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

Compatibility with Turing models #76

Open
DominiqueMakowski opened this issue Jul 20, 2024 · 2 comments
Open

Compatibility with Turing models #76

DominiqueMakowski opened this issue Jul 20, 2024 · 2 comments

Comments

@DominiqueMakowski
Copy link

Thanks forthe package!
Any plans to make Effects.jl work with (at least some simple) Turing models?
What would be the things needed for it to work?

@palday
Copy link
Member

palday commented Jul 23, 2024

This might be doable as a package extension. The main problem is that AFAIK Turing models don't support the StatsAPI.RegressionModel interface. The main functions we depend on are:

  • modelmatrix
  • coef
  • formula (to extract the model formula and then use that to generate typical values for marginalized terms)
  • vcov

I suspect that you would actually want to not use vcov directly, but rather do something more like we do in the experimental bootstrap support in the MixedModels extension, but using posterior draws instead of bootstrap replicates.

@DominiqueMakowski
Copy link
Author

Indeed, maybe a good start to add the above methods would be in TuringGLM falls more into a standard regression type.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants