Tim Vink

The more parameters we add to a class, the more intimidating it becomes for a new user to learn it. I prefer a simple knife as a tool. My point:...

A separate class for this makes sense. However, does the problem of having to keep at least one feature for every group occur often enough to outweigh the costs of...

After offline discussion, I understand there are teams that could benefit from this use-case. Separating it from `ShapRFECV`, I think we can manage to keep the most common use-case as...

I just watched a presentation by @nanne-aben on covariate shift that details a different approach: 1) train a resemblance model (he calls it an adversial model) between train and test...

Interesting discussion. Framed slightly differently, you could use adversarial/resemblance modelling to _calibrate_ your model as a last (retraining) step, in order to improve performance in production where there is a...

Hi @priyanshj Can you try using the latest version of matplotlib ? `v3.1.0` is almost 2 years old, the latest version is `v.3.4.3`. You can upgrade using `pip install --upgrade...

Thanks for reporting! We'll look into it somewhere this week.

Hi @bhishanpdl , I also found some time to have a look. Is the issue not being able to see the classes and functions exposed with `dir(probatus)` ? The fact...

> This is taking longer than usual. You might need to provide the dependency resolver with stricter constraints to reduce runtime. See https://pip.pypa.io/warnings/backtracking for guidance. This + the logs showing...

I see how training a resemblance model could help detect seasonality in the data. I think there should be more context provided before offering a tool like this: - Can...