Posted On: Aug 2, 2016
You can now access information on how long it took to create a datasource, model, evaluations, or batch predictions, in the Amazon Machine Learning (Amazon ML) console, the API, and in the AWS SDKs. Using this information, you can estimate how long it would take to create similar objects. Knowing the compute time, you can then estimate the cost of datasource creation, model training, or model evaluation based on similar objects created in the past. For regularly retrained models, you can use this information to detect problems in the retraining pipeline before they are deployed to production. This information is found on the summary page in the console for each object, or returned through the API or SDK as an additional value.
For more information about using Compute and Completion Time in Amazon ML, visit the Amazon ML documentation.