Releases: aws/sagemaker-python-sdk
Releases · aws/sagemaker-python-sdk
v2.89.0
Features
- Add PT 1.11 support
- add validation specification
Bug Fixes and Other Changes
- repack model locally when local_code local mode
Documentation Changes
- smdmp 1.8.1 release note
v2.88.3
Bug Fixes and Other Changes
- deprecate: Remove deprecated argument s3_data_distribution_type
- Feat/jumpstart model table update
v2.88.2
Bug Fixes and Other Changes
- Automl integ describe job check
- Implement subclass compatibility for workflow pipeline job steps
v2.88.1
Bug Fixes and Other Changes
- Add encryption setting to tar_and_upload_dir method
v2.88.0
v2.87.0
Features
- Add Jumpstart example notebooks
- add Tensorflow and Pytorch version for SM Training Compiler and expand to regular regions
Bug Fixes and Other Changes
- integs for training compiler in non-PDX regions
- TrainingStep cache misses due to timestamp based job name
- retry context delete
- Add more logging when unexpected number of artifacts found
v2.86.2
Bug Fixes and Other Changes
- #using uuid to randomize, otherwise system timestamp is used
v2.86.1
Bug Fixes and Other Changes
- xgboost, sklearn network isolation for jumpstart
Documentation Changes
- fix minor typo
v2.86.0
Features
- Adds Spark Processing Notebook to Notebook Tests
v2.85.0
Features
- update lambda code on pipeline create/update/upsert for Lamb…
- jumpstart model url
- add serverless inference image_uri retrieve support
Bug Fixes and Other Changes
- Add back the Fix for Pipeline variables related customer issues
- Support file URIs in ProcessingStep's code parameter