Finn Busch

Hm, maybe my explanation was a bit misleading. My input dimension is N. I want to feed 10 independent samples (so batch size 1?) into the GP. 10 x 1...

I'll post a fully reporducible example later, don't have access to a machine right now. Here's the currently used model, if that helps ``` class IndependentMultitaskGPModelApproximate(gpytorch.models.ApproximateGP): def __init__(self, inducing_points_num, input_dim,...

I see. Unfortunately, that won't really help in my case - I am referring to arxiv articles (e.g. [https://arxiv.org/pdf/2206.08077.pdf](https://arxiv.org/pdf/2206.08077.pdf)) where the pdf file name is pretty much useless, I'd rather...

Any news on this?

Hi Henry, Thank you for the help. I managed to install my C++ library **just** as a module. This leads to the function signatures being `modulename.classname`, because the module is...

I understand. I was attempting to extend your work to non-stationary fields by warping the inputs using parameters that are learnable. Consequently, this will lead to a changing warp function....