Sagemaker

Bug Fixes and Other Changes

* local mode deletion of temp files on job end
* Cron expression resetting on update monitor
* added support to update arguments in create_monitoring_schedule

Features

* Add python 3.9 and spark 3.2 support for spark processor
* Adding support for Multi Worker Mirrored Strategy in TF estimator

Bug Fixes and Other Changes

* tag permission issue - remove describe before create

Features

* feature store with_feature_group functionality changes
* Adding support for SageMaker Training Compiler PyTorch 1.13
* support of the intelligent stopping in the tuner
* AutoGluon 0.6.2 image_uris update
* Support for flexible instance types in the HPO
* Add business details and hyper parameters fields and update test_model_card.py

Bug Fixes and Other Changes

* disable the tuner test
* Skip test_run_from_transform_job integ test to unblock python-sdk code pipeline
* Revert "feature: feature store with_feature_group functionality changes"
* advanced inference recommendation jobs parameters check
* make model_config optional when predicted labels are provided for bias detection

Features

* support cluster lifecycle management for Sagemaker EMR step
* Inference recommendation id deployment support

Bug Fixes and Other Changes

* test dub gpu integs with p3
* fix(experiments/run.py): Stop duplication of RUN_TC_TAG on Consecutive Experiment Runs
* Enable load_run without name args in Transform env
* Remove confusing log line emitted during feature group ingestion
* Enable Experiment integ test on beta clients
* Make test_processor_with_role_as_pipeline_parameter more concrete

Documentation Changes

* add security note for the estimator hyperparameter arg
* SageMaker distributed - model parallism library release note
* Add a deprecation note for DetailedProfileConfig

Features

* Display file diff on black-check
* Support for environment variables in the HPO
* Support role as PipelineParameter in Processor class
* Add TrainingImageConfig support for SageMaker training jobs

Bug Fixes and Other Changes

* use FeatureGroup's Session in nonconcurrency ingestion
* Update feature_group.py ingest() description
* Do not use print function. User logger instead
* Add batch_get_record and search API for FeatureStore
* hashing problem for framework processors with identical source dirs