Adversarial Insight ML

Header 1

Description

AIML, short for Adversarial Insight ML, is your go-to Python package for gauging your machine learning image classification models' resilience against adversarial attacks. With AIML, you can effortlessly test your models against various adversarial examples, receiving crisp, insightful feedback. AIML strives for user-friendliness, making it accessible even to non-technical users. We've meticulously implemented various attack methods into AIML to provide a robust robustness assessment. AIML automatically generates adversarial examples (maliciously altered examples) with each attack method, tests their performance on your model and gives you a detailed result and the image samples for you to look at. Using AIML is as easy as installing it with pip and then calling the 'evaluate' function to get your evaluation results. Check out our repository's README.md for a detailed guide and demos.