Christian Lorentzen
Christian Lorentzen
**Preface:** *I still consider this a non-issue because it's mainly about cosmetics. The impact of different quantile methods on real use cases is well within the (usual) statistical uncertainty.* **Main...
@I-Al-Istannen Could you provide a minimal reproducible example that triggers the error?
Please something that can be copied and pasted here. Githib provides code blocks.
Sorry I read and wrote too fast. > This causes crashes when using the Imputer with normal int32 columns backed by numpy What exactly crashes? Calling `transform`? Here, an example...
Might be worth to consider https://github.com/Quantco/glum.
I close as kind of duplicate to #26821. It will not directly solve this issue here, but it will enable to achieve splits stratified by regression targets (which I would...
@jjerphan @Micky774 your inout might be valuable. I would recommend to invest some effort into an analysis of similarities and difference between `scipy.spatial.distance` and `sklearn.metrics.pairwise`. No need to have 2...
#22046 seems like a blocker
Very easy. Basically: - Allow a new loss function - Call inverse link function in `predict` - Think about tests