-
第92行的cache_hit_ratio（）函数里面是不是少了一个参数呀
-
This has to be treated differently for regression and classification but is very similar to the `predictive(..)` currently implemented. Basically `laplace.thompson_sample(x, n_samples=1)` should retur…
-
Thompson sampling may be better than UCT.
Convert policy priors into action value priors.
Scale the win rate and initial setup to maximise strength.
Optimistic Thompson scaling may improve perf…
-
We don't have an implementation of Thompson sampling, do we?
-
Hi,
Thanks a lot for developing this very neat package, and making it available to the community.
I had a quick question on Thompson sampling:
It seems that Thompson sampling is mentioned as on…
-
The Thompson sampling implementation is slow! This is due to the random normal number generation process; with many features (or many nodes in the last layer of a neural model), sample vectors need to…
-
It is not clear if uniform sampling is required for horvitz_thompson_estimator, or if just independant draws is enough ??
-
Hi I found a problem in bayesian_optimization.py where it pushes kwargs to AcquisitionOptimizer in wrong way such that GPyOpt users cannot specify which GPy models to use by their own appetite.
1) …
-
We have implemented the EI acquisition function until now.
Further implementation of acquisition functions to do are:
- [ ] PI
- [ ] Thompson sampling
- [ ] UCB
- [ ] GP-UCB
- [ ] Random s…
-
Since a lot of people are working on tuning FPU at the moment and some people are exploring tweaks to the search algorithm I wanted to share a few areas of research I was looking over this evening, in…