

Declared dependencies 4

Native requirements

4 direct targetsNot calculated transitive count

Package / target	Requirement	Relationship
torch Environment { "extras": "", "marker": null, "requires_python": ">=3.7", "evaluation": "NOT_EVALUATED" }	`torch>=1.8.0`	runtime
torchvision Environment { "extras": "", "marker": null, "requires_python": ">=3.7", "evaluation": "NOT_EVALUATED" }	`torchvision`	runtime
numpy Environment { "extras": "", "marker": null, "requires_python": ">=3.7", "evaluation": "NOT_EVALUATED" }	`numpy`	runtime
scipy Environment { "extras": "", "marker": null, "requires_python": ">=3.7", "evaluation": "NOT_EVALUATED" }	`scipy`	runtime