Manage Machine Learning Experiments with Amazon SageMaker Model Tracking Capability - Amazon SageMaker

To organize, find, and evaluate machine leaning model experiments, use Amazon SageMaker model tracking capabilities. Developing models typically requires extensive experimenting with different datasets, algorithms, and parameter values. Using the model tracking capability, you can search, filter and sort through hundreds and possibly thousands of experiments using model attributes such as parameters, metrics and tags. This helps you find the best model for your use case quickly. Find, organize, or evaluate training jobs using properties, hyperparameters, performance metrics, or any other metadata.