Docker Registry Paths and Example Code - Amazon SageMaker
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China.

Docker Registry Paths and Example Code

The following topics list the Docker registry path and other parameters for each of the Amazon SageMaker provided algorithms and Deep Learning Containers (DLC).

Use the path as follows:

  • To create a training job (create_training_job), specify the Docker registry path (TrainingImage) and the training input mode (TrainingInputMode) for the training image. You create a training job to train a model using a specific dataset.

  • To create a model (create_model), specify the Docker registry path (Image) for the inference image (PrimaryContainer Image). SageMaker launches machine learning compute instances that are based on the endpoint configuration and deploys the model, which includes the artifacts (the result of model training).


For the registry path, use the :1 version tag to ensure that you are using a stable version of the algorithm/DLC. You can reliably host a model trained using an image with the :1 tag on an inference image that has the :1 tag. Using the :latest tag in the registry path provides you with the most up-to-date version of the algorithm/DLC, but might cause problems with backward compatibility. Avoid using the :latest tag for production purposes.

For XGBoost, do not use :latest or :1. Use the specific version you require, such as :0.90-1-cpu-py3, :0.90-2-cpu-py3, :1.0-1-cpu-py3, or :1.2-1.

To find the registry path, choose the Amazon Region, then choose the algorithm or DLC.