You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Feb 23, 2023. It is now read-only.
I think in previous versions GPyOpt allowed users to tune the parameter acqusition_jitter, which is the degree of exploration of the acquisitions function. In the current version acquisition_jitter is removed and can only be used for modular Bayesian Opitimization.
Does anyone know why they removed it? And also can I still tune jitter without using modular Bayesian Optimization?
The text was updated successfully, but these errors were encountered:
As far as I am aware it wasn't removed, you would still pass it as a named argument to BayesianOptimization object. Does this give you an error, or is there any other reason you think it was removed?
In the documentationm jitter is no longer one of the parameters: [https://gpyopt.readthedocs.io/en/latest/GPyOpt.methods.html].. If I do pass it as argument there is no error, but I don't think it does anything either...
Ah, i see. That must be a side effect of some refactoring we did a while ago, when we separated code that handles optional parameters like that. I'll have a little think of how to bring this part of docs back. But I am pretty sure this parameter is still supported.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
I think in previous versions GPyOpt allowed users to tune the parameter acqusition_jitter, which is the degree of exploration of the acquisitions function. In the current version acquisition_jitter is removed and can only be used for modular Bayesian Opitimization.
Does anyone know why they removed it? And also can I still tune jitter without using modular Bayesian Optimization?
The text was updated successfully, but these errors were encountered: