pralab/secml
A Python library for Secure and Explainable Machine Learning
SecML helps data scientists and machine learning engineers assess how robust their AI models are against malicious attacks. You input your trained machine learning model and a dataset, and it shows you how easily an attacker could trick your model or inject bad data. This allows you to understand and improve your model's security before deployment.
191 stars. No commits in the last 6 months. Available on PyPI.
Use this if you need to evaluate the security of your machine learning models against adversarial attempts to fool or corrupt them.
Not ideal if you are looking for a general-purpose machine learning library for building and training models from scratch, as its focus is on security evaluation.
Stars
191
Forks
27
Language
Jupyter Notebook
License
Apache-2.0
Category
Last pushed
Jun 23, 2025
Commits (30d)
0
Dependencies
8
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/pralab/secml"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
google/scaaml
SCAAML: Side Channel Attacks Assisted with Machine Learning
Koukyosyumei/AIJack
Security and Privacy Risk Simulator for Machine Learning (arXiv:2312.17667)
AI-SDC/SACRO-ML
Collection of tools and resources for managing the statistical disclosure control of trained...
oss-slu/mithridatium
Mithridatium is a research-driven project aimed at detecting backdoors and data poisoning in...
matteonerini/pin-side-channel-attacks
Machine Learning for PIN Side-Channel Attacks Based on Smartphone Motion Sensors