v0.5.0

What's New

Features

• Session injection for SigV4 signing — New sagemaker_mlflow.use_session() and set_session() APIs allow callers to inject a custom boto3 Session for per-context credential injection. Thread-safe and asyncio-safe via contextvars. (#25)

• Inference image logging — Added support for logging inference image information alongside log_model, enabling deployment workflows that require image metadata. (#23)

• Python 3.12, 3.13, and 3.14 support — Expanded test matrix and added classifiers for Python 3.12, 3.13, and 3.14. (#27)

Full Changelog

v0.4.0...v0.5.0

v0.4.0

What's Changed

• Replace mlflow with mlflow-skinny, add [full] extra by @hashsaee in #22
• Fix tox.ini: pin mlflow-skinny to match mlflow version in test envs by @hashsaee in #24

New Contributors

• @hashsaee made their first contribution in #22

Full Changelog: v0.3.0...v0.4.0

v0.3.0

What's Changed

• feat: add presigned URL upload support for S3 artifacts by @henishborad in #19
• feat: add workspace and scorer store plugins for MLflow 3.10.0 support by @rd-pong in #20
• Update VERSION and prepare for release by @rd-pong in #21

New Contributors

• @henishborad made their first contribution in #19
• @rd-pong made their first contribution in #20

Full Changelog: v0.2.0...v0.3.0

v0.2.0

What's Changed

• Support Cross-Account Sagemaker MLFlow access using assume_role_arn by @bweigel in #5
• Add support for SageMaker MLflow App resource type by @rrrkharse in #11

New Contributors

• @bweigel made their first contribution in #5
• @rrrkharse made their first contribution in #11

Full Changelog: v0.1.1...v0.2.0

v0.1.1

What's Changed

• feat: add govcloud partition in #9

Full Changelog: v0.1.0...v0.1.1

v0.1.0

Initial source for sagemaker-mlflow (#1)

Co-authored-by: Ananth Bashyam <ananthbashyam1@gmail.com>