From the course: Microsoft Azure Data Scientist Associate (DP-100) Cert Prep: 4 Implement Responsible Machine Learning
Unlock this course with a free trial
Join today to access over 24,700 courses taught by industry experts.
Trigger an Azure Machine Learning pipeline, including from Azure DevOps or GitHub - Azure Tutorial
From the course: Microsoft Azure Data Scientist Associate (DP-100) Cert Prep: 4 Implement Responsible Machine Learning
Trigger an Azure Machine Learning pipeline, including from Azure DevOps or GitHub
- [Instructor] One of the more interesting things that you can do is use GitHub as a general worker farm to create actions that get deployed into Azure ML Studio. And because both products have great integration with each other, one of the things you can do is hook up GitHub via secrets. So create creature credentials via Azure ML Studio, hook those into the GitHub actions ecosystem, and then configure a Job using the GitHub actions YAML format. All right, let's go ahead and take a look at how that would work. So first up we have a repo here that I use for certain operations. And let's pretend that this is the repo that would be communicating with Azure ML Studio. All I would need to do is scroll down to the Setting section right here. And notice that under Secrets, under Actions, I could create a new repository secret. And this could be, for example, my Azure credentials. And I would put that secret right in here. Once…
Contents
-
-
Configure compute for a batch deployment2m 11s
-
(Locked)
Deploy a model to a batch endpoint4m 2s
-
Test a real-time deployed service4m 23s
-
(Locked)
Apply Machine Learning Operations (MLOps) practices4m 32s
-
(Locked)
Trigger an Azure Machine Learning pipeline, including from Azure DevOps or GitHub2m 36s
-
(Locked)
Conclusion1m 6s
-