tobe

It would be the problem of Python3. We will address this and fix it soon.

Great and thanks for contribution @zrcahyx 👍 I will have a look at this pull-request and test soon.

I remember that the problem of retuning multiple trials is not about running the GaussianProgress in multiple times. If we don't give the new data to run BayesianOptimization, the acquisition...

Yes, you know what I means @zrcahyx . But the acquisition function is continuous which means that if we get the highest point, the second highest point is next to...

Thanks for reporting @ed-word . You are right and we will fix it soon. The original through is to minimize the minus accuracy or maximize the accuracy.

Thanks for reporting and it is really appreciated to add these features for `Advisor`. In order to simplify the implementation and user's interface, we choose the classic UCB acquisition function....

It needs `docker` to start the server. Please install `docker` before running `advisor_admin server start `.

Since we don't want user to install extra dependencies to run server. `advisor_admin server start` will use the pre-built docker image to run. We may add more information for users...

I am using `scikit-learn 0.19.1` and did not have this warning. Can you provide the study configuration(search space) and I can try in local environment? @michellemay

The document from someone's blog https://dmsimard.com/2014/01/01/documentation-for-ceph-rest-api/