Jacob Schreiber

Yeah, I think this can be improved by first storing each of the distributions as a list and then storing the edges. If you'd like to take a stab at...

Howdy This is the right place to ask. While `sample` is implemented for every other model, there currently isn't an implementation of the `sample` method for Bayesian networks yet. This...

Do you mean only learn the variances and HMM transitions, or only learn the HMM transitions?

Unfortunately, you'll need to write out the HMM by hand (i.e., using the `add_edge` and `add_state` attributes) and pass in the mixture models when specifying the States objects. See the...

Sorry, I meant more that you can see a tutorial for how to write out a HMM using the `add_edge` and `add_state` methods.

Thanks for the report. I've never put pomegranate on conda (someone else must have), but I'll look into fixing it.

Thanks for sharing! I'll look into trying to incorporate this.

I tried getting this to work. Unfortunately, it looks like it no longer displays the image to the screen as most users might expect when they use `model.plot()`. That was...

Sure. In theory, one could come up with a new transition matrix that's based on each example. I've seen cool work on this (on vacation so I can't find the...

When you make predictions, each column of the returned matrix should correspond to the same column in your original data. What do you see as being the additional benefit of...