Name of XGBoost parameters in Java

#1

Hi,

I am trying to tune parameters for XGBoost’s regressor, and for another problem which has binary output(binary:hinge). Can somebody please point me to the parameters that I would need to tune?

For now I am trying to tune these params(for both),

  • “eval_metric”
  • “booster”
  • “max_delta_step”
  • “min_child_wt”
  • “subsample”
  • “rounds”
  • “eta”
  • “alpha”
  • “lambda”
  • “gamma”
  • “max_depth”
  • “n_estimators”

Please let me know if these are all the parameters I would need to tune or is there more that can be tuned to give me better results?

Also very importantly, the code is in Java and I would like to know if the name of the parameters are correct(eval_metric, booster, max_delta_step, min_child_wt, subsample, n_estimators, etc)?

Is there a place where I can see the parameters names for Java? I have this question since I sometimes see no effects in changing some parameters.

Thanks,
Ram

#2

Hi - Can you look at the below error in building the Java environment?
OS - macOS
Project - xgboost4j-example
[ ERROR ] Failed to execute goal org.scalastyle:scalastyle-maven-plugin:0.8.0:check (checkstyle) on project xgboost4j-example_2.12: Failed during scalastyle execution : Unable to find configuration file at location scalastyle-config.xml -> [Help 1]

I would appreciate any help in setting up the Java environment for xgboost, thanks, Arun