Skip to content

aws-samples/mlops-ramp

MLOPS-RAMP

RAMP Program is a customer-centered workshop program in which customers directly configure MLOps for automating the model's seamless workflow with AWS' model development environment through the use of hands-on AWS services. RAMP allows customers to experience real-world trial-and-error experiences by using AWS' services themselves, thereby further enhancing their understanding of AWS' services. In addition, RAMP can be used as a base architecture for future customer actual workloads based on the architecture created in the customer's account.

In the MLOPS RAMP Program, as a Data Scientist, customers develop models in Amazon SageMaker, or you actually do methods of model inference, and in the role of Machine Learning Platform Engineer/MLOps Engineer, you implement the end-to-end workflow from data preparation to model distribution.

1. VISION model and Step Function

Go reference architecture

This senario performs model development using the Object Detection model of Yolov5 and then builds the MLOps architecture directly using AWS Step Functions from the customer's account. All implementations are directly done by the customer and the issues that arise during the course of the program are shared by the SAs in charge and supported by debugging.

2. AutoGluon Model and SageMaker Pipeline

Go reference architecture

This senario performs model development using the Binary Classification model of Autogluon and then builds the MLOps architecture using SageMaker Pipelines from the customer's account. All implementations are directly done by the customer and the issues that arise during the course of the program are shared by the SAs in charge and supported by debugging.

Security

See CONTRIBUTING for more information.

License

This library is licensed under the MIT-0 License. See the LICENSE file.

About

No description, website, or topics provided.

Resources

License

Code of conduct

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •