XGBoost4j noobie questions: prediction and model IO

Hi all,
sorry that this is very basic, but I followed online tutorials as best I could, and I’m still confused about these two aspects. Basically, they both boil down to: What IS the Booster class?

  1. I’m trying to figure out how to get predictions from my trained model, for individual examples coming in. This tutorial: https://xgboost.readthedocs.io/en/stable/jvm/java_intro.html#prediction uses Booster.predict(DMatrix data). But this tutorial I followed first doesn’t use the Booster class at all: https://github.com/oracle/tribuo/blob/main/tutorials/regression-tribuo-v4.ipynb Would I need to use Model.predict(Example example) then? What exactly is the difference between the two predict methods? Is one of them faster?
  2. I’m also wondering about the Booster class because it is mentioned in the tutorial on model IO, and I don’t know if I can save my model without it (I trained it without a Booster object after all, with the tutorial mentioned above), or if not, how to create the proper Booster object for it.
    I hope someone can help me get started! :slight_smile: Thanks!