Hello,
I’m currently working on my undergraduate thesis related to XGBoost. In my thesis, I need to explain in detail the processes behind XGBoost training, ranging from how it builds trees to calculating leaf weights, etc. My question is, since I have set the subsample to 0.7, meaning that each tree randomly selects that portion of the training set on which to train, is there a way to see which samples were randomly selected for each tree?
Thank you.