mc-robinson

I just wanted to note that I have independently come across this bug recently. In particular, this functionality would be really useful when trying to de-salt molecules with enhanced stereochemistry...

Thanks for the detailed responses! And @jacobrgardner, I know it was a mistake haha, but I am also happy to (eventually) try to do a detailed notebook on heteroskedastic GPs...

Hi guys, thanks for the work you've already put into this. I'll try to update the example notebook soon, and I'm happy to try and help build out the heteroskedastic...

Hi all, Sorry for the delay on this one -- but I am finally attaching a slightly updated notebook: https://colab.research.google.com/drive/1osVrSIfnrm7WWXgsYgP1rR6FDukHNjQF In it is a first attempt at a `MostLikelyHeteroskedasticGP` class....

Ok great, thanks for all of the helpful comments. I'll try to update the notebook with these suggestions later today.

Alright, sorry for the delay on this one, but here is the updated notebook: https://colab.research.google.com/drive/1M--JgobEFdSnrke_j8X-kwNM6S-M9BrM I have to say I went down a bit of a rabbit hole trying to...

Thanks! And no I don't think the difference is that big in these small toy cases, but I am looking into the difference between the variance estimators for some more...

Cool, thanks for the advice. I'll try to implement these things and look into the effects of the kernel soon.

Hi, sorry this took so long -- but here is an updated notebook with the implemented "most likely heteroskedastic GP" from the paper https://colab.research.google.com/drive/1dvFA3LYdLcH0ObQhRvnG1yXCUNgeYIXz . The GP is benchmarked on...

Great, thanks for the feedback @Balandat and @eytan. I've been traveling a bit, but let me first try to get a PR going. After that, I'm happy to write up...