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

Feature Request: Implement Continuous Bernoulli and Continuous Categorical distributions #21

Open
wdkrnls opened this issue Apr 12, 2020 · 2 comments

Comments

@wdkrnls
Copy link

wdkrnls commented Apr 12, 2020

I'm a big fan of your package 👍. I was curious if you might consider adding two new exponential family distributions for percent or compositional data: the Continuous Bernoulli and Continuous Categorical which have been recently added into tensor flow. The paper for Continuous Categorical is here.

@twolodzko
Copy link
Owner

Thanks, it's nice to hear that you found it helpful. Unfortunately, currently I'm only maintaining the package, so it is unlikely that I'll add any new features. However I'll be happy to receive PR, if you, or someone else, would be willing to contribute.

@wdkrnls
Copy link
Author

wdkrnls commented Apr 13, 2020

Fair enough. Thanks for maintaining 👍. The Continuous Bernoulli is pretty simple to code up naively in base R, but I am not too familiar with Rcpp and am rusty with numerical tricks to minimize floating point issues. The Continuous Categorical distribution is more challenging, but sampling on the Simplex might benefit a lot from Rcpp AFAICT.

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

No branches or pull requests

2 participants