Adrin Jalali
Adrin Jalali
Yeah I'll sync that branch with `main` today.
`sample-props` branch is now synced with `main`.
Why did it not fail when @thomasjpfan pushes? 🤯
I just ran isort on the files, no manual change.
Merge this one?
I changed it to not touch the estimator, and just return the set of parameters with the `'warn'` ones replaced with their default values. It's much less intrusive and much...
The warning should be printed only once with the "default" policy, but it doesn't, and I don't understand why. The [doc says](https://docs.python.org/3/library/warnings.html#the-warnings-filter): > print the first occurrence of matching warnings...
Thanks for resurfacing this @jeremiedbb . I think independent of what we do with the warning, adding the description in the docstring is an improvement. I would however like to...
Just a note that the _if all steps support `sample_weight`_ is not trivial since a meta estimator may accept `**kwargs` and we don't know if `sample_weight` is really supported there...