icouckuy
icouckuy
I think supporting additional training data needs more thought and is a lot of work. Lets not fix this on a release version yet
You can install GPFlow 0.5.0 directly from git: https://github.com/GPflow/GPflow/tree/0.5.0
The dev-2.0 branch is not stable enough for production.
If the discrete variables are ordinal then you can treat them as continuous parameter for the modeling. When optimizing the acquisition function you can just evaluate the discrete parameter on...
1. I also would go for one OrdinalRangeParameter to hold both a range(7) or arbitrary list of integers. Maybe we should call it IntegerParameter? > Being able to split up...
There is not yet an ask/tell interface. This might be part of the next version. You can override the BayesianOptimizer class if you want to do something special for know,...
Thanks for the suggestions. - ask-and-tell (I call it the human simulator. i.e., it can also be time-consuming measurements) is definitely an interesting use case that will be considered. At...
What's the status on this? We are doing this for a while, but not sure if all the older code has been updated?
Hi @RafieeAshkan , we are jointly developing Trieste with Secondmind. Trieste is compatible with the latest Tensorflow and GPFlow: https://github.com/secondmind-labs/trieste
I skipped the ModelWrapper class in my review as it is a separate PR #64