Irrigation-district staff have a working scikit-learn pipeline and…

Question

Irrigation-district staff have a working scikit-learn pipeline and want it trained as a managed SageMaker job, not rewritten in XGBoost and not handed a blank CUDA kernel. Which path fits?

Accepted Answer

Train with the SageMaker scikit-learn estimator and container so the existing pipeline stays a common-library job. — Staff already have a working scikit-learn pipeline. Point SageMaker's sklearn estimator at it and leave the code alone. Forcing XGBoost or a blank CUDA kernel rewrites working code, and Translate is not that trainer.

Answer

Rewrite the pipeline in XGBoost even though the working code is scikit-learn.

Answer

Hand the team a blank CUDA kernel and skip the SageMaker scikit-learn container.

Answer

Amazon Translate, because language is a scikit-learn estimator.

Irrigation-district staff have a working scikit-learn pipeline and want it trained as a managed SageMaker job, not rewritten in XGBoost and not handed a blank CUDA kernel. Which path fits?

Short Explanation

Full Explanation