Overview:

Before exploring Azure ML Studio, let’s understand basic ML glossary.

Lets define these terms:

Azure Machine Learning glossary – Azure Machine Learning | Microsoft Learn

ML Workspace

The workspace is the top-level resource for Machine Learning. It provides a centralized place to work with all the artifacts you create when you use Machine Learning. The workspace keeps a history of all jobs, including logs, metrics, output, and a snapshot of your scripts. The workspace stores references to resources like datastores and compute. It also holds all assets like models, environments, components, and data assets.

Components:

A Machine Learning component is a self-contained piece of code that does one step in a machine learning pipeline. Components are the building blocks of advanced machine learning pipelines. Components can do tasks such as data processing, model training, and model scoring. A component is analogous to a function. It has a name and parameters, expects input, and returns output.

Machine Learning allows you to work with different types of data:

• URIs (a location in local or cloud storage):
  • uri_folder
  • uri_file
• Tables (a tabular data abstraction):
  • mltable
• Primitives:
  • string
  • boolean
  • number

For most scenarios, you use URIs (uri_folder and uri_file) to identify a location in storage that can be easily mapped to the file system of a compute node in a job by either mounting or downloading the storage to the node.

The mltable parameter is an abstraction for tabular data that’s used for automated machine learning (AutoML) jobs, parallel jobs, and some advanced scenarios. If you’re starting to use Machine Learning and aren’t using AutoML, we strongly encourage you to begin with URIs.

Datastore:

Machine Learning datastores securely keep the connection information to your data storage on Azure so that you don’t have to code it in your scripts. You can register and create a datastore to easily connect to your storage account and access the data in your underlying storage service. The Azure Machine Learning CLI v2 and SDK v2 support the following types of cloud-based storage services:

• Azure Blob Storage container
• Azure Files share
• Azure Data Lake Storage
• Azure Data Lake Storage Gen2

Environment:

Machine Learning environments are an encapsulation of the environment where your machine learning task happens. They specify the software packages, environment variables, and software settings around your training and scoring scripts. The environments are managed and versioned entities within your Machine Learning workspace. Environments enable reproducible, auditable, and portable machine learning workflows across various computes.

 Machine Learning allows you to create your own environment by using:

• A Docker image.
• A base Docker image with a conda YAML to customize further.
• A Docker build context.

Models:

Machine Learning models consist of the binary files that represent a machine learning model and any corresponding metadata. You can create models from a local or remote file or directory. For remote locations, https, wasbs, and azureml locations are supported. The created model is tracked in the workspace under the specified name and version. Machine Learning supports three types of storage formats for models:

• custom_model
• mlflow_model
• triton_model

Compute:

A compute is a designated compute resource where you run your job or host your endpoint. Machine Learning supports the following types of compute:

• Compute cluster: A managed-compute infrastructure that you can use to easily create a cluster of CPU or GPU compute nodes in the cloud.
• Compute instance: A fully configured and managed development environment in the cloud. You can use the instance as a training or inference compute for development and testing. It’s similar to a virtual machine in the cloud.
• Kubernetes cluster: Used to deploy trained machine learning models to Azure Kubernetes Service (AKS). You can create an AKS cluster from your Machine Learning workspace or attach an existing AKS cluster.
• Attached compute: You can attach your own compute resources to your workspace and use them for training and inference.

Demo: Exploring Azure Machine Learning Studio

You’ll create the resources you need to start working with Azure Machine Learning.

• A workspace. To use Azure Machine Learning, you’ll first need a workspace. The workspace is the central place to view and manage all the artifacts and resources you create.
• A compute instance. A compute instance is a pre-configured cloud-computing resource that you can use to train, automate, manage, and track machine learning models. A compute instance is the quickest way to start using the Azure Machine Learning SDKs and CLIs. You’ll use it to run Jupyter notebooks and Python scripts in the rest of the tutorials.

ML Workspace

A workspace is a centralized environment for managing all aspects of a machine learning project, including data, experiments, models, and deployments

Components

Components are reusable, modular blocks used to build machine learning pipelines, encapsulating data preparation, model training, and other processes.

MLWorkspace

Model catalog:

The model catalog is a repository for storing and managing trained machine learning models, allowing for versioning and retrieval of models for deployment

Notebook:

Notebooks are interactive coding environments that allow data scientists to write and execute code, visualize data, and share insights within a document.

Automated ML

Automated ML refers to the process of automating the selection, composition, and parameterization of machine learning models to streamline model development.

Designer:

The designer is a graphical interface that enables users to construct machine learning models and pipelines using a drag-and-drop approach.

Data:

In the context of machine learning, data refers to the raw information used to train models, validate their accuracy, and make predictions.

Dataset:

A dataset is a collection of data that has been processed and formatted in a way that is ready for use in training and evaluating machine learning models.

Jobs:

Jobs are individual instances of a machine learning process, such as training or inference, that are executed with specific configurations and resources.

Environment:

An environment specifies the software dependencies and configurations necessary to run machine learning code consistently across different compute targets

Models:

Models are the output of machine learning algorithms that have been trained on data to recognize patterns and make predictions or decisions.

Endpoints:

Endpoints are interfaces through which external applications can access and use the functionality of deployed machine learning models.

Compute:

Compute refers to the processing resources, such as CPUs or GPUs, allocated for running machine learning workloads and experiments.

Conclusion:

This post guide is an invaluable resource for beginners, providing clear instructions on setting up a workspace and compute instance, which are fundamental for any machine learning project.

The guide emphasizes the importance of the workspace as the central place for managing machine learning artifacts and the compute instance as a key resource for model training and management.

With these resources in place, users can confidently proceed to explore the full capabilities of Azure Machine Learning, leveraging its powerful tools for innovative and efficient machine learning solutions.

References:

Tutorial: Create workspace resources – Azure Machine Learning | Microsoft Learn

Azure Machine Learning glossary – Azure Machine Learning | Microsoft Learn