Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

UniqueCombinations constraint with handling_strategy = 'transform' yields synthetic data with nan values #521

Closed
tjhallum opened this issue Jul 21, 2021 · 1 comment · Fixed by #538
Assignees
Labels
bug Something isn't working
Milestone

Comments

@tjhallum
Copy link

tjhallum commented Jul 21, 2021

Environment Details

SDV version: 0.11.0
Python version: 3.6.8
Operating System: Windows 10

Error Description

When applying a UniqueCombinations constraint with handling_strategy = 'transform', the CTGAN model produces some synthetic observations that exhibit nan values for those columns specified within the constraint (despite training the CTGAN model on nan-free data).

This screen capture illustrates the bug:

image

Steps to reproduce

The full Jupyter Notebook that reproduces this bug is attached (the notebook pulls it's own data from the web):
generate_adversarial_data-ctgan.zip

@csala
Copy link
Contributor

csala commented Aug 2, 2021

A full explanation of what caused this issue has been added in the PR that closed it.

@csala csala added this to the 0.12.0 milestone Aug 2, 2021
@csala csala assigned csala and fealho Aug 2, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants