Changelogs » Sagemaker

PyUp Safety actively tracks 274,040 Python packages for vulnerabilities and notifies you when to upgrade.

Sagemaker

2.18.0

Features
  
  * all de/serializers support content type
  * warn on 'Stopped' (non-Completed) jobs
  * all predictors support serializer/deserializer overrides
  
  Bug Fixes and Other Changes
  
  * v2 upgrade tool should ignore cell starting with '%'
  * use iterrows to iterate pandas dataframe
  * check for distributions in TF estimator
  
  Documentation Changes
  
  * Update link to Sagemaker PyTorch Docker Containers
  * create artifact restricted to SM context note
  
  Testing and Release Infrastructure
  
  * remove flaky assertion in test_integ_history_server
  * adjust assertion of TensorFlow MNIST test

2.17.0

Features
  
  * bump minor version for re:Invent 2020 features

2.16.4

Features
  
  * Add re:Invent 2020 features
  
  Bug Fixes and Other Changes
  
  * use eia python version fixture in integration tests
  * bump version to 2.17.0 for re:Invent-2020
  
  Documentation Changes
  
  * add feature store documentation

2.16.3.post0

Testing and Release Infrastructure
  
  * use ECR-hosted image for ubuntu:16.04

2.16.3

Bug Fixes and Other Changes
  
  * fix failures for multiple spark run() invocations

2.16.2

Bug Fixes and Other Changes
  
  * create default bucket only if needed

2.16.1

Bug Fixes and Other Changes
  
  * ensure 1p algos are compatible with forward-port

2.16.0.post0

Documentation Changes
  
  * clarify non-breaking changes after v1 forward port

2.16.0

Features
  
  * update image uri for neo tensorflow

2.15.4

Bug Fixes and Other Changes
  
  * add kms_key optional arg to Pipeline.deploy()
  
  Documentation Changes
  
  * Debugger API - improve docstrings and add examples

2.15.3

Bug Fixes and Other Changes
  
  * refactor _create_model_request

2.15.2

Bug Fixes and Other Changes
  
  * preserve model_dir bool value
  * refactor out batch transform job input generation

2.15.1

Bug Fixes and Other Changes
  
  * include more notebook tests, logger to warn
  * include managed spot training notebook test
  * add missing account IDs for af-south-1 and eu-south-1

2.15.0

Features
  
  * add network isolation support for PipelineModel
  * forward-port v1 names as deprecated aliases
  
  Bug Fixes and Other Changes
  
  * include additional docstyle improvements
  * check optional keyword before accessing
  * use local updated args; use train_max_wait
  * cross-platform file URI for Processing
  * update kwargs target attribute
  
  Documentation Changes
  
  * fix Spark class links
  * kwargs descriptions include clickable links
  * fix broken link to moved notebook

2.14.0

Features
  
  * upgrade Neo MxNet to 1.7
  
  Bug Fixes and Other Changes
  
  * add a condition to retrieve correct image URI for xgboost

2.13.0

Features
  
  * add xgboost framework version 1.2-1
  
  Bug Fixes and Other Changes
  
  * revert "feature: upgrade Neo MxNet to 1.7 (1928)"

2.12.0

Features
  
  * upgrade Neo MxNet to 1.7

2.11.0

Features
  
  * Add SDK support for SparkML Serving Container version 2.4
  
  Bug Fixes and Other Changes
  
  * pin pytest version <6.1.0 to avoid pytest-rerunfailures breaking changes
  * temporarily skip the MxNet Neo test until we fix them
  
  Documentation Changes
  
  * fix conda setup for docs

2.10.0

Features
  
  * add inferentia pytorch inference container config

2.9.2

Bug Fixes and Other Changes
  
  * allow kms encryption upload for processing

2.9.1

Bug Fixes and Other Changes
  
  * update spark image_uri config with eu-north-1 account

2.9.0

Features
  
  * add MXNet 1.7.0 images
  
  Documentation Changes
  
  * removed Kubernetes workflow content

2.8.0

Features
  
  * add spark processing support to processing jobs
  
  Bug Fixes and Other Changes
  
  * remove DataFrame assert from unrelated test

2.7.0

Features
  
  * reshape Parents into experiment analytics dataframe

2.6.0

Features
  
  * add model monitor image accounts for af-south-1 and eu-south-1
  
  Bug Fixes and Other Changes
  
  * enforce some docstyle conventions
  
  Documentation Changes
  
  * fix CSVSerializer typo in v2.rst

2.5.5

Bug Fixes and Other Changes
  
  * update PyTorch 1.6.0 inference image uri config
  * set use_spot_instances and max_wait as init params from job description
  * run integ tests when image_uri_config jsons are changed
  * Revert "fix: update pytorch inference 1.6 image uri config (1873)"
  * update pytorch inference 1.6 image uri config
  
  Documentation Changes
  
  * fix typo in v2.rst
  
  Testing and Release Infrastructure
  
  * fix PyTorch inference packed model integ test

2.5.4

Bug Fixes and Other Changes
  
  * update max_run_wait to max_wait in v2.rst for estimator parameters
  * Updating regional account ids for af-south-1 and eu-south-1
  * add account ids for af-south-1 and eu-south-1 for debugger rules

2.5.3

Bug Fixes and Other Changes
  
  * Revert "change: update image uri config for pytorch 1.6.0 inference (1864)"
  * update image uri config for pytorch 1.6.0 inference
  * add missing framework version image uri config

2.5.2

Bug Fixes and Other Changes
  
  * refactor normalization of args for processing
  * set TF 2.1.1 as highest py2 version for TF
  * decrease integ test concurrency and increase delay between retries

2.5.1

Bug Fixes and Other Changes
  
  * formatting changes from updates to black

2.5.0

Features
  
  * add mypy tox target
  
  Bug Fixes and Other Changes
  
  * break out methods to get processing arguments
  * break out methods to get train arguments

2.4.2

Bug Fixes and Other Changes
  
  * check ast node on later renamers for cli v2 updater
  
  Documentation Changes
  
  * Clarify removals in v2

2.4.1

Bug Fixes and Other Changes
  
  * update rulesconfig to 0.1.5

2.4

===
  
  Release of sagemaker-sparkml-serving-container, supporting Spark major version 2.4.
  
  
  
  Changelog

2.4.0

Features
  
  * Neo algorithm accounts for af-south-1 and eu-south-1
  
  Bug Fixes and Other Changes
  
  * upgrade pytest and other deps, tox clean-up
  * upgrade airflow to 1.10.11
  * update exception assertion with new api change
  * docs: Add SerDe documentation

2.3

===
  
  Release of sagemaker-sparkml-serving-container, supporting Spark major version 2.3.

2.3.0

Features
  
  * support TF training 2.3
  
  Documentation Changes
  
  * update 1p estimators class description

2.2

===
  
  Initial release of sagemaker-sparkml-serving-container, supporting Spark major version 2.2.

2.2.0

Features
  
  * new 1P algorithm accounts for af-south-1 and eu-south-1
  
  Bug Fixes and Other Changes
  
  * update debugger us-east-1 account
  * docs: Add information on Amazon SageMaker Operators usage in China

2.1.0

Features
  
  * add DLC account numbers for af-south-1 and eu-south-1

2.0.1

Bug Fixes and Other Changes
  
  * use pathlib.PurePosixPath for S3 URLs and Unix paths
  * fix regions for updated RL images
  
  Documentation Changes
  
  * update CHANGELOG to reflect v2.0.0 changes
  
  Testing and Release Infrastructure
  
  * remove v2-incompatible notebooks from notebook build

2.0.0

Breaking Changes
  
  * rename s3_input to TrainingInput
  * Move _NumpyDeserializer to sagemaker.deserializers.NumpyDeserializer
  * rename numpy_to_record_serializer to RecordSerializer
  * Move _CsvDeserializer to sagemaker.deserializers and rename to CSVDeserializer
  * Move _JsonSerializer to sagemaker.serializers.JSONSerializer
  * Move _NPYSerializer to sagemaker.serializers and rename to NumpySerializer
  * Move _JsonDeserializer to sagemaker.deserializers.JSONDeserializer
  * Move _CsvSerializer to sagemaker.serializers.CSVSerializer
  * preserve script path when S3 source_dir is provided
  * use image_uris.retrieve() for XGBoost URIs
  * deprecate sagemaker.amazon.amazon_estimator.get_image_uri()
  * deprecate fw_registry module and use image_uris.retrieve() for SparkML
  * deprecate Python SDK CLI
  * Remove the content_types module
  * deprecate unused parameters
  * deprecate fw_utils.create_image_uri()
  * use images_uris.retrieve() for Debugger
  * deprecate fw_utils.parse_s3_url in favor of s3.parse_s3_url
  * deprecate unused functions from utils and fw_utils
  * Remove content_type and accept parameters from Predictor
  * Add parameters to deploy and remove parameters from create_model
  * Add LibSVM serializer for XGBoost predictor
  * move ShuffleConfig from sagemaker.session to sagemaker.inputs
  * deprecate get_ecr_image_uri_prefix
  * rename estimator.train_image() to estimator.training_image_uri()
  * deprecate is_version_equal_or_higher and is_version_equal_or_lower
  * default wait=True for HyperparameterTuner.fit() and Transformer.transform()
  * remove unused bin/sagemaker-submit file
  
  Features
  
  * start new module for retrieving prebuilt SageMaker image URIs
  * handle separate training/inference images and EI in image_uris.retrieve
  * add support for Amazon algorithms in image_uris.retrieve()
  * Add pandas deserializer
  * Remove LegacySerializer and LegacyDeserializer
  * Add sparse matrix serializer
  * Add v2 SerDe compatability
  * Add JSON Lines serializer
  * add framework upgrade tool
  * add 1p algorithm image_uris migration tool
  * Update migration tool to support breaking changes to create_model
  * support PyTorch 1.6 training
  
  Bug Fixes and Other Changes
  
  * handle named variables in v2 migration tool
  * add modifier for s3_input class
  * add XGBoost support to image_uris.retrieve()
  * add MXNet configuration to image_uris.retrieve()
  * add remaining Amazon algorithms for image_uris.retrieve()
  * add PyTorch configuration for image_uris.retrieve()
  * make image_scope optional for some images in image_uris.retrieve()
  * separate logs() from attach()
  * use image_uris.retrieve instead of fw_utils.create_image_uri for DLC frameworks
  * use images_uris.retrieve() for scikit-learn classes
  * use image_uris.retrieve() for RL images
  * Rename BaseDeserializer.deserialize data parameter
  * Add allow_pickle parameter to NumpyDeserializer
  * Fix scipy.sparse imports
  * Improve code style of SerDe compatibility
  * use image_uris.retrieve for Neo and Inferentia images
  * use generated RL version fixtures and update Ray version
  * use image_uris.retrieve() for ModelMonitor default image
  * use _framework_name for 'protected' attribute
  * Fix JSONLinesDeserializer
  * upgrade TFS version and fix py_versions KeyError
  * Fix PandasDeserializer tests to more accurately mock response
  * don't require instance_type for image_uris.retrieve() if only one option
  * ignore code cells with shell commands in v2 migration tool
  * Support multiple Accept types
  
  Documentation Changes
  
  * fix pip install command
  * document name changes for TFS classes
  * document v2.0.0 changes
  * update KFP full pipeline
  
  Testing and Release Infrastructure
  
  * generate Chainer latest version fixtures from config
  * use generated TensorFlow version fixtures
  * use generated MXNet version fixtures

2.0.0.rc1

Breaking Changes
  
  * Move StreamDeserializer to sagemaker.deserializers
  * Move StringDeserializer to sagemaker.deserializers
  * rename record_deserializer to RecordDeserializer
  * remove "train_" where redundant in parameter/variable names
  * Add BytesDeserializer
  * rename image to image_uri
  * rename image_name to image_uri
  * create new inference resources during model.deploy() and model.transformer()
  * rename session parameter to sagemaker_session in S3 utility classes
  * rename distributions to distribution in TF/MXNet estimators
  * deprecate update_endpoint arg in deploy()
  * create new inference resources during estimator.deploy() or estimator.transformer()
  * deprecate delete_endpoint() for estimators and HyperparameterTuner
  * refactor Predictor attribute endpoint to endpoint_name
  * make instance_type optional for Airflow model configs
  * refactor name of RealTimePredictor to Predictor
  * remove check for Python 2 string in sagemaker.predictor._is_sequence_like()
  * deprecate sagemaker.utils.to_str()
  * drop Python 2 support
  
  Features
  
  * add BaseSerializer and BaseDeserializer
  * add Predictor.update_endpoint()
  
  Bug Fixes and Other Changes
  
  * handle "train_*" renames in v2 migration tool
  * handle image_uri rename for Session methods in v2 migration tool
  * Update BytesDeserializer accept header
  * handle image_uri rename for estimators and models in v2 migration tool
  * handle image_uri rename in Airflow model config functions in v2 migration tool
  * update migration tool for S3 utility functions
  * set _current_job_name and base_tuning_job_name in HyperparameterTuner.attach()
  * infer base name from job name in estimator.attach()
  * ensure generated names are < 63 characters when deploying compiled models
  * add TF migration documentation to error message
  
  Documentation Changes
  
  * update documentation with v2.0.0.rc1 changes
  * remove 'train_*' prefix from estimator parameters
  * update documentation for image_name/image --> image_uri
  
  Testing and Release Infrastructure
  
  * refactor matching logic in v2 migration tool
  * add cli modifier for RealTimePredictor and derived classes
  * change coverage settings to reduce intermittent errors
  * clean up pickle.load logic in integ tests
  * use fixture for Python version in framework integ tests
  * remove assumption of Python 2 unit test runs

2.0.0.rc0

Breaking Changes
  
  * remove estimator parameters for TF legacy mode
  * remove legacy `TensorFlowModel` and `TensorFlowPredictor` classes
  * force image URI to be passed for legacy TF images
  * rename `sagemaker.tensorflow.serving` to `sagemaker.tensorflow.model`
  * require `framework_version` and `py_version` for framework estimator and model classes
  * change `Model` parameter order to make `model_data` optional
  
  Bug Fixes and Other Changes
  
  * add v2 migration tool
  
  Documentation Changes
  
  * update TF documentation to reflect breaking changes and how to upgrade
  * start v2 usage and migration documentation
  
  Testing and Release Infrastructure
  
  * remove scipy from dependencies
  * remove TF from optional dependencies

1.72.0

Features
  
  * Neo: Add Granular Target Description support for compilation
  
  Documentation Changes
  
  * Add xgboost doc on bring your own model
  * fix typos on processing docs

1.71.1

Bug Fixes and Other Changes
  
  * remove redundant information from the user_agent string.
  
  Testing and Release Infrastructure
  
  * use unique model name in TFS integ tests
  * use pytest-cov instead of coverage

1.71.0

Features
  
  * Add mpi support for mxnet estimator api
  
  Bug Fixes and Other Changes
  
  * use 'sagemaker' logger instead of root logger
  * account for "py36" and "py37" in image tag parsing

1.70.2

Bug Fixes and Other Changes
  
  * convert network_config in processing_config to dict
  
  Documentation Changes
  
  * Add ECR URI Estimator example

1.70.1

Bug Fixes and Other Changes
  
  * Nullable fields in processing_config

1.70.0

Features
  
  * Add model monitor support for us-gov-west-1
  * support TFS 2.2
  
  Bug Fixes and Other Changes
  
  * reshape Artifacts into data frame in ExperimentsAnalytics
  
  Documentation Changes
  
  * fix MXNet version info for requirements.txt support

1.69.0

Features
  
  * Add ModelClientConfig Fields for Batch Transform
  
  Documentation Changes
  
  * add KFP Processing component

1.68.0

Features
  
  * add spot instance support for AlgorithmEstimator
  
  Documentation Changes
  
  * add xgboost documentation for inference

1.67.1.post0

Documentation Changes
  
  * add Step Functions SDK info

1.67.1

Bug Fixes and Other Changes
  
  * add deprecation warnings for estimator.delete_endpoint() and tuner.delete_endpoint()

1.67.0

Features
  
  * Apache Airflow integration for SageMaker Processing Jobs
  
  Bug Fixes and Other Changes
  
  * fix punctuation in warning message
  
  Testing and Release Infrastructure
  
  * address warnings about pytest custom marks, error message checking, and yaml loading
  * mark long-running cron tests
  * fix tox test dependencies and bump coverage threshold to 86%

1.66.0

Features
  
  * add 3.8 as supported python version
  
  Testing and Release Infrastructure
  
  * upgrade airflow to latest stable version
  * update feature request issue template

1.65.1.post1

Testing and Release Infrastructure
  
  * add py38 to buildspecs

1.65.1.post0

Documentation Changes
  
  * document that Local Mode + local code doesn't support dependencies arg
  
  Testing and Release Infrastructure
  
  * upgrade Sphinx to 3.1.1

1.65.1

Bug Fixes and Other Changes
  
  * remove include_package_data=True from setup.py
  
  Documentation Changes
  
  * add some clarification to Processing docs
  
  Testing and Release Infrastructure
  
  * specify what kinds of clients in PR template

1.65.0

Features
  
  * support for describing hyperparameter tuning job
  
  Bug Fixes and Other Changes
  
  * update distributed GPU utilization warning message
  * set logs to False if wait is False in AutoML
  * workflow passing spot training param to training job

1.64.1

Bug Fixes and Other Changes
  
  * include py38 tox env and some dependency upgrades

1.64.0

Features
  
  * add support for SKLearn 0.23

1.63.0

Features
  
  * Allow selecting inference response content for automl generated models
  * Support for multi variant endpoint invocation with target variant param
  
  Documentation Changes
  
  * improve docstring and remove unavailable links

1.62.0

Features
  
  * Support for multi variant endpoint invocation with target variant param
  
  Bug Fixes and Other Changes
  
  * Revert "feature: Support for multi variant endpoint invocation with target variant param (1571)"
  * make instance_type optional for prepare_container_def
  * docs: workflows navigation
  
  Documentation Changes
  
  * fix typo in MXNet documentation

1.61.0

Features
  
  * Use boto3 DEFAULT_SESSION when no boto3 session specified.
  
  Bug Fixes and Other Changes
  
  * remove v2 Session warnings
  * upgrade smdebug-rulesconfig to 0.1.4
  * explicitly handle arguments in create_model for sklearn and xgboost

1.60.2

Bug Fixes and Other Changes
  
  * [doc] Added Amazon Components for Kubeflow Pipelines

1.60.1.post0

Documentation Changes
  
  * clarify that entry_point must be in the root of source_dir (if applicable)

1.60.1

Bug Fixes and Other Changes
  
  * refactor the navigation
  
  Documentation Changes
  
  * fix undoc directive; removes extra tabs

1.60.0.post0

Documentation Changes
  
  * remove some duplicated documentation from main README
  * fix TF requirements.txt documentation

1.60.0

Features
  
  * support TensorFlow training 2.2
  
  Bug Fixes and Other Changes
  
  * blacklist unknown xgboost image versions
  * use format strings instead of os.path.join for S3 URI in S3Downloader
  
  Documentation Changes
  
  * consolidate framework version and image information

1.59.0

Features
  
  * MXNet elastic inference support
  
  Bug Fixes and Other Changes
  
  * add Batch Transform data processing options to Airflow config
  * add v2 warning messages
  * don't try to use local output path for KMS key in Local Mode
  
  Documentation Changes
  
  * add instructions for how to enable 'local code' for Local Mode

1.58.4

Bug Fixes and Other Changes
  
  * update AutoML default max_candidate value to use the service default
  * add describe_transform_job in session class
  
  Documentation Changes
  
  * clarify support for requirements.txt in Tensorflow docs
  
  Testing and Release Infrastructure
  
  * wait for DisassociateTrialComponent to take effect in experiment integ test cleanup

1.58.3

Bug Fixes and Other Changes
  
  * update DatasetFormat key name for sagemakerCaptureJson
  
  Documentation Changes
  
  * update Processing job max_runtime_in_seconds docstring

1.58.2.post0

Documentation Changes
  
  * specify S3 source_dir needs to point to a tar file
  * update PyTorch BYOM topic

1.58.2

Bug Fixes and Other Changes
  
  * address flake8 error

1.58.1

Bug Fixes and Other Changes
  
  * upgrade boto3 to 1.13.6

1.58.0

Features
  
  * support inter container traffic encryption for processing jobs
  
  Documentation Changes
  
  * add note that v2.0.0 plans have been posted

1.57.0

Features
  
  * add tensorflow training 1.15.2 py37 support
  * PyTorch 1.5.0 support

1.56.3

Bug Fixes and Other Changes
  
  * update xgboost latest image version

1.56.2

Bug Fixes and Other Changes
  
  * training_config returns MetricDefinitions
  * preserve inference script in model repack.
  
  Testing and Release Infrastructure
  
  * support Python 3.7

1.56.1.post1

Documentation Changes
  
  * document model.tar.gz structure for MXNet and PyTorch
  * add documentation for EstimatorBase parameters missing from docstring

1.56.1.post0

Testing and Release Infrastructure
  
  * add doc8 check for documentation files

1.56.1

Bug Fixes and Other Changes
  
  * add super() call in Local Mode DataSource subclasses
  * fix xgboost image incorrect latest version warning
  * allow output_path without trailing slash in Local Mode training jobs
  * allow S3 folder input to contain a trailing slash in Local Mode
  
  Documentation Changes
  
  * Add namespace-based setup for SageMaker Operators for Kubernetes
  * Add note about file URLs for Estimator methods in Local Mode

1.56.0

Features
  
  * add EIA support for TFS 1.15.0 and 2.0.0
  
  Bug Fixes and Other Changes
  
  * use format strings intead of os.path.join for Unix paths for Processing Jobs

1.55.4

Bug Fixes and Other Changes
  
  * use valid encryption key arg for S3 downloads
  * update sagemaker pytorch containers to external link
  * allow specifying model name when creating a Transformer from an Estimator
  * allow specifying model name in create_model() for TensorFlow, SKLearn, and XGBoost
  * allow specifying model name in create_model() for Chainer, MXNet, PyTorch, and RL
  
  Documentation Changes
  
  * fix wget endpoints
  * add Adobe Analytics; upgrade Sphinx and docs environment
  * Explain why default model_fn loads PyTorch-EI models to CPU by default
  * Set theme in conf.py
  * correct transform()'s wait default value to "False"
  
  Testing and Release Infrastructure
  
  * move unit tests for updating an endpoint to test_deploy.py
  * move Neo unit tests to a new file and directly use the Model class
  * move Model.deploy unit tests to separate file
  * add Model unit tests for delete_model and enable_network_isolation
  * skip integ tests in PR build if only unit tests are modified
  * add Model unit tests for prepare_container_def and _create_sagemaker_model
  * use Model class for model deployment unit tests
  * split model unit tests by Model, FrameworkModel, and ModelPackage
  * add Model unit tests for all transformer() params
  * add TF batch transform integ test with KMS and network isolation
  * use pytest fixtures in batch transform integ tests to train and upload to S3 only once
  * improve unit tests for creating Transformers and transform jobs
  * add PyTorch + custom model bucket batch transform integ test

1.55.3

Bug Fixes and Other Changes
  
  * remove .strip() from batch transform
  * allow model with network isolation when creating a Transformer from an Estimator
  * add enable_network_isolation to EstimatorBase

1.55.2

Bug Fixes and Other Changes
  
  * use .format instead of os.path.join for Processing S3 paths.
  
  Testing and Release Infrastructure
  
  * use m5.xlarge instances for "ap-northeast-1" region integ tests.

1.55.1

Bug Fixes and Other Changes
  
  * correct local mode behavior for CN regions

1.55.0.post0

Documentation Changes
  
  * fix documentation to provide working example.
  * add documentation for XGBoost
  * Correct comment in SKLearn Estimator about default Python version
  * document inferentia supported version
  * Merge Amazon Sagemaker Operators for Kubernetes and Kubernetes Jobs pages
  
  Testing and Release Infrastructure
  
  * turn on warnings as errors for docs builds

1.55.0

Features
  
  * support cn-north-1 and cn-northwest-1

1.54.0

Features
  
  * inferentia support

1.53.0

Features
  
  * Allow setting S3 endpoint URL for Local Session
  
  Bug Fixes and Other Changes
  
  * Pass kwargs from create_model to Model constructors
  * Warn if parameter server is used with multi-GPU instance

1.52.1

Bug Fixes and Other Changes
  
  * Fix local _SageMakerContainer detached mode (aws1374)

1.52.0.post0

Documentation Changes
  
  * Add docs for debugger job support in operator

1.52.0

Features
  
  * add us-gov-west-1 to neo supported regions

1.51.4

Bug Fixes and Other Changes
  
  * Check that session is a LocalSession when using local mode
  * add tflite to Neo-supported frameworks
  * ignore tags with 'aws:' prefix when creating an EndpointConfig based on an existing one
  * allow custom image when calling deploy or create_model with various frameworks
  
  Documentation Changes
  
  * fix description of default model_dir for TF
  * add more details about PyTorch eia

1.51.3

Bug Fixes and Other Changes
  
  * make repack_model only removes py file when new entry_point provided

1.51.2

Bug Fixes and Other Changes
  
  * handle empty inputs/outputs in ProcessingJob.from_processing_name()
  * use DLC images for GovCloud
  
  Testing and Release Infrastructure
  
  * generate test job name at test start instead of module start

1.51.1

Bug Fixes and Other Changes
  
  * skip pytorch ei test in unsupported regions
  
  Documentation Changes
  
  * correct MultiString/MULTI_STRING docstring

1.51.0

Features
  
  * pytorch 1.3.1 eia support
  
  Documentation Changes
  
  * Update Kubernetes Operator default tag
  * improve docstring for tuner.best_estimator()

1.50.18.post0

Documentation Changes
  
  * correct Estimator code_location default S3 path

1.50.18

Bug Fixes and Other Changes
  
  * change default compile model max run to 15 mins

1.50.17.post0

Testing and Release Infrastructure
  
  * fix PR builds to run on changes to their own buildspecs
  * programmatically determine partition based on region

1.50.17

Bug Fixes and Other Changes
  
  * upgrade framework versions

1.50.16

Bug Fixes and Other Changes
  
  * use sagemaker_session when initializing Constraints and Statistics
  * add sagemaker_session parameter to DataCaptureConfig
  * make AutoML.deploy use self.sagemaker_session by default
  
  Testing and Release Infrastructure
  
  * unset region during integ tests
  * use sagemaker_session fixture in all Airflow tests
  * remove remaining TF legacy mode integ tests

1.50.15

Bug Fixes and Other Changes
  
  * enable Neo integ tests

1.50.14.post0

Testing and Release Infrastructure
  
  * remove TF framework mode notebooks from PR build
  * don't create docker network for all integ tests

1.50.14

Bug Fixes and Other Changes
  
  * don't use os.path.join for S3 path when repacking TFS model
  * dynamically determine AWS domain based on region

1.50.13

Bug Fixes and Other Changes
  
  * allow download_folder to download file even if bucket is more restricted
  
  Testing and Release Infrastructure
  
  * configure pylint to recognize boto3 and botocore as third-party imports
  * add multiple notebooks to notebook PR build

1.50.12

Bug Fixes and Other Changes
  
  * enable network isolation for amazon estimators
  
  Documentation Changes
  
  * clarify channel environment variables in PyTorch documentation

1.50.11

Bug Fixes and Other Changes
  
  * fix HyperparameterTuner.attach for Marketplace algorithms
  * move requests library from required packages to test dependencies
  * create Session or LocalSession if not specified in Model
  
  Documentation Changes
  
  * remove hardcoded list of target devices in compile()
  * Fix typo with SM_MODEL_DIR, missing quotes

1.50.10.post0

Documentation Changes
  
  * add documentation guidelines to CONTRIBUTING.md
  * Removed section numbering

1.50.10

Bug Fixes and Other Changes
  
  * remove NEO_ALLOWED_TARGET_INSTANCE_FAMILY

1.50.9.post0

Documentation Changes
  
  * remove labels from issue templates

1.50.9

Bug Fixes and Other Changes
  
  * account for EI and version-based ECR repo naming in serving_image_uri()
  
  Documentation Changes
  
  * correct broken AutoML API documentation link
  * fix MXNet version lists

1.50.8

Bug Fixes and Other Changes
  
  * disable Debugger defaults in unsupported regions
  * modify session and kms_utils to check for S3 bucket before creation
  * update docker-compose and PyYAML dependencies
  * enable smdebug for Horovod (MPI) training setup
  * create lib dir for dependencies safely (only if it doesn't exist yet).
  * create the correct session for MultiDataModel
  
  Documentation Changes
  
  * update links to the local mode notebooks examples.
  * Remove outdated badges from README
  * update links to TF notebook examples to link to script mode examples.
  * clean up headings, verb tenses, names, etc. in MXNet overview
  * Update SageMaker operator Helm chart installation guide
  
  Testing and Release Infrastructure
  
  * choose faster notebook for notebook PR build
  * properly fail PR build if has-matching-changes fails
  * properly fail PR build if has-matching-changes fails

1.50.7

Bug fixes and other changes
  
  * do not use script for TFS when entry_point is not provided
  * remove usage of pkg_resources
  * update py2 warning message since python 2 is deprecated
  * cleanup experiments, trials, and trial components in integ tests

1.50.6.post0

Documentation changes
  
  * add additional information to Transformer class transform function doc string

1.50.6

Bug fixes and other changes
  
  * Append serving to model framework name for PyTorch, MXNet, and TensorFlow

1.50.5

Bug fixes and other changes
  
  * Use serving_image_uri for Airflow
  
  Documentation changes
  
  * revise Processing docstrings for formatting and class links
  * Add processing readthedocs

1.50.4

Bug fixes and other changes
  
  * Remove version number from default version comment
  * remove remaining instances of python-dateutil pin
  * upgrade boto3 and remove python-dateutil pin
  
  Documentation changes
  
  * Add issue templates and configure issue template chooser
  * Update error type in delete_endpoint docstring
  * add version requirement for using "requirements.txt" when serving an MXNet model
  * update container dependency versions for MXNet and PyTorch
  * Update supported versions of PyTorch

1.50.3

Bug fixes and other changes
  
  * ignore private Automatic Model Tuning hyperparameter when attaching AlgorithmEstimator
  
  Documentation changes
  
  * add Debugger API docs

1.50.2

Bug fixes and other changes
  
  * add tests to quick canary
  * honor 'wait' flag when updating endpoint
  * add default framework version warning message in Model classes
  * Adding role arn explanation for sagemaker role
  * allow predictor to be returned from AutoML.deploy()
  * add PR checklist item about unique_name_from_base()
  * use unique_name_from_base for multi-algo tuning test
  * update copyright year in license header
  
  Documentation changes
  
  * add version requirement for using "requirement.txt" when serving a PyTorch model
  * add SageMaker Debugger overview
  * clarify requirements.txt usage for Chainer, MXNet, and Scikit-learn
  * change "associate" to "create" for OpenID connector
  * fix typo and improve clarity on installing packages via "requirements.txt"

1.50.1

Bug fixes and other changes
  
  * fix PyTorchModel deployment crash on Windows
  * make PyTorch empty framework_version warning include the latest PyTorch version

1.50.0

Features
  
  * allow disabling debugger_hook_config
  
  Bug fixes and other changes
  
  * relax urllib3 and requests restrictions.
  * Add uri as return statement for upload_string_as_file_body
  * refactor logic in fw_utils and fill in docstrings
  * increase poll from 5 to 30 for DescribeEndpoint lambda.
  * fix test_auto_ml tests for regions without ml.c4.xlarge hosts.
  * fix test_processing for regions without m4.xlarge instances.
  * reduce test's describe frequency to eliminate throttling error.
  * Increase number of retries when describing an endpoint since tf-2.0 has larger images and takes longer to start.
  
  Documentation changes
  
  * generalize Model Monitor documentation from SageMaker Studio tutorial

1.49.0

Features
  
  * Add support for TF-2.0.0.
  * create ProcessingJob from ARN and from name
  
  Bug fixes and other changes
  
  * Make tf tests tf-1.15 and tf-2.0 compatible.
  
  Documentation changes
  
  * add Model Monitor documentation
  * add link to Amazon algorithm estimator parent class to clarify **kwargs

1.48.1

Bug fixes and other changes
  
  * use name_from_base in auto_ml.py but unique_name_from_base in tests.
  * make test's custom bucket include region and account name.
  * add Keras to the list of Neo-supported frameworks
  
  Documentation changes
  
  * add link to parent classes to clarify **kwargs
  * add link to framework-related parent classes to clarify **kwargs

1.48.0

Features
  
  * allow setting the default bucket in Session
  
  Bug fixes and other changes
  
  * set integration test parallelization to 512
  * shorten base job name to avoid collision
  * multi model integration test to create ECR repo with unique names to allow independent parallel executions

1.47.1

Bug fixes and other changes
  
  * Revert "feature: allow setting the default bucket in Session (1168)"
  
  Documentation changes
  
  * add AutoML README
  * add missing classes to API docs

1.47.0

Features
  
  * allow setting the default bucket in Session
  
  Bug fixes and other changes
  
  * allow processing users to run code in s3

1.46.0

Features
  
  * support Multi-Model endpoints
  
  Bug fixes and other changes
  
  * update PR template with items about tests, regional endpoints, and API docs

1.45.2

Bug fixes and other changes
  
  * modify schedule cleanup to abide by latest validations
  * lower log level when getting execution role from a SageMaker Notebook
  * Fix "ValueError: too many values to unpack (expected 2)" is occurred in windows local mode
  * allow ModelMonitor and Processor to take IAM role names (in addition to ARNs)
  
  Documentation changes
  
  * mention that the entry_point needs to be named inference.py for tfs

1.45.1

Bug fixes and other changes
  
  * create auto ml job for tests that based on existing job
  * fixing py2 support for latest TF version
  * fix tags in deploy call for generic estimators
  * make multi algo integration test assertion less specific

1.45.0

Features
  
  * add support for TF 1.15.0, PyTorch 1.3.1 and MXNet 1.6rc0.
  * add S3Downloader.list(s3_uri) functionality
  * introduce SageMaker AutoML
  * wrap up Processing feature
  * add a few minor features to Model Monitoring
  * add enable_sagemaker_metrics flag
  * Amazon SageMaker Model Monitoring
  * add utils.generate_tensorboard_url function
  * Add jobs list to Estimator
  
  Bug fixes and other changes
  
  * remove unnecessary boto model files
  * update boto version to >=1.10.32
  * correct Debugger tests
  * fix bug in monitor.attach() for empty network_config
  * Import smdebug_rulesconfig from PyPI
  * bump the version to 1.45.0 (publishes 1.46.0) for re:Invent-2019
  * correct AutoML imports and expose current_job_name
  * correct Model Monitor eu-west-3 image name.
  * use DLC prod images
  * remove unused env variable for Model Monitoring
  * aws model update
  * rename get_debugger_artifacts to latest_job_debugger_artifacts
  * remove retain flag from update_endpoint
  * correct S3Downloader behavior
  * consume smdebug_ruleconfig .whl for ITs
  * disable DebuggerHook and Rules for TF distributions
  * incorporate smdebug_ruleconfigs pkg until availability in PyPI
  * remove pre/post scripts per latest validations
  * update rules_config .whl
  * remove py_version from SKLearnProcessor
  * AutoML improvements
  * stop overwriting custom rules volume and type
  * fix tests due to latest server-side validations
  * Minor processing changes
  * minor processing changes (instance_count + docs)
  * update api to latest
  * Eureka master
  * Add support for xgboost version 0.90-2
  * SageMaker Debugger revision
  * Add support for SageMaker Debugger [WIP]
  * Fix linear learner crash when num_class is string and predict type is `multiclass_classifier`
  * Additional Processing Jobs integration tests
  * Migrate to updated Processing Jobs API
  * Processing Jobs revision round 2
  * Processing Jobs revision
  * remove instance_pools parameter from tuner
  * Multi-Algorithm Hyperparameter Tuning Support
  * Import Processors in init files
  * Remove SparkML Processors and corresponding unit tests
  * Processing Jobs Python SDK support

1.44.4

Bug fixes and other changes
  
  * Documentation for Amazon Sagemaker Operators

1.44.3

Bug fixes and other changes
  
  * move sagemaker config loading to LocalSession since it is only used for local code support.
  
  Documentation changes
  
  * fix docstring wording.

1.44.2

Bug fixes and other changes
  
  * add pyyaml dependencies to the required list.
  
  Documentation changes
  
  * Correct info on code_location parameter

1.44.1

Bug fixes and other changes
  
  * Remove local mode dependencies from required.

1.44.0

Features
  
  * separating sagemaker dependencies into more use case specific installable components.
  
  Bug fixes and other changes
  
  * remove docker-compose as a required dependency.

1.43.5

Bug fixes and other changes
  
  * remove red from possible colors when streaming logs

1.43.4.post1

Documentation changes
  
  * clarify that source_dir can be an S3 URI

1.43.4.post0

Documentation changes
  
  * clarify how to use parameter servers with distributed MXNet training

1.43.4

Bug fixes and other changes
  
  * use regional endpoint for STS in builds and tests
  
  Documentation changes
  
  * update link to point to ReadTheDocs

1.43.3

Bug fixes and other changes
  
  * exclude regions for P2 tests

1.43.2

Bug fixes and other changes
  
  * add support for me-south-1 region

1.43.1

Bug fixes and other changes
  
  * validation args now use default framework_version for TensorFlow

1.43.0

Features
  
  * Add support for PyTorch 1.2.0

1.42.9

Bug fixes and other changes
  
  * use default bucket for checkpoint_s3_uri integ test
  * use sts regional endpoint when creating default bucket
  * use us-west-2 endpoint for sts in buildspec
  * take checkpoint_s3_uri and checkpoint_local_path in Framework class

1.42.8

Bug fixes and other changes
  
  * add kwargs to create_model for 1p to work with kms

1.42.7

Bug fixes and other changes
  
  * paginating describe log streams

1.42.6.post0

Documentation changes
  
  * model local mode

1.42.6

Bug fixes and other changes
  
  * update tfs documentation for requirements.txt
  * support content_type in FileSystemInput
  * allowing account overrides in special regions

1.42.5

Bug fixes and other changes
  
  * update using_mxnet.rst

1.42.4

Bug fixes and other changes
  
  * Revert "fix issue-987 error by adding instance_type in endpoint_name (1058)"
  * fix issue-987 error by adding instance_type in endpoint_name

1.42.3

Bug fixes and other changes
  
  * preserve EnableNetworkIsolation setting in attach
  * enable kms support for repack_model
  * support binary by NoneSplitter.
  * stop CI unit test code checks from running in parallel

1.42.2

Bug fixes and other changes
  
  * re-enable airflow_config tests

1.42.1

Bug fixes and other changes
  
  * lazy import of tensorflow module
  * skip airflow_config tests as they're blocking the release build
  * skip lda tests in regions that does not support it.
  * add airflow_config tests to canaries
  * use correct STS endpoint for us-iso-east-1

1.42.0

Features
  
  * add estimator preparation to airflow configuration
  
  Bug fixes and other changes
  
  * correct airflow workflow for BYO estimators.

1.41.0

Features
  
  * enable sklearn for network isolation mode

1.40.2

Bug fixes and other changes
  
  * use new ECR images in us-iso-east-1 for TF and MXNet

1.40.1

Bug fixes and other changes
  
  * expose kms_key parameter for deploying from training and hyperparameter tuning jobs
  
  Documentation changes
  
  * Update sklearn default predict_fn

1.40.0

Features
  
  * add support to TF 1.14 serving with elastic accelerator.

1.39.4

Bug fixes and other changes
  
  * pass enable_network_isolation when creating TF and SKLearn models

1.39.3

Bug fixes and other changes
  
  * expose vpc_config_override in transformer() methods
  * use Estimator.create_model in Estimator.transformer

1.39.2

Bug fixes and other changes
  
  * pass enable_network_isolation in Estimator.create_model
  * use p2 instead of p3 for the Horovod test

1.39.1

Bug fixes and other changes
  
  * copy dependencies into new folder when repacking model
  * make get_caller_identity_arn get role from DescribeNotebookInstance
  * add https to regional STS endpoint
  * clean up git support integ tests

1.39.0

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

1.38.6

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

1.38.5

Bug fixes and other changes
  
  * clean up resources created by file system set up when setup fails

1.38.4

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

1.38.3

Bug fixes and other changes
  
  * change AMI ids in tests to be dynamic based on regions

1.38.2

Bug fixes and other changes
  
  * skip efs tests in non us-west-2 regions
  * refactor tests to use common retry method

1.38.1

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

1.38.0

Features
  
  * support training inputs from EFS and FSx

1.37.2

Bug fixes and other changes
  
  * Add support for Managed Spot Training and Checkpoint support
  * Integration Tests now dynamically checks AZs

1.37.1

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

1.37.0

Features
  
  * add XGBoost Estimator as new framework
  
  Bug fixes and other changes
  
  * fix tests for new regions
  * add update_endpoint for PipelineModel
  
  Documentation changes
  
  * refactor the using Chainer topic

1.36.4

Bug fixes and other changes
  
  * region build from staging pr
  
  Documentation changes
  
  * Refactor Using PyTorch topic for consistency

1.36.3

Bug fixes and other changes
  
  * fix integration test failures masked by timeout bug
  * prevent multiple values error in sklearn.transformer()
  * model.transformer() passes tags to create_model()

1.36.2

Bug fixes and other changes
  
  * rework CONTRIBUTING.md to include a development workflow

1.36.1

Bug fixes and other changes
  
  * prevent integration test's timeout functions from hiding failures
  
  Documentation changes
  
  * correct typo in using_sklearn.rst

1.36.0

Features
  
  * support for TensorFlow 1.14
  
  Bug fixes and other changes
  
  * ignore FI18 flake8 rule
  * allow Airflow enabled estimators to use absolute path entry_point

1.35.1

Bug fixes and other changes
  
  * update sklearn document to include 3p dependency installation
  
  Documentation changes
  
  * refactor and edit using_mxnet topic

1.35.0

Features
  
  * allow serving image to be specified when calling MXNet.deploy

1.34.3

Bug fixes and other changes
  
  * waiting for training tags to propagate in the test

1.34.2

Bug fixes and other changes
  
  * removing unnecessary tests cases
  * Replaced generic ValueError with custom subclass when reporting unexpected resource status
  
  Documentation changes
  
  * correct wording for Cloud9 environment setup instructions

1.34.1

Bug fixes and other changes
  
  * enable line-too-long Pylint check
  * improving Chainer integ tests
  * update TensorFlow script mode dependency list
  * improve documentation of some functions
  * update PyTorch version
  * allow serving script to be defined for deploy() and transformer() with frameworks
  * format and add missing docstring placeholders
  * add MXNet 1.4.1 support
  
  Documentation changes
  
  * add instructions for setting up Cloud9 environment.
  * update using_tensorflow topic

1.34.0

Features
  
  * Git integration for CodeCommit
  * deal with credentials for Git support for GitHub
  
  Bug fixes and other changes
  
  * modify TODO on disabled Pylint check
  * enable consider-using-ternary Pylint check
  * enable chained-comparison Pylint check
  * enable too-many-public-methods Pylint check
  * enable consider-using-in Pylint check
  * set num_processes_per_host only if provided by user
  * fix attach for 1P algorithm estimators
  * enable ungrouped-imports Pylint check
  * enable wrong-import-order Pylint check
  * enable attribute-defined-outside-init Pylint check
  * enable consider-merging-isinstance Pylint check
  * enable inconsistent-return-statements Pylint check
  * enable simplifiable-if-expression pylint checks
  * fix list serialization for 1P algos
  * enable no-else-return and no-else-raise pylint checks
  * enable unidiomatic-typecheck pylint check

1.33.0

Features
  
  * git support for hosting models
  * allow custom model name during deploy
  
  Bug fixes and other changes
  
  * remove TODO comment on import-error Pylint check
  * enable wrong-import-position pylint check
  * Revert "change: enable wrong-import-position pylint check (907)"
  * enable signature-differs pylint check
  * enable wrong-import-position pylint check
  * enable logging-not-lazy pylint check
  * reset default output path in Transformer.transform
  * Add ap-northeast-1 to Neo algorithms region map

1.32.2

Bug fixes and other changes
  
  * enable logging-format-interpolation pylint check
  * remove superfluous parens per Pylint rule
  
  Documentation changes
  
  * add pypi, rtd, black badges to readme

1.32.1

Bug fixes and other changes
  
  * correct code per len-as-condition Pylint check
  * tighten pylint config and expand C and R exceptions
  * Update displaytime.sh
  * fix notebook tests
  * separate unit, local mode, and notebook tests in different buildspecs
  
  Documentation changes
  
  * refactor the overview topic in the sphinx project

1.32.0

Features
  
  * support Endpoint_type for TF transform
  
  Bug fixes and other changes
  
  * fix git test in test_estimator.py
  * Add ap-northeast-1 to Neo algorithms region map

1.31.1

Bug fixes and other changes
  
  * print build execution time
  * remove unnecessary failure case tests
  * build spec improvements.

1.31.0

Features
  
  * use deep learning images
  
  Bug fixes and other changes
  
  * Update buildspec.yml
  * allow only one integration test run per time
  * remove unnecessary P3 tests from TFS integration tests
  * add pytest.mark.local_mode annotation to broken tests

1.30.0

Features
  
  * add TensorFlow 1.13 support
  * add git_config and git_clone, validate method
  
  Bug fixes and other changes
  
  * add pytest.mark.local_mode annotation to broken tests

1.29.0

Features
  
  * network isolation mode in training
  
  Bug fixes and other changes
  
  * Integrate black into development process
  * moving not canary TFS tests to local mode

1.28.3

Bug fixes and other changes
  
  * update Sagemaker Neo regions and instance families
  
  Documentation changes
  
  * fix punctuation in MXNet version list
  * clean up MXNet and TF documentation

1.28.2

Bug fixes and other changes
  
  * prevent race condition in vpc tests

1.28.1

Bug fixes and other changes
  
  * Update setup.py

1.28.0

Features
  
  * Add DataProcessing Fields for Batch Transform

1.27.0

Features
  
  * add wait argument to estimator deploy
  
  Bug fixes and other changes
  
  * fix logger creation in Chainer integ test script

1.26.0

Features
  
  * emit estimator transformer tags to model
  * Add extra_args to enable encrypted objects upload
  
  Bug fixes and other changes
  
  * downgrade c5 in integ tests and test all TF Script Mode images
  
  Documentation changes
  
  * include FrameworkModel and ModelPackage in API docs

1.25.1

Bug fixes and other changes
  
  * use unique job name in hyperparameter tuning test

1.25.0

Features
  
  * repack_model support dependencies and code location
  
  Bug fixes and other changes
  
  * skip p2 tests in ap-south-east
  * add better default transform job name handling within Transformer
  
  Documentation changes
  
  * TFS support for pre/processing functions

1.24.0

Features
  
  * add region check for Neo service

1.23.0

Features
  
  * support MXNet 1.4 with MMS
  
  Documentation changes
  
  * update using_sklearn.rst parameter name

1.22.0

Features
  
  * add encryption option to "record_set"
  
  Bug fixes and other changes
  
  * honor source_dir from S3

1.21.2

Bug fixes and other changes
  
  * set _current_job_name in attach()
  * emit training jobs tags to estimator

1.21.1

Bug fixes and other changes
  
  * repack model function works without source directory

1.21.0

Features
  
  * Support for TFS preprocessing

1.20.3

Bug fixes and other changes
  
  * run tests if buildspec.yml has been modified
  * skip local file check for TF requirements file when source_dir is an S3 URI
  
  Documentation changes
  
  * fix docs in regards to transform_fn for mxnet

1.20.2

Bug fixes and other changes
  
  * pin pytest version to 4.4.1 to avoid pluggy version conflict

1.20.1

Bug fixes and other changes
  
  * update TrainingInputMode with s3_input InputMode

1.20.0

Features
  
  * add RL Ray 0.6.5 support
  
  Bug fixes and other changes
  
  * prevent false positive PR test results
  * adjust Ray test script for Ray 0.6.5

1.19.1

Bug fixes and other changes
  
  * add py2 deprecation message for the deep learning framework images

1.19.0

Features
  
  * add document embedding support to Object2Vec algorithm

1.18.19

Bug fixes and other changes
  
  * skip p2/p3 tests in eu-central-1

1.18.18

Bug fixes and other changes
  
  * add automatic model tuning integ test for TF script mode

1.18.17

Bug fixes and other changes
  
  * use unique names for test training jobs

1.18.16

Bug fixes and other changes
  
  * add KMS key option for Endpoint Configs
  * skip p2 test in regions without p2s, freeze urllib3, and specify allow_pickle=True for numpy
  * use correct TF version in empty framework_version warning
  * remove logging level overrides
  
  Documentation changes
  
  * add environment setup instructions to CONTRIBUTING.md
  * add clarification around framework version constants
  * remove duplicate content from workflow readme
  * remove duplicate content from RL readme

1.18.15

Bug fixes and other changes
  
  * fix propagation of tags to SageMaker endpoint

1.18.14.post1

Documentation changes
  
  * remove duplicate content from Chainer readme

1.18.14.post0

Documentation changes
  
  * remove duplicate content from PyTorch readme and fix internal links

1.18.14

Bug fixes and other changes
  
  * make Local Mode export artifacts even after failure

1.18.13

Bug fixes and other changes
  
  * skip horovod p3 test in region with no p3
  * use unique training job names in TensorFlow script mode integ tests

1.18.12

Bug fixes and other changes
  
  * add integ test for tagging
  * use unique names for test training jobs
  * Wrap horovod code inside main function
  * add csv deserializer
  * restore notebook test

1.18.11

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

1.18.10

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

1.18.9.post1

Documentation changes
  
  * spelling error correction

1.18.9.post0

Documentation changes
  
  * move RL readme content into sphinx project

1.18.9

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

1.18.8

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()

1.18.7

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

1.18.6.post0

Documentation changes
  
  * add mandatory sagemaker_role argument to Local mode example.

1.18.6

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

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

1.18.4

* doc-fix: Remove incorrect parameter for EI TFS Python README
  * feature: ``Predictor``: delete SageMaker model
  * feature: ``PipelineModel``: delete SageMaker model
  * bug-fix: Estimator.attach works with training jobs without hyperparameters
  * doc-fix: remove duplicate content from mxnet/README.rst
  * doc-fix: move overview content in main README into sphynx project
  * bug-fix: pass accelerator_type in ``deploy`` for REST API TFS ``Model``
  * doc-fix: move content from tf/README.rst into sphynx project
  * doc-fix: move content from sklearn/README.rst into sphynx project
  * doc-fix: Improve new developer experience in README
  * feature: Add support for Coach 0.11.1 for Tensorflow

1.18.3.post1

* doc-fix: fix README for PyPI

1.18.3

* doc-fix: update information about saving models in the MXNet README
  * doc-fix: change ReadTheDocs links from latest to stable
  * doc-fix: add ``transform_fn`` information and fix ``input_fn`` signature in the MXNet README
  * feature: add support for ``Predictor`` to delete endpoint configuration by default when calling ``delete_endpoint()``
  * feature: add support for ``Model`` to delete SageMaker model
  * feature: add support for ``Transformer`` to delete SageMaker model
  * bug-fix: fix default account for SKLearnModel

1.18.2

* enhancement: Include SageMaker Notebook Instance version number in boto3 user agent, if available.
  * feature: Support for updating existing endpoint

1.18.1

* enhancement: Add ``tuner`` to imports in ``sagemaker/__init__.py``

1.18.0

* bug-fix: Handle StopIteration in CloudWatch Logs retrieval
  * feature: Update EI TensorFlow latest version to 1.12
  * feature: Support for Horovod

1.17.2

* feature: HyperparameterTuner: support VPC config

1.17.1

* enhancement: Workflow: Specify tasks from which training/tuning operator to transform/deploy in related operators
  * feature: Supporting inter-container traffic encryption flag

1.17.0

* bug-fix: Workflow: Revert appending Airflow retry id to default job name
  * feature: support for Tensorflow 1.12
  * feature: support for Tensorflow Serving 1.12
  * bug-fix: Revert appending Airflow retry id to default job name
  * bug-fix: Session: don't allow get_execution_role() to return an ARN that's not a role but has "role" in the name
  * bug-fix: Remove ``__all__`` from ``__init__.py`` files
  * doc-fix: Add TFRecord split type to docs
  * doc-fix: Mention ``SM_HPS`` environment variable in MXNet README
  * doc-fix: Specify that Local Mode supports only framework and BYO cases
  * doc-fix: Add missing classes to API docs
  * doc-fix: Add information on necessary AWS permissions
  * bug-fix: Remove PyYAML to let docker-compose install the right version
  * feature: Update TensorFlow latest version to 1.12
  * enhancement: Add Model.transformer()
  * bug-fix: HyperparameterTuner: make ``include_cls_metadata`` default to ``False`` for everything except Frameworks

1.16.3

* bug-fix: Local Mode: Allow support for SSH in local mode
  * bug-fix: Workflow: Append retry id to default Airflow job name to avoid name collisions in retry
  * bug-fix: Local Mode: No longer requires s3 permissions to run local entry point file
  * feature: Estimators: add support for PyTorch 1.0.0
  * bug-fix: Local Mode: Move dependency on sagemaker_s3_output from rl.estimator to model
  * doc-fix: Fix quotes in estimator.py and model.py

1.16.2

* enhancement: Check for S3 paths being passed as entry point
  * feature: Add support for AugmentedManifestFile and ShuffleConfig
  * bug-fix: Add version bound for requests module to avoid conflicts with docker-compose and docker-py
  * bug-fix: Remove unnecessary dependency tensorflow
  * doc-fix: Change ``distribution`` to ``distributions``
  * bug-fix: Increase docker-compose http timeout and health check timeout to 120.
  * feature: Local Mode: Add support for intermediate output to a local directory.
  * bug-fix: Update PyYAML version to avoid conflicts with docker-compose
  * doc-fix: Correct the numbered list in the table of contents
  * doc-fix: Add Airflow API documentation
  * feature: HyperparameterTuner: add Early Stopping support

1.16.1.post1

* Documentation: add documentation for Reinforcement Learning Estimator.
  * Documentation: update TensorFlow README for Script Mode

1.16.1

* feature: update boto3 to version 1.9.55

1.16.0

* feature: Add 0.10.1 coach version
  * feature: Add support for SageMaker Neo
  * feature: Estimators: Add RLEstimator to provide support for Reinforcement Learning
  * feature: Add support for Amazon Elastic Inference
  * feature: Add support for Algorithm Estimators and ModelPackages: includes support for AWS Marketplace
  * feature: Add SKLearn Estimator to provide support for SciKit Learn
  * feature: Add Amazon SageMaker Semantic Segmentation algorithm to the registry
  * feature: Add support for SageMaker Inference Pipelines
  * feature: Add support for SparkML serving container

1.15.2

* bug-fix: Fix FileNotFoundError for entry_point without source_dir
  * doc-fix: Add missing feature 1.5.0 in change log
  * doc-fix: Add README for airflow

1.15.1

* enhancement: Local Mode: add explicit pull for serving
  * feature: Estimators: dependencies attribute allows export of additional libraries into the container
  * feature: Add APIs to export Airflow transform and deploy config
  * bug-fix: Allow code_location argument to be S3 URI in training_config API
  * enhancement: Local Mode: add explicit pull for serving

1.15.0

* feature: Estimator: add script mode and Python 3 support for TensorFlow
  * bug-fix: Changes to use correct S3 bucket and time range for dataframes in TrainingJobAnalytics.
  * bug-fix: Local Mode: correctly handle the case where the model output folder doesn't exist yet
  * feature: Add APIs to export Airflow training, tuning and model config
  * doc-fix: Fix typos in tensorflow serving documentation
  * doc-fix: Add estimator base classes to API docs
  * feature: HyperparameterTuner: add support for Automatic Model Tuning's Warm Start Jobs
  * feature: HyperparameterTuner: Make input channels optional
  * feature: Add support for Chainer 5.0
  * feature: Estimator: add support for MetricDefinitions
  * feature: Estimators: add support for Amazon IP Insights algorithm

1.14.2

* bug-fix: support ``CustomAttributes`` argument in local mode ``invoke_endpoint`` requests
  * enhancement: add ``content_type`` parameter to ``sagemaker.tensorflow.serving.Predictor``
  * doc-fix: add TensorFlow Serving Container docs
  * doc-fix: fix rendering error in README.rst
  * enhancement: Local Mode: support optional input channels
  * build: added pylint
  * build: upgrade docker-compose to 1.23
  * enhancement: Frameworks: update warning for not setting framework_version as we aren't planning a breaking change anymore
  * feature: Estimator: add script mode and Python 3 support for TensorFlow
  * enhancement: Session: remove hardcoded 'training' from job status error message
  * bug-fix: Updated Cloudwatch namespace for metrics in TrainingJobsAnalytics
  * bug-fix: Changes to use correct s3 bucket and time range for dataframes in TrainingJobAnalytics.
  * enhancement: Remove MetricDefinition lookup via tuning job in TrainingJobAnalytics

1.14.1

* feature: Estimators: add support for Amazon Object2Vec algorithm

1.14.0

* feature: add support for sagemaker-tensorflow-serving container
  * feature: Estimator: make input channels optional

1.13.0

* feature: Estimator: add input mode to training channels
  * feature: Estimator: add model_uri and model_channel_name parameters
  * enhancement: Local Mode: support output_path. Can be either file:// or s3://
  * enhancement: Added image uris for SageMaker built-in algorithms for SIN/LHR/BOM/SFO/YUL
  * feature: Estimators: add support for MXNet 1.3.0, which introduces a new training script format
  * feature: Documentation: add explanation for the new training script format used with MXNet
  * feature: Estimators: add ``distributions`` for customizing distributed training with the new training script format

1.12.0

* feature: add support for TensorFlow 1.11.0

1.11.3

* feature: Local Mode: Add support for Batch Inference
  * feature: Add timestamp to secondary status in training job output
  * bug-fix: Local Mode: Set correct default values for additional_volumes and additional_env_vars
  * enhancement: Local Mode: support nvidia-docker2 natively
  * warning: Frameworks: add warning for upcoming breaking change that makes framework_version required

1.11.2

* enhancement: Enable setting VPC config when creating/deploying models
  * enhancement: Local Mode: accept short lived credentials with a warning message
  * bug-fix: Local Mode: pass in job name as parameter for training environment variable

1.11.1

* enhancement: Local Mode: add training environment variables for AWS region and job name
  * doc-fix: Instruction on how to use preview version of PyTorch - 1.0.0.dev.
  * doc-fix: add role to MXNet estimator example in readme
  * bug-fix: default TensorFlow json serializer accepts dict of numpy arrays

1.11.0

* bug-fix: setting health check timeout limit on local mode to 30s
  * bug-fix: make Hyperparameters in local mode optional.
  * enhancement: add support for volume KMS key to Transformer
  * feature: add support for GovCloud

1.10.1

* feature: add train_volume_kms_key parameter to Estimator classes
  * doc-fix: add deprecation warning for current MXNet training script format
  * doc-fix: add docs on deploying TensorFlow model directly from existing model
  * doc-fix: fix code example for using Gzip compression for TensorFlow training data

1.10.0

* feature: add support for TensorFlow 1.10.0

1.9.3.1

* doc-fix: fix rst warnings in README.rst

1.9.3

* bug-fix: Local Mode: Create output/data directory expected by SageMaker Container.
  * bug-fix: Estimator accepts the vpc configs made capable by 1.9.1

1.9.2

* feature: add support for TensorFlow 1.9

1.9.1

* bug-fix: Estimators: Fix serialization of single records
  * bug-fix: deprecate enable_cloudwatch_metrics from Framework Estimators.
  * enhancement: Enable VPC config in training job creation

1.9.0

* feature: Estimators: add support for MXNet 1.2.1

1.8.0

* bug-fix: removing PCA from tuner
  * feature: Estimators: add support for Amazon k-nearest neighbors(KNN) algorithm

1.7.2

* bug-fix: Prediction output for the TF_JSON_SERIALIZER
  * enhancement: Add better training job status report

1.7.1

* bug-fix: get_execution_role no longer fails if user can't call get_role
  * bug-fix: Session: use existing model instead of failing during ``create_model()``
  * enhancement: Estimator: allow for different role from the Estimator's when creating a Model or Transformer

1.7.0

* feature: Transformer: add support for batch transform jobs
  * feature: Documentation: add instructions for using Pipe Mode with TensorFlow

1.6.1

* feature: Added multiclass classification support for linear learner algorithm.

1.6.0

* feature: Add Chainer 4.1.0 support

1.5.4

* feature: Added Docker Registry for all 1p algorithms in amazon_estimator.py
  * feature: Added get_image_uri method for 1p algorithms in amazon_estimator.py
  * Support SageMaker algorithms in FRA and SYD regions

1.5.3

* bug-fix: Can create TrainingJobAnalytics object without specifying metric_names.
  * bug-fix: Session: include role path in ``get_execution_role()`` result
  * bug-fix: Local Mode: fix RuntimeError handling

1.5.2

* Support SageMaker algorithms in ICN region

1.5.1

* enhancement: Let Framework models reuse code uploaded by Framework estimators
  * enhancement: Unify generation of model uploaded code location
  * feature: Change minimum required scipy from 1.0.0 to 0.19.0
  * feature: Allow all Framework Estimators to use a custom docker image.
  * feature: Option to add Tags on SageMaker Endpoints

1.5.0

* feature: Add Support for PyTorch Framework
  * feature: Estimators: add support for TensorFlow 1.7.0
  * feature: Estimators: add support for TensorFlow 1.8.0
  * feature: Allow Local Serving of Models in S3
  * enhancement: Allow option for ``HyperparameterTuner`` to not include estimator metadata in job
  * bug-fix: Estimators: Join tensorboard thread after fitting

1.4.2

* bug-fix: Estimators: Fix attach for LDA
  * bug-fix: Estimators: allow code_location to have no key prefix
  * bug-fix: Local Mode: Fix s3 training data download when there is a trailing slash

1.4.1

* bug-fix: Local Mode: Fix for non Framework containers

1.4.0

* bug-fix: Remove __all__ and add noqa in __init__
  * bug-fix: Estimators: Change max_iterations hyperparameter key for KMeans
  * bug-fix: Estimators: Remove unused argument job_details for ``EstimatorBase.attach()``
  * bug-fix: Local Mode: Show logs in Jupyter notebooks
  * feature: HyperparameterTuner: Add support for hyperparameter tuning jobs
  * feature: Analytics: Add functions for metrics in Training and Hyperparameter Tuning jobs
  * feature: Estimators: add support for tagging training jobs

1.3.0

* feature: Add chainer

1.2.5

* bug-fix: Change module names to string type in __all__
  * feature: Save training output files in local mode
  * bug-fix: tensorflow-serving-api: SageMaker does not conflict with tensorflow-serving-api module version
  * feature: Local Mode: add support for local training data using file://
  * feature: Updated TensorFlow Serving api protobuf files
  * bug-fix: No longer poll for logs from stopped training jobs

1.2.4

* feature: Estimators: add support for Amazon Random Cut Forest algorithm

1.2.3

* bug-fix: Fix local mode not using the right s3 bucket

1.2.2

* bug-fix: Estimators: fix valid range of hyper-parameter 'loss' in linear learner

1.2.1

* bug-fix: Change Local Mode to use a sagemaker-local docker network

1.2.0

* feature: Add Support for Local Mode
  * feature: Estimators: add support for TensorFlow 1.6.0
  * feature: Estimators: add support for MXNet 1.1.0
  * feature: Frameworks: Use more idiomatic ECR repository naming scheme

1.1.3

* bug-fix: TensorFlow: Display updated data correctly for TensorBoard launched from ``run_tensorboard_locally=True``
  * feature: Tests: create configurable ``sagemaker_session`` pytest fixture for all integration tests
  * bug-fix: Estimators: fix inaccurate hyper-parameters in kmeans, pca and linear learner
  * feature: Estimators: Add new hyperparameters for linear learner.

1.1.2

* bug-fix: Estimators: do not call create bucket if data location is provided

1.1.1

* feature: Estimators: add ``requirements.txt`` support for TensorFlow

1.1.0

* feature: Estimators: add support for TensorFlow-1.5.0
  * feature: Estimators: add support for MXNet-1.0.0
  * feature: Tests: use ``sagemaker_timestamp`` when creating endpoint names in integration tests
  * feature: Session: print out billable seconds after training completes
  * bug-fix: Estimators: fix LinearLearner and add unit tests
  * bug-fix: Tests: fix timeouts for PCA async integration test
  * feature: Predictors: allow ``predictor.predict()`` in the JSON serializer to accept dictionaries

1.0.4

* feature: Estimators: add support for Amazon Neural Topic Model(NTM) algorithm
  * feature: Documentation: fix description of an argument of sagemaker.session.train
  * feature: Documentation: add FM and LDA to the documentation
  * feature: Estimators: add support for async fit
  * bug-fix: Estimators: fix estimator role expansion

1.0.3

* feature: Estimators: add support for Amazon LDA algorithm
  * feature: Hyperparameters: add data_type to hyperparameters
  * feature: Documentation: update TensorFlow examples following API change
  * feature: Session: support multi-part uploads
  * feature: add new SageMaker CLI

1.0.2

* feature: Estimators: add support for Amazon FactorizationMachines algorithm
  * feature: Session: correctly handle TooManyBuckets error_code in default_bucket method
  * feature: Tests: add training failure tests for TF and MXNet
  * feature: Documentation: show how to make predictions against existing endpoint
  * feature: Estimators: implement write_spmatrix_to_sparse_tensor to support any scipy.sparse matrix

1.0.1

* api-change: Model: Remove support for 'supplemental_containers' when creating Model
  * feature: Documentation: multiple updates
  * feature: Tests: ignore tests data in tox.ini, increase timeout for endpoint creation, capture exceptions during endpoint deletion, tests for input-output functions
  * feature: Logging: change to describe job every 30s when showing logs
  * feature: Session: use custom user agent at all times
  * feature: Setup: add travis file

1.0.0

* Initial commit

Links

Releases