Create Databricks model serving endpoint in Azure DevOps yaml

[afdadfadsfadsf]

Hello,

I need to create and destroy a model endpoint as part of CI/CD. 

I tried with mlflow deployments create-endpoint, giving databricks as --target however it errors saying that --endpoint is not a known argument when clearly --endpoint is required. I tried this in DevOps and also in Databricks proper.

I tried running Python inside yaml but mlflow.login() fails with interactive=False and several different configurations for .databrickscfg file (host and token, host username and password).

Is there a way to create an endpoint via yaml? Do you have a working example?

[Kaniz_Fatma]

Hi @afdadfadsfadsf, Creating and managing model endpoints as part of your CI/CD pipeline is essential for deploying machine learning models. I can provide some guidance on how to set up a CI/CD pipeline using YAML in Azure DevOps.

You can adapt these principles to your specific use case.

1. Azure DevOps YAML Pipelines:

   • Azure DevOps allows you to define your build and release pipelines using YAML (YAML Ain’t Markup Language). This approach enables you to manage your pipeline configuration as code, just like any other source file.
   • To get started, create a YAML build definition by adding a YAML file to the root of your repository. You can use the same pipeline features available in the visual designer but with a markup file.
   • Here’s an example of configuring CI/CD pipelines as code with YAML in Azure DevOps:
     • Azure DevOps Hands-on-Labs: Configuring CI/CD Pipelines as Code with YAML¹

2. Sample ASP.NET Core Project with YAML Pipeline:

   • If you’re working with ASP.NET Core, you can refer to this sample project on GitHub that includes a Core 2.2 web project and a sample YAML pipeline:
     • Web Project with YAML Pipeline²

3. GitLab CI/CD Minimal Example:

   • While this example is from GitLab, the basic structure of a GitLab CI/CD pipeline defined in YAML can serve as a reference. It sets up a single-stage pipeline with one job:
     • Understanding YAML and CI/CD³

4. Amazon SageMaker CI/CD Pipeline:

   • If you’re using Amazon SageMaker, this blog post demonstrates how to build a CI/CD pipeline for deploying custom machine learning models using AWS services:
     • Build a CI/CD Pipeline for Deploying Custom ML Models⁴

Remember to adapt these examples to your specific MLflow deployment scenario. While YAML pipelines provide flexibility, you may need to customize the steps based on your requirements. Good luck with your CI/CD setup! 😊

To ensure we provide you with the best support, could you please take a moment to review the response and choose the one that best answers your question? Your feedback not only helps us assist you better but also benefits other community members who may have similar questions in the future.

If you found the answer helpful, consider giving it a kudo. If the response fully addresses your question, please mark it as the accepted solution. This will help us close the thread and ensure your question is resolved.

We appreciate your participation and are here to assist you further if you need it!