You are about setting up a machine learning environment.
You already have a workspace where you need to configure the compute resources for your experiments.
You are going to make use of the capabilities of Azure's AutoML feature and you want to use ML pipelines to organize your workflow, for which you want to use the ML Designer.
Which compute resource should you choose?
Click on the arrows to vote for the correct answer
A. B. C. D.Answer: D.
Option A is incorrect because the ML compute instance is good for both AutoML and for training run of pipelines, it is not suitable for ML Designer.
Option B is incorrect because while HDInsights is capable of running pipelines, it is not suitable for AutoML and for ML Designer.
Option C is incorrect because remote VMs cannot be used together with ML Designer.
Option D is CORRECT because Azure ML compute cluster is the only option suitable for AutoML, for running pipelines as well as to exploit the capabilities of the graphical ML Designer.
Reference:
To configure the compute resources for your machine learning experiments on Azure, you can choose one of the following options: Azure ML compute instance, Azure HDInsights, Remote VM, or Azure ML compute cluster.
Since you want to use the ML pipelines and the ML Designer to organize your workflow, you should choose the Azure ML compute cluster as the compute resource. The Azure ML compute cluster is a fully-managed, cloud-based service that provides scalable compute resources for your machine learning workloads. It can automatically scale up or down based on the needs of your workload, and you can easily configure it using Azure Machine Learning Studio or the Azure CLI.
The Azure ML compute cluster is designed specifically for machine learning workloads, and it supports a wide range of machine learning frameworks and libraries, including TensorFlow, PyTorch, scikit-learn, and many others. With the Azure ML compute cluster, you can easily provision and manage compute resources for your machine learning experiments, and you can take advantage of the AutoML feature to automate the process of building and deploying machine learning models.
In summary, if you want to use the ML pipelines and the ML Designer to organize your workflow, and you want to take advantage of the AutoML feature on Azure, you should choose the Azure ML compute cluster as your compute resource.