hyperparameters evaluation

Please check following link for using SDK method for best training jobs: https://sagemaker.readthedocs.io/en/stable/api/training/tuner.html

Also, please refer to following blog for using model attributes to track training: https://aws.amazon.com/blogs/machine-learning/amazon-sagemaker-now-comes-with-new-capabilities-for-accelerating-machine-learning-experimentation/

Basically SageMaker persists hyperparameters and model S3 ARN for training job in its metadata store. Data is not versioned but only S3 ARNs

Some links for reference:

https://docs.aws.amazon.com/sagemaker/latest/dg/automatic-model-tuning-how-it-works.html

https://docs.aws.amazon.com/sagemaker/latest/dg/xgboost-tuning.html

https://aws.amazon.com/blogs/machine-learning/optimize-hyperparameters-with-amazon-sagemaker-automatic-model-tuning/

https://aws.amazon.com/blogs/machine-learning/tune-ml-models-for-additional-objectives-like-fairness-with-sagemaker-automatic-model-tuning/