Skip to content

Latest commit

 

History

History
58 lines (40 loc) · 1.21 KB

README.rst

File metadata and controls

58 lines (40 loc) · 1.21 KB

Sagemill

Sagemill lets you to execute your notebook as SageMaker jobs in the similar way to papermill

Overall features

  • parameterize notebooks
  • execute notebooks as SageMaker Trainig/Processing Job

Installatioin

$ pip install sagemill

Prerequsites

  • Python>=3.5

Required IAM policy

  • ecr
    • create-repository
    • get-authorization-token
  • sts
    • get-caller-identity
  • permission to run Sagemaker jobs
    • write to cloudwatch logs
    • access to some S3 buckets
    • start training/proceessing jobs
    • etc.

Example notebooks

These notebook are assumed to be conda_python3 in SageMaker Notebook instance If you run it on different environments, install conda and run pip install sagemaker