The probabilities variable is not passed to the fitter when running multiple trials.
Added missing function argument
Thanks for the contribution. I think this does indeed fix an issue, but it looks like it breaks one of the HMM test cases. Maybe the test case is wrong. I hope to have a chance to look into it today or in the next couple days and I will update you with what I find out.
Thanks for taking the time to report this!
Well it was more than a day or two until I could look into it, but yes, the test case was written incorrectly. I've merged your fix and a fix for the test in a05072a, and added your name to the list of contributors.
Thanks very much for the patch!
Hi Ryan, glad I could help! Best wishes.