Andrei Paleyes
Andrei Paleyes
The notebook became very flaky, almost every other build was failing due to it. So the notebook was removed from tests in #366 . We shall take another look at...
It seems that none of the initial designs checks for constraints. It isn't obvious how to fix it though. For random things are pretty clear: consider points individually, and just...
This issue surfaced from conversation with @henrymoss . Emukit does not support 2-stage optimization procedures. For example, consider the following scenario: a multifidelity function f(x,z) where z is the fidelity,...
Implementation of GPy model wrapper uses private stuff like `model._fixes_` and `model._trigger_params_changed`. Can we avoid using it? If not, should it be public in GPy, and can we contribute this...
Our optimizers have the following signature: def optimize(self, x0: np.array, f: Callable=None, df: Callable=None, f_df: Callable=None) where: - f is the function - df is the gradient of the function...
> Conversion complete: 1 warning; 5 errors; 5 undefined macros[\And, \thisfloatsetup, \FBwidth, \floatbox, \capbeside]; 1 missing file[floatrow.sty] So it couldn't load a package `floatrow`, and a couple of errors came...
**Exact location of issue** https://ar5iv.org/abs/2108.04105 https://arxiv.org/abs/2108.04105 **Problem details** ar5iv worked perfectly on this document, amazing work team! The only issue is that the arXiv original was updated to a second...
Hi! I am trying to setup a Robust optimisation experiment with Ax. There is no tutorial on how to do it, so i pieced something together from unit tests. However,...
This paper from people behind BoTorch suggests LogEI acquisition function that can be used as a stand-in replacement for EI, with the performance being equal or better: https://arxiv.org/abs/2310.20708 I feel...
Apparently `np.bool` alias was deprecated in 1.24 for good: https://numpy.org/devdocs/release/1.24.0-notes.html . Using it now gives an error. We don't use it, but GPy does (https://github.com/SheffieldML/GPy/search?q=np.bool , as of this post...