kernc

@pep8speaks in https://github.com/OrkoHunter/pep8speaks/issues/95 says it should follow [setup.cfg flake8 configuration](https://github.com/scikit-optimize/scikit-optimize/pull/1074/files#diff-fa602a8a75dc9dcc92261bac5f533c2a85e34fcceaff63b3a3a81d9acde2fc52), but apparently not the case? I'd kindly motion toggling off the noise in favor of a [more simple lint check](https://github.com/scikit-optimize/scikit-optimize/pull/1074/files#diff-b803fcb7f17ed9235f1e5cb1fcd2f5d3b2838429d4368ae4c57ce4436577f03fR15-R39)....

`n_points` controls the [_beam search_](https://en.wikipedia.org/wiki/Beam_search) _width_. Each computed `n_points` batch "eats into" total `n_iter`. It's not evident that `n_points = cpu_count` should produce the optimal results, only that `n_iter` will...

The fix looks good. Can you also write a simple unit test for it?

`BayesSearchCV` had been [recently revamped](https://github.com/scikit-optimize/scikit-optimize/commit/9461bfecc12c8cab7ae18a55e44e1d258b3c09c9). Make sure you're testing with the git master rather than the currently released version.

I'm not that familiar with that bit of codebase, but why not turn the proposed solution into a proposed [PR](https://docs.github.com/en/github/collaborating-with-issues-and-pull-requests/about-pull-requests) so that it can get more readily reviewed and, if...

Another workaround is to use [`DeltaXStopper`](https://scikit-optimize.github.io/stable/modules/generated/skopt.callbacks.DeltaXStopper.html#skopt.callbacks.DeltaXStopper) callback: ```py gp_minimize(..., callback=DeltaXStopper(1e-8)) ```

If you could provide a simple code snippet that someone can use to reproduce the error, that would be helpful.

> simply check a callable whether a new individual is within the constraints and keep on drafting until that function returns 'True'. Is that correct? With this implementation, as well...

To get 0-100 plus three decimals, you could instead ask for `Integer(0, 100_000)` and manually divide the value by 1000 in your objective function.

FWIW this late, I'm 90% through in https://github.com/kernc/scikit-optimize/pull/3, which is quite severe, but promising. @matthewfeickert Would love your insight or another pair of hands. :sweat_smile: