Main Repository: Codebase for "Time-series Generative Adversarial Networks (TimeGAN)- (NeurIPS), 2019. Authors: Jinsung Yoon, Daniel Jarrett, Mihaela van der Schaar Paper Link: https://papers.nips.cc/paper/8789-time-series-generative-adversarial-networks
This directory contains testing of TimeGAN for synthetic time-series data generation using a real-world dataset from a wearable device.
- dehydration monitoring data: https://www.mdpi.com/1424-8220/22/5/1887/htm, you can contact the author to get access to the data.
To run the pipeline for training and evaluation on TimeGAN framwork, simply run python3 -m main_timegan.py or see jupyter-notebook tutorial of TimeGAN in tutorial_timegan.ipynb.