Jacob Schreiber

Thanks for the contribution. I'm trying to understand the code now. I'll get back to you soon about this PR.

It would also be helpful if you could add some unit tests to ensure that sampling (with a fixed seed) gives the expected behavior.

That's fine with me. If you could add unit tests and appropriate documentation on your additions as well that would be great.

What is the status of this? Are you looking into adding Gibbs sampling or do you want to move forward with this addition? If so, can you add unit tests?

pomegranate does not support complex numbers.

I've seen this in a few examples. I'm not precisely sure what's happening but suspect that it's a result of the cholesky decomposition I do for the covariance matrix estimation....

Sorry for taking a while to get to this. You've defined your distributions to be a list with a single element in it. You need to define it to be...

Howdy Sorry you've encountered these. I'm not sure where the memory leak is and don't really experience it myself. if you could email me your data and model that'd help...

@yh21equn to confirm, you've downloaded the latest code from GitHub, installed it successfully, and are still seeing these issues?

Sorry you encountered issues. Multivariable data needs to have three dimensions either as a fixed-dimension array with dimensions (n_samples, n_observations, n_dimensions) or as a list of 2D arrays where each...