aws / sagemaker-python-sdk

A library for training and deploying machine learning models on Amazon SageMaker
https://sagemaker.readthedocs.io/
Apache License 2.0
2.09k stars 1.13k forks source link

Programmatic way to get available training/inference containers. #4318

Open Alex-Wenner-FHR opened 9 months ago

Alex-Wenner-FHR commented 9 months ago

Describe the feature you'd like It would be very helpful to be able to programmatically get all available framerworks, py versions etc, prior to sagemaker.image_uris.retrieve(...). This would allow people to see what containers are available, without having to dig through repositories and try to find supported images or continuously hit errors until the right permutation of framework, framework version, and py version are achieved.

How would this feature be used? Please describe. Something like sagemaker.image_uris.show_images() or something like that would be desired. This ideally would take no parameters and show maybe a dictionary of all the frameworks, framework versions, and respective py verisons.

Describe alternatives you've considered Considering programmatically parsing: https://github.com/aws/deep-learning-containers/blob/master/available_images.md

But this does not feel like a good solution at all.

Alex-Wenner-FHR commented 9 months ago

Related Issue: https://github.com/aws/deep-learning-containers/issues/2732