Xgboost: Unstable Output


#1

Hi,

I am using Xgboost for a Machine Learning task… My outcomes are differing when I save the model and try to load it again to obtain predictions for the dependent variable in validation set.

My validation set is exactly similar to my test set which I have created using 70:30

PS: I am working on Categorical data. I have not used One Hot Encoding.

Please help me identify the reasons for unstable results.