Andrew
Andrew
Wondering if a fix for this ever emerged? `num_chains>1` still fails when run in a jupyter notebook.
Ah, bummer, I typically like to build models in jupyter notebooks before migrating to a proper script, guess I'll have to rework my workflow a little. Thanks!
Even with that fix, there are dependency issues on import: `ContextualVersionConflict: (scikit-learn 1.2.2 (/usr/local/lib/python3.10/dist-packages), Requirement.parse('scikit-learn~=1.3'), {'matminer'})`
Curious if this was ever implemented? I still find the default SKLearn optimizer performs better at finding optimal hyperparameters for GPs compared to `fit_gpytorch_mll` or `fit_gpytorch_model`.
Thanks for the timely responses! I was going through my code and found that the data was being scaled in an odd way between iterations and throwing off my hyperparameter...
This would be a good addition. We can probably craft a tutorial around a few other underused features to increase our total coverage as well. We currently don't have a...
Yeah, I can take a stab at it!