Verbose for Gaussian Process regression

[ithakis]

Describe the workflow you want to enable

I would like to monitor the progress of my fit function for a Gaussian Process.

Describe your proposed solution

if verbose >= 1 print the parameters "theta" and the log marginal likelihood. Maybe offer the option to return the log likelihood with a dict for the tried thetas.

• In general any verbose would be very welcome. To get a rough estimate of the progress.

Please and thank you.

Describe alternatives you've considered, if relevant

No response

Additional context

No response

[ogrisel]

if verbose >= 1 print the parameters "theta" and the log marginal likelihood.

Feel free to open a PR for this. This should be pretty uncontroversial to review and merge.

Maybe offer the option to return the log likelihood with a dict for the tried thetas.

Assuming those are not numerous values, we could probably store 2 arrays as fitted attributes on the estimator for post-fit inspection.

self.explored_thetas_ = ...
self.explored_thetas_loglikelihoods_ = ...

[ROMEEZHOU]

Hi! Is someone working on this issue? If not, I would be glad to help.

[ROMEEZHOU]

/take