v2.23.4.post0

Documentation Changes

• update predict_fn implementation for PyTorch EIA 1.5.1

v2.23.4

Bug Fixes and Other Changes

• remove captureWarninig setting

v2.23.3

Bug Fixes and Other Changes

• improve optional dependency error message
• add debugger rule container account in PDT
• assert step execution first in pipeline test
• add service inserted fields to generated Hive DDL

Documentation Changes

• fix description for max_wait
• use correct classpath in V2 alias documentation.
• Bad arg name in feat-store ingestion manager

v2.23.2

Bug Fixes and Other Changes

• remove shell=True in subprocess.check_output
• use SecurityConfig dict key

Documentation Changes

• remove D212 from ignore to comply with PEP257 standards

v2.23.1

Bug Fixes and Other Changes

• update git utils temp file
• Allow online store only FeatureGroups

Documentation Changes

• inform contributors when not to mark integration tests as canaries
• adding change log for smd model parallel

v2.23.0

Features

• Add support for actions in debugger rules.

Bug Fixes and Other Changes

• include sparkml 2.4 in image uri config properly
• Mount metadata dir only if it exists
• allow urllib3 1.26

v2.22.0

Features

• Support local mode for Amazon SageMaker Processing jobs

Bug Fixes and Other Changes

• Add API enhancements for SMP
• adjust naming convention; fix links
• lower value used in featurestore test

Documentation Changes

• Update GTDD instructions

v2.21.0

Features

• remove D205 to enable PEP257 Docstring Conventions

Bug Fixes and Other Changes

• Pin smdebug-rulesconfig to 1.0.0
• use itertuples to ingest pandas dataframe to FeatureStore

v2.20.0

Features

• add dataset definition support for processing jobs

Bug Fixes and Other Changes

• include workflow integ tests with clarify and debugger enabled
• only run DataParallel and EdgePackaging tests in supported regions

Documentation Changes

• fix smp code example, add note for CUDA 11 to sdp
• adding note about CUDA 11 to SMP. Small title update PyTorch

v2.19.0

Features

• add tensorflow 1.15.4 and 2.3.1 as valid versions
• add py36 as valid python version for pytorch 1.6.0
• auto-select container version for p4d and smdistributed
• add edge packaging job support
• Add Clarify Processor, Model Bias, Explainability, and Quality Monitors support. (#494)
• add model parallelism support
• add data parallelism support (#454) (#511)
• support creating and updating profiler in training job (#444) (#526)

Bug Fixes and Other Changes

• bump boto3 and smdebug_rulesconfig versions for reinvent and enable data parallel integ tests
• run UpdateTrainingJob tests only during allowed secondary status
• Remove workarounds and apply fixes to Clarify and MM integ tests
• add p4d to smdataparallel supported instances
• Mount metadata directory when starting local mode docker container
• add integ test for profiler
• Re-enable model monitor integration tests.

Documentation Changes

• add SageMaker distributed libraries documentation
• update documentation for the new SageMaker Debugger APIs
• minor updates to doc strings