David Walz

I just want to comment that in my field we often have n_observations / n_parameters ~ 1 (ARD=True). In that setting the default priors for the lengthscale are too tight...

Yep, I can confirm that increasing the limit for non-lazy evaluation fixes the issue. Thanks!

Duplicate of #1287.

Mmh I think either this ticket or #1287 should stay open until the issue is fixed without changing gpytorch's defaults.

Did you by chance install "sonnet" from pip? https://pypi.python.org/pypi/sonnet/0.1.6

@hkothari thanks for your implementation! @ev-br as far as I understand the remaining issue is the lack of broadcasting support for evaluating pmf and cdf on tensor-shaped `x`, right?

@CuriousSingularity BoTorch has a flexible support for multiobjective optimization https://github.com/pytorch/botorch.

I cannot reproduce. The gurobipy example you tried first can also run without license, so make that you see gurobi connecting to the cloud. Instead of https://www.gurobi.com/documentation/9.1/quickstart_linux/software_installation_guid.html you can also...

Great! Looking at the paper, it does shed a bad light on the reviewers who apparently didn't even bother checking out the code. I do appreciate your effort , and...