Azure Machine Learning Service Part-1: Creating Workspace
In a series of blog posts, I am planning to write down my experiences of training, deploying and managing models and running pipelines with Azure Machine Learning Service. This is part-1 where I will be walking you through the creation of workspace in Azure ML service
About Azure Machine Learning Service
Azure Machine Learning Service is a cloud based platform from Microsoft to train, deploy, automate, manage and track ML models. It has a facility to build models by using drag-drop components in Designer along with traditional code based model building. Azure ML service makes our job very ease in maintaining developed models and also helps in hassle free deployment of models in lower(QA, Unit) and higher(Prod) environments as APIs. It is integrated with various components in Azure like Azure Kubernetes Services, Azure Databricks, Azure Monitor, Azure Storage accounts, Azure Pipelines, MLFlow, Kubeflow to carry out various activities which will be discussed in upcoming posts.
Why Azure Machine Learning Service
In the process of building models, one need to play around with various hyperparameters and use various techniques. Also one need to scale out the resources for training the model if the dataset is huge. Bringing your model development and deployment to cloud makes your job easy. In particular Azure Machine Learning Service has below advantages.
- Simplifies model management
- Automated machine learning simplifies model building
- Scales out training to GPU cluster or CPU cluster or Azure Databricks whenever needed with inbuilt integration
- Deployment of models to production with Azure Kubernetes Service or Azure IOT edge is very simple.
Azure Machine Learning Service Terms
Let me list down some of the common terms that you come across while working with Azure Machine Learning Service.
- Workspace: Workspace is a high-level resource for Azure ML. It gives centralised place to work with various components that are part of Azure ML Service.
- Experiment: An experiment is a group of runs from a specified script.
- Run: Each run of the model(script) is termed as Run in the Azure ML workspace. For example lets think that you are trying to build a model to classify images in MNIST dataset. Experiment-1 going to be ResNet-18 model in which you will have different runs with different set of hyperparameters. Experiment-2 is going to be a VGG-16 model which has different runs based on the set of hyperparameters chosen.
- Compute Targets: Compute target are the machines used for training and inferencing jobs. It could be a GPU cluster or CPU cluster or Azure Kubernetes Service cluster or Azure Databricks
- Datastore: Datastore is a term for storage space for storing data with Azure ML. The datastore can use either an Azure blob container or an Azure file share as the back-end storage. Each workspace has a default datastore and datastores can also be added.
Below is the taxonomy of Azure Machine Learning Workspace
Steps to Create Azure Machine Learning Workspace through Azure Portal
- Login into your Azure account in Azure Portal
- Click on ‘Create a resource’ button as shown below image
3. Search for Machine Learning as show in below image
4. Click on Create button to create Azure Machine Learning Workspace
5. Enter the details as shown in below image and click on ‘Review & Create’
6. After the successful deployment of your request for Azure ML workspace, please login into ML Portal and select the details from above including the workspace created in the above step as shown in below image and ‘Get Started’ button.
7. You should be able to see below page with all the component of Azure Machine Learning Service.
With this I will be ending this part of the series.
Next:
In the next part, I will be discussing about training machine learning model in Azure Machine Learning Service.
You can reach me in LinkedIn
