Can we run XGBoost with Spark on a single node with multiple GPUs?

#1

Are we able to achieve this using the xgboost4j library plus https://xgboost.readthedocs.io/en/latest/gpu/index.html?

Or what might be a better suggested way to do so?

#2

If you are running on a single node I don’t see a reason for using Spark, just use the xgboost executable.