Scott Linderman
Scott Linderman
Thanks Ziqiang. @mattjj, can you update pip with the latest pylds so that we don't have to install it from source? In the meantime, I'll remove the hips dependency. On...
Ah, good point. The HSMM + SLDS is a relatively untested combination. I'll look into the generation code and see if I can debug it. Best, Scott On Mon, Feb...
@noahcowan1 I just saw this note to myself from 2023. Did you happen to play around with the solver and/or specification of the objective? Maybe we should make a note...
I checked again and this issue seems to be resolved in TFP version 0.25.0.
At this point I think it's fair to assume that numpy < 2.0 will be a requirement of dynamax for the foreseeable future. It doesn't seem to be a high...
Thank you, @osorensen and @thomaspinder! We will start working on these issues and suggestions ASAP, and I'll keep you posted.
Hi @osorensen, thanks for the reminder. @gdalle, sorry for the rendering issue. I thought we had fixed this by pinning to Numpy
Hi @osorensen, sorry for the delay. We pinned Numpy < 2.0 and recreated the documentation, and now almost all of the rendering issues should be fixed. There is a known...
Hi @gdalle, of course. Please see https://github.com/probml/dynamax/releases/tag/0.1.5, which is now available on [PyPI](https://pypi.org/project/dynamax/) as well.
Thank you @osorensen and @gdalle! We are starting to work on these issues and hope to have a response in a couple weeks. We greatly appreciate your suggestions.