This directory contains examples of running Ray on SageMaker
- hello_ray runs a basic hello world example where after a Ray cluster is created, a remote function is executed to return the IP address of each worker node.
- distributed_xgboost demonstrates how Ray can accelerate training of XGBoost models on large scale data through distributed CPU or GPU training
- pytorch_lightning leverages the RayLightning plugin to accelerate training and tuning of Deep Learning models on PyTorch Lightning
- jax_alpa_language_model shows how Ray with alpa can accelerate training of large scale language models
- distributed_tabnet provides an example of training a TabNet tabular model on a parquet dataset using PyTorch and TensorFlow