Releases: aws/sagemaker-python-sdk
Releases · aws/sagemaker-python-sdk
v1.18.11
Bug fixes and other changes
- local data source relative path includes the first directory
- upgrade pylint and fix tagging with SageMaker models
Documentation changes
- add info about unique job names
v1.18.10
Bug fixes and other changes
- make start time, end time and period configurable in sagemaker.analytics.TrainingJobAnalytics
Documentation changes
- fix typo of argument spelling in linear learner docstrings
v1.18.9.post1
Documentation changes
- spelling error correction
v1.18.9.post0
Documentation changes
- move RL readme content into sphinx project
v1.18.9
Bug fixes
- hyperparameter query failure on script mode estimator attached to complete job
Other changes
- add EI support for TFS framework
Documentation changes
- add third-party libraries sections to using_chainer and using_pytorch topics
v1.18.8
Bug fixes
- fix ECR URI validation
- remove unrestrictive principal * from KMS policy tests.
Documentation changes
- edit description of local mode in overview.rst
- add table of contents to using_chainer topic
- fix formatting for HyperparameterTuner.attach()
v1.18.7
Other changes
- add pytest marks for integ tests using local mode
- add account number and unit tests for govcloud
Documentation changes
- move chainer readme content into sphinx and fix broken link in using_mxnet
v1.18.6.post0
Documentation changes
- add mandatory sagemaker_role argument to Local mode example.
v1.18.6
Changes
- enable new release process
- Update inference pipelines documentation
- Migrate content from workflow and pytorch readmes into sphinx project
- Propagate Tags from estimator to model, endpoint, and endpoint config
Sagemaker Python SDK 1.18.5
- bug-fix: pass kms id as parameter for uploading code with Server side encryption
- feature:
PipelineModel
: Create a Transformer from a PipelineModel - bug-fix:
AlgorithmEstimator
: Make SupportedHyperParameters optional - feature:
Hyperparameter
: Support scaling hyperparameters - doc-fix: Remove duplicate content from main README.rst, /tensorflow/README.rst, and /sklearn/README.rst and add links to readthedocs content