An alternative to proprietary privacy auditing tools for statistical and deep learning algorithms. It quantifies privacy risks using membership inference attacks to measure data leakage in trained models.
Tech Stack
Jupyter NotebookAI, ML & Data
Why ml_privacy_meter?
• Quantitative privacy scoring
• Supports membership inference attacks
• Extensible for custom algorithms
Limitations
• High mathematical barrier
• Python-only implementation
• Computationally expensive audits
3/3/2026
Last Update
150
Forks
10
Issues
MIT
License
Financial Leak Detected
Stop the "SaaS Tax"
Your team could be burning cash. Switching to ml_privacy_meter instantly boosts your runway.
Competitor Cost
-$1,440
/ year (est. based on Google Cloud Data Loss Prevention)
