API Reference

This section provides comprehensive API references for all privacy-preserving libraries and tools. Each library offers differentially private implementations of popular data science and machine learning frameworks, enabling secure analysis while maintaining data privacy.

Pandas

Differentially private implementation of Pandas with PrivateDataFrame and PrivateSeries for secure data manipulation and analysis.

Diffprivlib

Differentially private version of scikit-learn with privacy-preserving machine learning algorithms and statistical analysis tools.

OpenDP

Powerful library for privacy-preserving data analysis with comprehensive functions and methods for sensitive data protection.

SmartNoise SQL

SQL querying framework for differentially private database analysis and statistical operations.

SmartNoise Synth

Data synthesizer built on OpenDP for generating synthetic datasets with privacy guarantees.

Opacus

Meta-developed library for differentially private training of PyTorch models with PrivacyEngine and data loaders.

TensorFlow Privacy

Google-developed library for training machine learning models with differential privacy using PrivateKerasModel and DP-SGD.

Record Linkage

Comprehensive toolkit for differentially private record linkage with preprocessing, indexing, and comparing capabilities.

SPlink

Probabilistic record linkage tool with blocking rules, comparisons, and linker functionality for accurate data matching.