v1.3.0

v1.3.0 (November 22, 2024)

New features

• Add support of PyTorch distributed checkpoints (#269)
• Extend benchmark framework to support distributed checkpoints (#269)
• Add support of distributed training to S3IterableDataset (#269)

Breaking changes

• No breaking changes.

v1.2.7

v1.2.7 (October 29, 2024)

New features

• Add support for CRT retries (awslabs/mountpoint-s3#1069).
• Add support for CopyObject API (#242).

v1.2.6

v1.2.6 (October 9, 2024)

New features

• Add support of PyTorch Lightning checkpoints to benchmark suit (#226).

Bug fixes

• Fix potential race condition while instantiating the S3Client (#237).

Breaking changes

• No breaking changes.

v1.2.5

New features

• Enhanced error logging.
• Support tell for S3writer.
• Path-style addressing support.
• Update crates and Mountpoint dependencies.

Breaking changes

• No breaking changes.

Full Changelog: v1.2.4...v1.2.5

PyPi Project

https://pypi.org/project/s3torchconnector/

v1.2.4

New features

• Update crates and Mountpoint dependencies.

Breaking changes

• No breaking changes.

Full Changelog: v1.2.3...v1.2.4

PyPi Project

https://pypi.org/project/s3torchconnector/

v1.2.3

v1.2.3

New features

• Update S3ClientConfig to pass in the configuration for allowing unsigned requests, under boolean flag unsigned.
• Improve the performance of S3Reader when utilized with pytorch.load by incorporating support for the readinto method.
• [Experimental] Add support for passing an optional custom endpoint to S3LightningCheckpoint constructor method.

Breaking changes

• No breaking changes.

Full Changelog: v1.2.2...v1.2.3

PyPi Project

https://pypi.org/project/s3torchconnector/

v1.2.2

v1.2.2

New features

• Expose a new class, S3ClientConfig, with throughput_target_gbps and part_size parameters of the inner S3 client.

Breaking changes

• No breaking changes.

Full Changelog: v1.2.1...v1.2.2

PyPi Project

https://pypi.org/project/s3torchconnector/

v1.2.1

v1.2.1

Breaking changes

• Separate completely Rust logs and Python logs. Logs from Rust components used for debugging purposes
  are configured through the following environment variables: S3_TORCH_CONNECTOR_DEBUG_LOGS,
  S3_TORCH_CONNECTOR_LOGS_DIR_PATH

Full Changelog: v1.2.0...v1.2.1

PyPi Project

https://pypi.org/project/s3torchconnector/

v1.2.0

v1.2.0 (March 13, 2024)

New features

• Add PyTorch Lightning checkpoints support

Bug Fixes / Improvements

• Fix deadlock when enabling CRT debug logs. Removed former experimental method _enable_debug_logging().
• Refactor User-Agent setup for extensibility.
• Update lightning User-Agent prefix to s3torchconnector/{__version__} (lightning; {lightning.__version__}.

Breaking changes

• No breaking changes.

Full Changelog: v1.1.4...v1.2.0

PyPi Project

https://pypi.org/project/s3torchconnector/

v1.1.4

v1.1.4

New features

• Support for Python 3.12.
• Additional logging when constructing Datasets, and when making requests to S3.
• Provide tooling for running benchmarks for S3 Connector for Pytorch.
• Update crates and Mountpoint dependencies.
• [Experimental] Allow passing in the S3 endpoint URL to Dataset constructors.

Bug Fixes

• HeadObject is no longer called when constructing datasets with from_prefix and seeking relative to end of file.

Breaking changes

• No breaking changes.

Full Changelog: v1.1.3...v1.1.4

PyPi Project

https://pypi.org/project/s3torchconnector/