Tobias Pitters

Thanks for your efforts and this PR, we appreciate it. Due to lack of activity I'll close this for now. Feel free to come back, reopen and finish this whenever...

Thanks for the report, what platform are you using?

I don't have this problem on windows and therefore cannot reproduce it. Can you pls create a new env and show the specific commands you ran to arrive at the...

Thanks for your report. Shap returns values that are compatible with a `predict_proba` call from sklearn. Is something else stated in the README? If so could you please point me...

Ok, sorry for the confusion I created. I did not check that thoroughly, but meant that the output is shap values are generated for each class, similar to how predict_proba...

The problem I see here is that we are not consistent with the output shapes of xgboost, lightgbm and random forest classifiers which might have led to the confusion here.

@SFatemehM thanks for the PR, would like to review this. Do you have a code snippet I could use to test this?

Just looked at this and I think it would be really good if we could test this function. I hope we can somehow write a test that is shorter than...

> > This looks pretty good. Since we do not add a unit test I would like to get the notebook `PyTorch Deep Explainer MNIST example` back to work. You...

> > Just looked at this and I think it would be really good if we could test this function. I hope we can somehow write a test that is...