Quentin Soubeyran

`scikit-optimize` version 0.8.1 is not compatible with `scikit-learn` 0.24.2, as `scikit-learn` 0.24+ changed some APIs that scikit-optimize uses. You need to either downgrade Scikit-Learn, or update Scikit-optimize to 0.9 from...

Hello, I could not reproduce that using the script below, using either `scikit-optimize==0.8.1` or `master` branch. Could you provide a minimum example producing the bug ? ```python import scipy import...

I still can't reproduce the issue, using your script and provided versions. I see you are using `ipython` from the stacktrace: are you perhaps using `load_ext autoreload` ? This could...

The colab is a good idea ! So .... it's my fault that I couldn't reproduce: I reformatted the code, and in the process did one modification that had an...

#1066 fixes that. You work around the issue by manually specifying `plot_dims` in `plot_evaluations` to all the dimensions of your search space. This will override the logic that skips "constant"...

After testing I can't seem to sidestep the issue either, you'll have to wait for the fix.

With the current code, there is no way to short-hand a `Categorical` of two numbers: even `list` or `numpy.ndarray` object go through the inference that is described for `tuple` objects....

@kernc @glouppe For some reasons CircleCI doesn't let me see the details of the failed tests (I get a "your session has expired" error). If you can tell me what's...

I'm afraid the library hasn't been updated for other a year, so this is unlikely. The last 0.9.0 release is quite dated, and there hasn't been any activity from the...