Skip to main content

Guides

Working with Antigranular​

Antigranular (AG) is a platform that utilises differential privacy to handle data securely.

Private Python​

AG’s Python client and Jupyter extension can bring your script to life in a secure environment by using a specialised version of Python created by AG This Python version operates under restricted conditions, allowing only methods that guarantee differential privacy. A simple cell magic command %%ag is all you need to transport your code block into a secure and privacy-preserving space. 

Explore the first steps to start working with AG.

Packages and Libraries​

AG supports a collection of external Python libraries for data analysis and privacy. Below is a list of supported libraries and packages. Use the links to access more details about each package. The prefix op_ denotes a Private Python wrapped version of a library.

  • pandas: An adaptable data manipulation library offering efficient data structures and tools for data analysis and manipulation.
  • op_pandas: A wrapped library designed for differentially private data manipulation within the pandas framework. It enhances privacy-preserving techniques and enables privacy-aware data processing.
  • op_diffprivib: A differentially private library based on DiffPrivLib that provides various privacy-preserving algorithms and mechanisms for machine learning and data analysis tasks.
  • op_opendp: A library that offers differentially private data analysis and algorithms based on the OpenDP project. It provides privacy-preserving methods and tools for statistical analysis.
  • op_snsql: A library focused on privacy-preserving SQL execution using the SmartNoise framework.
  • op_snsynth: A library focused on privacy-preserving synthethic data generators for tabular data using the SmartNoise framework.
  • op_opacus: Opacus is a library allowing for differentially private training of PyTorch models.
  • op_tensorflow: Tensorflow Privacy is a library enabling differentially private training of TensorFlow models.
  • op_recordlinkage: The RecordLinkage Python toolkit is a versatile library for efficiently linking and deduplicating records in diverse datasets, offering powerful record linkage capabilities for data integration and quality improvement.
  • op_splink: The Splink Python toolkit is an advanced probabilistic record linkage library that enables accurate and customisable record linkage while preserving data privacy. It is an essential tool for data integration and analysis.