A regression task that the y value’s range is around a * 10^-5 to b * 10^-4 and trying to use XGBoost to handle this question.

And the weird thing is, when finished model training, the model predictions on test data are always the same value.

I suspected that the y range is too small, so I do a data preprocessing to multiply 1000 to every y value and after that, the model’s prediction back to normal. It can predict different output for test data, what’s going on here?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.