Christian Lorentzen

Yes `"native"` is fine, even good! I'm happy with any other default value than `"default"`, which I would consider bad design because it does not tell anything about the actual...

I merge. In case that [#78](https://github.com/scikit-learn/enhancement_proposals/pull/78) concludes to still change the default value, we can do that in a new PR.

@mayer79 Do you want to review?

This functionality is available via `scipy.spatial.distance.cdist(X, Y, "minkowski", p=0.1)` and `KNeighborsClassifier` accepts a callable as parameter. I'm fine with either the current behaviour (not allowing p

What's our conclusion? 1. Do not allow p

@ogrisel What is your favorite option for which you would vote?

I could live with that. What do the others think? @raymondj-pace, @jjerphan, @adrinjalali @eschibli?

For larger pipelines, option 1 is my personal favorite as a user.

Can we close as [SLEP018](https://scikit-learn-enhancement-proposals.readthedocs.io/en/latest/slep018/proposal.html) was accepted?

Based on our example from sklearn.linear_model import [PoissonRegressor](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.PoissonRegressor.html#sklearn.linear_model.PoissonRegressor), see also #21787. ```python import numpy as np from sklearn.compose import ColumnTransformer from sklearn.datasets import fetch_openml from sklearn.linear_model import PoissonRegressor from sklearn.pipeline...