Marko Toplak

Thanks for the report. In this case the output scores would need to be the same.

It might not be a bug of the model, it could also be a bug in data processing in Orange. When models are saved their domain is saved as well....

Thanks! Oh, yes, I missed Feature constructor, that is certainly the one that introduce variable computation.

Thanks! What you describe is standard Orange behavior :) but there seems to be a bug in the variables constructed by Feature Constructor. This widget was worked on quite a...

Thank your for preparing such an easy test case. I tried your workflow: - with 3.32.0 I can confirm the bug - with the current master branch this bug is...

Why is setting the regularization to very weak insufficient in your case? What is your use case?

Actually, blame Orange for this one. Scikit-learn allows no-regularization. I vote for adding the no regularization option into the widget. I like how explicit that is, and might be useful...

@dsanalytics, thanks for the reminder. But your tone offends me. Who are you to boss us around? What makes you think that you can do that? Have you contributed meaningfully...

Orange3-text feedstock is correct. It specifies `orange3 >=3.25.0`. The problem here appears because you have a non-conda installed orange3 in a conda environment. Still, you shouldn't be seeing this. BTW,...

Thanks, appreciate the bug report. It works for me though. Let's try to find what is different. What is your system locale (which is the language you are using)? Are...