Fast, flexible, and scalable data loading for ML training with Ray Data

Data loading and preprocessing can easily become the performance bottleneck in actual ML pipelines. Data preprocessing is also becoming more complex. With Ray Data, data loading can be fast, flexible, and scalable.