Process distributed ML data pipelines with Ray Data across CPU/GPU clusters.
npx skills add https://github.com/ovachiever/droid-tings --skill ray-data
Please help me install this Agent Skill. Skill: ray-data Source: https://github.com/ovachiever/droid-tings/tree/main/skills/ray-data Command: npx skills add https://github.com/ovachiever/droid-tings --skill ray-data
Ray Data offers scalable, distributed data processing for ML pipelines, enabling streaming, multi-modal data loading, and integration with PyTorch/TensorFlow across CPU/GPU clusters.
Read Parquet data, apply a map_batches transformation, and iterate batches.
High-intent search queries and answers about installing and using this skill.
Ray Data enables distributed processing of Parquet files by reading data into batches and applying transformations across a cluster. It scales from a single laptop to hundreds of nodes, handling the distributed coordination automatically.
Ray Data integrates directly with PyTorch and TensorFlow, enabling you to load multi-modal data (Parquet, CSV, JSON, images) and feed batches into training pipelines with GPU acceleration support.
Ray Data supports streaming execution and batched transforms on distributed clusters, allowing you to preprocess images, text, and structured data in parallel before feeding results to inference models.
Ray Data reads Parquet, CSV, JSON, and image files natively. You can apply map_batches transformations and convert between formats as part of your ETL workflow.
Ray Data supports GPU acceleration across streaming and batch pipelines, enabling distributed ETL workloads to leverage GPU resources on clusters ranging from small to hundreds of nodes.
ihatesea69Process large datasets with distributed Ray Data pipelines for ML workloads.
Clay-HHKProcess large-scale ML datasets across clusters for distributed preprocessing and batch inference.
MesferAliProcesses large-scale ML datasets with distributed streaming across CPUs and GPUs.
qinyan-aiScale pandas and NumPy workflows to datasets that exceed memory.
