Rhys Goodall

Yep, will await new codebase and see if it can be back ported. Is new code also TF based?

Re: deepchem/deepchem#1676: Having looked at the FermiNet code from DM and the PauliNet code here, There's definitely a lot of technical overlap but the engineering/structure of the codebases appears not...

potentially related issues: #170 #184

I am not fully suggesting that we add this to the scikit-learn API yet but more that accumulating more information on the nodes could allow for wrapping the base models...

@jacobrgardner Do you have any suggested literature covering the insight here that; > variational GPs inherently allow for some degree of heteroscedasticity via the variational covariance matrix S. This is...

> @Balandat I'm digging up a really old PR here but it seemed like the right place to ask after doing a bit of searching for the right topic. It...

https://github.com/pytorch/botorch/issues/861 - This is the potential API issue I think is related as when trying to piece this together it links to issues where the noise is still logged internally...

> If anyone needs this model it would be great if they could take a stab at it; we're very happy to provide feedback and input on it! Would you...