how to trigger sagemaker pipeline via code change in github?

0

based on the aws docs and sample code provided here, https://github.com/aws/amazon-sagemaker-examples/blob/6299535b80b44ef0b61b95c979b1511157965810/sagemaker-pipelines/tabular/customizing_build_train_deploy_project/modelbuild/pipelines/customer_churn/pipeline.py, I can wrap different ML steps like training into a step , in a sagemaker pipeline code. after that, i can define sagemaker projects to create a CI/CD pipeline to build and deploy models. the examples i saw builds/deploy model in aws code commit/pipeline. is it possible to move this part to github or gitlab for repository and then deploy from gitlab or other tools like jenkins. are there any examples or code samples to do this?? also, instead of working in sagemaker studio IDE, is it possible to set up code repo in github or gitlab and we use our own IDE to push changes to the pipeline or sagemkaer projects code to build/deploy model? and somehow hook that into studio such that for example , we make a change to hyperparameter value in gitlab, that will trigger the pipeline in studio?

2 Answers
0

You can create SageMaker projects using third party source code repositories such as Gitlab, Github etc. Here's a blog that walks through a solution - https://aws.amazon.com/blogs/machine-learning/create-amazon-sagemaker-projects-using-third-party-source-control-and-jenkins/ If you did not want to use Projects, you can still use pipelines with your source code. You can create an Amazon EventBridge rule to trigger a SageMaker pipeline execution based on events or a given schedule. See an example here - https://docs.aws.amazon.com/sagemaker/latest/dg/pipeline-eventbridge.html

AWS
Durga_S
answered 2 years ago
0

You can use the Sagemaker Project, which includes templates for building MLOps on GitHub. The reference link is as follows: https://catalog.us-east-1.prod.workshops.aws/workshops/44d3e2a0-ec6f-44df-9397-bcfdf129cadf/en-US

answered a year ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions