Manuel

The following line makes me thing that you are missing the header files for Python (`python3-dev` in Ubuntu; I don't know about Amazon Linux 2). ``` shap/_cext.cc:3:10: fatal error: Python.h:...

For anyone else, quick fix in your own code: ``` import category_encoders as ce class CatBoostEnc(ce.CatBoostEncoder): def fit_transform(self, X, y): return super().fit(X, y).transform(X) cbe = CatBoostEnc() ``` There you have...

I have a question regarding the way 1) of doing this, which I'm going to try since I need to get this working even naively soon. 1) The deprojection function...

Thank you very much @lukicdarkoo! Greatly appreciated.

It worked and my simulation slowed down from > 0.95x to around 0.20x when I subscribe to the topic. I see an obvious increase in the processor usage of the...

Thank you for the suggestion of vectorizing the operation with Numpy. It got me to to somewhere around 0.42x. Surprisingly enough, when I leave it running for one minute and...

I tried it with this line instead: ``` ... explainer = shap.TreeExplainer(model, X[:50], feature_perturbation='interventional', model_output='probability') ... ``` The output is this: ``` What I expect as base value (from labels):...

I just tried the same example in an AWS Jupyter notebook and got the exact same results: ``` What I expect as base value (from labels): 0.186 What I expect...

@slundberg What setup are you using? Maybe I can replicate it to have exact results like yours.

Hi @mvollrath - Why could there be some lag between the publisher and the web bridge? - What kind of throttling are you referring to? To the parameter of the...