Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Synthcity: facilitating innovative use cases of synthetic data in different data modalities

About

Synthcity is an open-source software package for innovative use cases of synthetic data in ML fairness, privacy and augmentation across diverse tabular data modalities, including static data, regular and irregular time series, data with censoring, multi-source data, composite data, and more. Synthcity provides the practitioners with a single access point to cutting edge research and tools in synthetic data. It also offers the community a playground for rapid experimentation and prototyping, a one-stop-shop for SOTA benchmarks, and an opportunity for extending research impact. The library can be accessed on GitHub (https://github.com/vanderschaarlab/synthcity) and pip (https://pypi.org/project/synthcity/). We warmly invite the community to join the development effort by providing feedback, reporting bugs, and contributing code.

Zhaozhi Qian, Bogdan-Constantin Cebere, Mihaela van der Schaar• 2023

Related benchmarks

TaskDatasetResultRank
Membership Inference AttackAbalone
AUC-ROC60
10
Membership Inference AttackCA Housing
AUC-ROC0.7
8
Membership Inference AttackCASP
AUC-ROC0.72
8
Membership Inference AttackDiabetes
AUC-ROC0.66
8
Membership Inference AttackFaults
AUC-ROC0.61
8
Synthetic Data Generation (Tabular Classification Utility)Average of 5 datasets
Synthetic Score79.7
5
Synthetic Data Privacy EvaluationMultiple datasets Average
Discriminator AUC0.91
5
Synthetic Data GenerationMultiple benchmark datasets Average
Training Time (s)268.5
5
Tabular Data Synthesis Fidelity5 datasets Average performance
Overall Score0.832
5
Showing 9 of 9 rows

Other info

Follow for update