/SynthPop

Code for the article "DATGAN: Integrating Expert Knowledge into Deep Learning for Synthetic Tabular Data"

Primary LanguageJupyter Notebook

DATGAN: Integrating Expert Knowledge into Deep Learning for Synthetic Tabular Data

This is the repository associated with the article "DATGAN: Integrating Expert Knowledge into Deep Learning for Synthetic Tabular Data". All the code that has been developed for this article is given in this repository. If you want to use any piece of code in this repository, feel free to do it. However, please cite our article:

"Lederrey G., Hillel T., Bierlaire M., DATGAN: Integrating Expert Knowledge into Deep Learning for Synthetic Tabular Data, arXiv preprint arXiv:2203.03489 (2022)"

Since this repository is a bit messy, we have created a pypi version of the DATGAN. You can visit the DATGAN Github repository or the pypi page.

Content of the repository

  • code: Repository with all the main code used for the article
    • modules: repository for the core code of the DATGAN and the ML efficacy method
    • notebooks: Jupyter notebooks used for developing the DATGAN
      • datgan_dev: Some notebooks used when developing the DATGAN model
      • results: Final repositories to create the results in the article
      • tests: Notebooks used to assess the synthetic datasets
      • training: Notebooks used to train some models
  • figures: Bunch of figures
  • data (not shown here): Available on demand. Please contact me!

If you have any questions about the article or the code in this repository, please contact Gael Lederrey.