A tabular data workload example built with Ray Train.
In this Workload module, you’ll learn how to scale a tabular XGBoost forest-cover classification pipeline from local training to a distributed Ray Train V2 job on an Anyscale cluster. You’ll ingest the UCI Cover Type dataset, persist train/validation Parquet splits to shared storage, and train/evaluate the model using Ray Datasets and distributed execution.
