XGBOOSTClassifier

[Sbellart-dev]

There is a problem with boolean condition when I want to import a model of XGBOOSTClassifier

I trained a XGBOOSTClassifier with some booleans features (with true and false and not 1 and 0 as possible values)

After the training, XGBOOST didn't registered a "split_condition" that pyxai try to read since it's implicit to check is the bool is set to true or not and pyxai failed to import my model

[szczepanskiNicolas]

Hello @Sbellart-dev

Sorry for the delay in replying.

In PyXAI, we use the version 1.7.3 of xgboost. This may not work if you've used a higher version to create the model. Could you please tell me the version with which the model was created?

We'll make this compatible in a future version of pyxai.

Thank you for your interest in the project.