Open Issues Need Help
View All on GitHubPart of the sai3 project that delivers multi-protocol storage access for AI/ML workflows. This project provides a CLI, along with Rust and Python libraries for AI/ML storage workflows. Supporting S3, File and Azure Blob using the latest Rust SDKs.
Part of the sai3 project that delivers multi-protocol storage access for AI/ML workflows. This project provides a CLI, along with Rust and Python libraries for AI/ML storage workflows. Supporting S3, File and Azure Blob using the latest Rust SDKs.
Part of the sai3 project that delivers multi-protocol storage access for AI/ML workflows. This project provides a CLI, along with Rust and Python libraries for AI/ML storage workflows. Supporting S3, File and Azure Blob using the latest Rust SDKs.
AI Summary: Implement a framework-agnostic AI/ML data loader for the existing S3 Python library. This data loader should randomly access and return batches of objects from an S3 bucket, supporting multiple epochs. The implementation should be compatible with various AI/ML frameworks like PyTorch and TensorFlow.
Part of the sai3 project that delivers multi-protocol storage access for AI/ML workflows. This project provides a CLI, along with Rust and Python libraries for AI/ML storage workflows. Supporting S3, File and Azure Blob using the latest Rust SDKs.
AI Summary: Optimize the `GET` operation in the s3dlio library to improve its performance. This involves investigating and implementing the recommendations from the discussion thread linked in the issue description, aiming to achieve closer to the maximum GET performance observed in comparative testing.
Part of the sai3 project that delivers multi-protocol storage access for AI/ML workflows. This project provides a CLI, along with Rust and Python libraries for AI/ML storage workflows. Supporting S3, File and Azure Blob using the latest Rust SDKs.