Sagemaker Experiment SDK v0.1.35

SageMaker Experiments v0.1.35 release!

Changes:

• Delete cloudformation.schema.json
• Delete sam.schema.json
• doc: remove references to CreateArtifact restriction and clarify where graph artifacts are rendered in Studio

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.34

SageMaker Experiments v0.1.34 release!

Changes:

• clarify log_metric behavior in Tracker documentation

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.33

SageMaker Experiments v0.1.33 release!

Changes:

• doc: add note regarding enable_sagemaker_metrics

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.32

SageMaker Experiments v0.1.32 release!

Changes:

• doc: update graph artifact documentation

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.31

SageMaker Experiments v0.1.31 release!

Changes:

• doc: clarify receiver operating characteristic curve doc
• Revert "remove unneccessary codcov token"
• remove unneccessary codcov token

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.30

SageMaker Experiments v0.1.30 release!

Changes:

• fix(tracker): log_metrics dataframe argument check allows df to be passed

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.29

SageMaker Experiments v0.1.29 release!

Changes:

• add no skill to P/R curve graph artifact
• add documentation regarding limitations of log_metric method

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.28

SageMaker Experiments v0.1.28 release!

Changes:

• feature: add log confusion matrix
• add confusion matrix json schema

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.27

SageMaker Experiments v0.1.27 release!

Changes:

• enable log of precision recall graphs

You can upgrade from PyPI via:

pip install -U sagemaker-experiments

Sagemaker Experiment SDK v0.1.26

SageMaker Experiments v0.1.26 release!

Changes:

• add wait for trial component to be created to improve test reliability
• add collect tests tox env
• list trials under trial component
• Log all files in a directory
• handle max input/output artifacts on TC
• add note to methods that call restricted API CreateArtifact
• add sleep in integ test for consistent results
• add docker-based test instructions
• revert use of CMH in test setup

You can upgrade from PyPI via:

pip install -U sagemaker-experiments