Releases: aws/sagemaker-python-sdk
Releases · aws/sagemaker-python-sdk
v1.39.0
Features
- Estimator.fit like logs for transformer
- handler for stopping transform job
Bug fixes and other changes
- remove hardcoded creds from integ test
- remove hardcoded creds from integ test
- Fix get_image_uri warning log for default xgboost version.
- add enable_network_isolation to generic Estimator class
- use regional endpoint when creating AWS STS client
- update Sagemaker Neo regions
- use cpu_instance_type fixture for stop_transform_job test
- hyperparameter tuning with spot instances and checkpoints
- skip efs and fsx integ tests in all regions
Documentation changes
- clarify some Local Mode limitations
v1.38.6
Bug fixes and other changes
- update: disable efs fsx integ tests in non-pdx regions
- fix canary test failure issues
- use us-east-1 for PR test runs
Documentation changes
- updated description for "accept" parameter in batch transform
v1.38.5
Bug fixes and other changes
- clean up resources created by file system set up when setup fails
v1.38.4
Bug fixes and other changes
- skip EFS tests until they are confirmed fixed.
Documentation changes
- add note to CONTRIBUTING to clarify automated formatting
- add checkpoint section to using_mxnet topic
v1.38.3
Bug fixes and other changes
- change AMI ids in tests to be dynamic based on regions
v1.38.2
Bug fixes and other changes
- skip efs tests in non us-west-2 regions
- refactor tests to use common retry method
v1.38.1
Bug fixes and other changes
- update py2 warning message
- add logic to use asimov image for TF 1.14 py2
Documentation changes
- changed EFS directory path instructions in documentation and Docstrings
v1.38.0
Features
- support training inputs from EFS and FSx
v1.37.2
Bug fixes and other changes
- Add support for Managed Spot Training and Checkpoint support
- Integration Tests now dynamically checks AZs
v1.37.1
Bug fixes and other changes
- eliminate dependency on mnist dataset website
Documentation changes
- refactor using_sklearn and fix minor errors in using_pytorch and using_chainer