Introducing Azure Machine Learning Studio
Overview
Following a recent client engagement our team were tasked with evaluating the machine learning capabilities offered by Azure. Our client has a vast amount of structured data and was looking to implement machine learning practices to help them better forecast future demand and reduce business risk.
The client already has a number of workloads and services running on Azure therefore the requirement was to evaluate solutions within Azure.
One of the product offerings from Azure is their Azure Machine Learning Studio. We decided to share some insights following our time using the platform.
Evaluating the Azure Machine Learning Studio
Setting up a new learning workspace is very simple to complete, select your subscription and the required resource group. You can then define all of the workspace settings such as storage, authenication and the region.
Key service capabilities for the full machine learning lifecycle
Data labelling
Create, manage and monitor labelling projects, and automate iterative tasks with machine learning–assisted labelling.
Data preparation
Quickly iterate on data preparation at scale on Apache Spark clusters within Azure Machine Learning, interoperable with Azure Synapse Analytics.
Collaborative notebooks
Maximise productivity with IntelliSense, easy compute and kernel switching, and offline notebook editing. Launch your notebook in Visual Studio Code for a rich development experience, including secure debugging and support for Git source control.
Automated machine learning
Rapidly create accurate models for classification, regression, time-series forecasting, natural language processing tasks, and computer vision tasks. Use model interpretability to understand how the model was built.
Drag-and-drop machine learning
Use machine learning tools like designer for data transformation, model training, and evaluation, or to easily create and publish machine learning pipelines.
Reinforcement learning
Scale reinforcement learning to powerful compute clusters, support multiple-agent scenarios, and access open-source reinforcement-learning algorithms, frameworks, and environments.
Responsible building
Get model transparency at training and inferencing with interpretability capabilities. Assess model fairness through disparity metrics and mitigate unfairness. Improve model reliability and identify and diagnose model errors with the error analysis toolkit. Help protect data with differential privacy.
Experimentation
Manage and monitor runs or compare multiple runs for training and experimentation. Create customised dashboards and share them with your team.
Registries
Use organisation-wide repositories to store and share models, pipelines, components, and datasets across multiple workspaces. Automatically capture lineage and governance data using the audit trail feature.
Git and GitHub
Use Git integration to track work and GitHub Actions support to implement machine learning workflows.
Managed endpoints
Use managed endpoints to operationalize model deployment and scoring, log metrics, and perform safe model rollouts.
Autoscaling compute
Use managed compute to distribute training and to rapidly test, validate, and deploy models. Share CPU and GPU clusters across a workspace and automatically scale to meet your machine learning needs.
Interoperability with other Azure services
Accelerate productivity with Microsoft Power BI and services such as Azure Synapse Analytics, Azure Cognitive Search, Azure Data Factory, Azure Data Lake, Azure Arc, Azure Security Centre and Azure Databricks.
Hybrid and multicloud support
Run machine learning on existing Kubernetes clusters on premises, in multicloud environments, and at the edge with Azure Arc. Use the simple machine learning agent to start training models more securely, wherever your data lives.
Enterprise-grade security
Build and deploy models more securely with network isolation and end-to-end private IP capabilities, role-based access control for resources and actions, customised roles, and managed identity for compute resources.
Cost management
Reduce IT costs and better manage resource allocations for compute instances, with workspace and resource-level quota limits and automatic shutdown.
Want to learn more?
We recommend A Cloud Guru for all of your cloud based training requirements and can highly recommend the Data Science Basics course. Michael Blythe from our team completed the in December 2022 and highly recommends it.
