sagemaker-python-sdk

Bug Fixes and Other Changes

• TGI 1.3.3
• skip failing integs
• query hf api for model md
• update image_uri_configs 12-29-2023 06:17:34 PST

Features

• support local mode in SageMaker Studio (#1300)
• Supporting tbac in load_run

Bug Fixes and Other Changes

• update image_uri_configs 12-25-2023 06:17:33 PST
• Disable failed test in IR
• Raise Exception for debug
• create role if needed in get_execution_role

Bug Fixes and Other Changes

• update image_uri_configs 12-22-2023 06:17:35 PST
• update model path in local mode
• Using logging instead of prints

Documentation Changes

• update issue template.

Features

• support remote debug for sagemaker training job

Bug Fixes and Other Changes

• update image_uri_configs 12-21-2023 08:32:41 PST
• Update tblib constraint

Features

• Use specific images for SMP v2 jobs
• Added update for model package

Bug Fixes and Other Changes

• Add write permission to job output dirs for remote and step decorator running on non-root job user
• Move func and args serialization of function step to step level

Documentation Changes

• SMP v2 doc updates (#1423)
• fix ModelBuilder sample notebook links

Bug Fixes and Other Changes

• Merge branch 'master-rba' into local_merge
• Fix user agent tag issue
• update image_uri_configs 12-13-2023 14:04:54 PST
• update image_uri_configs 12-13-2023 12:23:06 PST

Deprecations and Removals

• remove explicit partitions key requirement on pysdk side.

Features

• add non-repeating config logger
• Add SageMaker Geospatial verison 1.x images
• TGI 1.2.0 Image Uri
• support model.register() with triton model
• Enable retrieving function step results for local mode

Bug Fixes and Other Changes

• TGI 1.3.1
• excessive jumpstart instance type logging
• Support local mode for remote function
• Session.download_data can not download nested objects
• Fix writing into non-closed file with git clone command
• mitigation of xgboost container incompatibility with new version
• update image and hardware validation with inf and graviton
• remove two setuptools deprecations
• minor jumpstart dev ex improvements
• save utils fix
• Correct DJL neuronx regions
• delete unused file inference-experience-dev-tester.sh
• Fix Experiment Run integ test w.r.t unexpected boto3 version
• Bump test dependencies versions
• fast follow on js uncompressed support - ModelBuilder
• Modify Region List for Neuron Images (HF neuron/neuronx, PT Neuron)

Documentation Changes

• Mention for custom Docker Image

Features

• Update boto3 version to 1.33.3
• Goldfinch InferenceComponent integration
• Add Pipeline step decorator, NotebookJobStep, and scheduler
• ModelBuilder for simplified model testing and deployment

Bug Fixes and Other Changes

• Skip failing integ tests
• chore: add jumpstart gated model integration tests
• disable integration tests for Inference Component based endpoint for non supported regions
• remove unnecessary whl file - Galactus
• refactor endpoint type enums, comments, docstrings, method names…
• Address SA feedback regarding deployment straight to Endpoint Mode - Galactus
• fix HuggingFace GEN2 model deployment arguments
• when customer role does not have permission to read logs from CW, default to standard logging - Galactus
• Add pagination for ListInferenceComponent API responses, address non-blocking comments
• Create CSVSerializerWrapper - Galactus
• Model builder Final Fixes
• remote function include_local_workdir default value
• use copy instead of move in bootstrap script
• WaiterError on failed pipeline execution. results()
• Add a unit test for consistency between step and remote decorator
• merge WorkdirConfig and custom_filter_filter parameters
• Add back mixed step type integ test
• do not delete temp folder generated by sdist
• make collect_parameters decorator as private
• HMAC signing for ModelBuilder Triton python backend

Documentation Changes

• update docstring for Generation two endpoint and validation message
• galactus doc updates
• CustomFileFilter doc updates
• NotebookJobSteps class parameter severity update

Features

• re:Invent 2023 keynote1

Features

• PT2.1 SM Training/Inference DLC Release

Bug Fixes and Other Changes

• Release HuggingFace PT Neuronx training image 1.13.1
• HuggingFace PT Neuronx release in SDK

Features

• inference instance type conditioned on training instance type

Bug Fixes and Other Changes

• improved jumpstart tagging

Bug Fixes and Other Changes

• Allow either instance_type or instance_group to be defined in…
• enhance image_uris unit tests

Features

• jumpstart gated model artifacts
• jumpstart extract generated text from response
• jumpstart contruct payload utility

Bug Fixes and Other Changes

• relax upper bound on urllib in local mode requirements
• bump urllib3 version
• allow smdistributed to be enabled with torch_distributed.
• fix URL links

Documentation Changes

• remove python 2 reference
• update framework version links

Features

• Added register step in Jumpstart model
• jumpstart instance specific metric definitions

Bug Fixes and Other Changes

• Updates for DJL 0.24.0 Release
• use getter for resource-metadata dict
• add method to Model class to check if repack is needed

Features

• jumpstart model artifact instance type variants
• jumpstart instance specific hyperparameters
• Feature Processor event based triggers (#1132)
• Support job checkpoint in remote function
• jumpstart model package arn instance type variants

Bug Fixes and Other Changes

• Fix hyperlinks in feature_processor.scheduler parameter descriptions
• add image_uris_unit_test pytest mark
• bump apache-airflow to v2.7.2
• clone distribution in validate_distribution
• fix flaky Inference Recommender integration tests

Documentation Changes

• Update PipelineModel.register documentation
• specify that input_shape in no longer required for torch 2.0 mod…

Bug Fixes and Other Changes

• update local mode schema
• import error in unsupported js regions
• Update Ec2 instance type to g5.4xlarge in test_huggingface_torch_distributed.py

Features

• jumpstart estimator enable infra check flag
• jumpstart default payloads
• allow non-python files in job dependencies
• allow configuring docker container in local mode

Bug Fixes and Other Changes

• js tagging s3 prefix
• Batch transform: Add support for split_type == "None" in local mode
• use correct line endings and s3 uris on windows
• Fixed bug in _create_training_details
• DJL Neuronx 0.24.0

Documentation Changes

• Include FeatureGroup's load_feature_definitions API documentation

Features

• Selective Step Execution milestone 2 features
• feature-processor extra data sources support

Features

• Add support for in-memory feature groups and collection type features in Feature Store.

Bug Fixes and Other Changes

• chore: xfail resource in use failure for specific test
• Add missing API docs for processors

Documentation Changes

• Bring back (de)serializers documentation
• Add missing AirFlow operators + link to airflow documentation

Features

• add feature processor APIs to public doc
• s3 prefix model data for JumpStartModel
• Model Package support for updating approval

Bug Fixes and Other Changes

• Add bucket owner check
• transform step unit test
• Release TGI 1.1.0 Image