Exam DP-100 topic 4 question 34 discussion

From CI/CD pipeline perspective, the e2e process should be: 1. Create a project. 2. Create service connections. 3. Create build pipeline. 4. Create release pipeline. I don't think it's necessary to install ml extension as a separate step since it's in pipeline script code. https://www.azuredevopslabs.com/labs/vstsextend/aml/

Looking at the doc below, the correct sequence should be: 1) create devops project [STEP 2] 2) create service connection [STEP 3] 3) create release pipeline [STEP 4] 4) install SDK extension [STEP 6] https://docs.microsoft.com/en-us/azure/devops/pipelines/targets/azure-machine-learning?view=azure-devops

1. Create an Azure DevOps project: This is the first step, where you set up a new Azure DevOps project to manage your build and release pipelines. 2. Install the Machine Learning extension for Azure Pipelines: This step ensures that the necessary tools are in place for working with Azure Machine Learning in your DevOps environment. 3. Create a service connection: This creates the link between Azure DevOps and your Azure Machine Learning workspace, allowing the pipelines to interact with your models and other resources. 4. Create a release pipeline: This pipeline is used to automate the deployment of the model to the target environment whenever a new model version is registered.

To configure Azure Pipelines to deploy a model from an Azure Machine Learning workspace, you should follow a specific sequence of actions. Here's the correct order based on the provided options: Create an Azure DevOps project: This is the initial step where you set up the project in Azure DevOps. Install the Machine Learning extension for Azure Pipelines: This step ensures that you have the necessary tools to work with Azure Machine Learning in your pipeline. Create a service connection: This step involves creating a connection to the Azure Machine Learning workspace so that the pipeline can interact with it. Create a build pipeline: This is where you define the build process, which includes steps to register the model. Create a release pipeline: Finally, set up the release pipeline to deploy the registered model to the target environment.

On exam 2023-03-27

Create a service connection: A service connection is required to connect Azure DevOps to Azure Machine Learning workspace. Install the Machine learning extension for Azure pipelines: This extension is required to use the Azure Machine Learning tasks in the release pipeline. Create a build pipeline: This step is optional, but it can be useful to create a build pipeline to build and package the model. Create a release pipeline: This step is essential to create a release pipeline that deploys the model to the target environment. Create an Azure DevOps project is not necessary since you have already created an Azure DevOps organization.