Source Scripts

ML scripts executed inside SageMaker containers during pipeline runs. These are uploaded to S3 by the training buildspec and referenced by the pipeline steps in ml_pipelines/training/pipeline.py.

Script	Pipeline Step	Container
preprocessing.py	PreprocessAbaloneData	SageMaker Processing (sklearn)
evaluate.py	EvaluateAbaloneModel	SageMaker Processing (xgboost)

Adding a new script

1. Add the Python file to this directory
2. Reference it via ProcessingStep.code in your pipeline definition
3. Ensure the buildspec uploads it to S3 before start-pipeline-execution