news
VS Code tools for Azure Machine Learning now shipping after 7 years in preview
Nearly seven years after debuting as a preview, the Azure Machine Learning Visual Studio Code extension is now generally available.
“You can use your favorite VS Code setup, either desktop or web, to build, train, deploy, debug, and manage machine learning models with Azure Machine Learning from within VS Code,” Microsoft's Leo Yao said in yesterday's GA announcement.
Formerly known as “Visual Studio Code Tools for AI,” the Visual Studio Code Marketplace extension has achieved nearly 2.4 million installs since its debut in September 2017. The extension provides the following benefits to developers:
- Build and train machine learning models faster and easily deploy them to the cloud or edge.
- It uses the latest open-source technologies such as TensorFlow, PyTorch, and Jupyter.
- Experiment locally and then quickly scale up or out using large GPU-enabled clusters in the cloud.
- Accelerate data science with automated machine learning and hyperparameter tuning.
- Track your experiments, manage your models, and easily deploy them with integrated CI/CD tools.
The extension leverages the cross-platform capabilities of VS Code to support Windows, macOS, Linux Ubuntu, as well as other Linux distributions.
The documentation lists many more features in a separate bullet list.
- Manage Azure Machine Learning resources (experiments, virtual machines, models, deployments, etc.).
- Develop locally with a remote compute instance
- Train a machine learning model
- Debugging machine learning experiments locally
- Schema-based language support, auto-completion and diagnostics for creating spec files
This preview did not come with a service level agreement and certain features may not be supported or may have limited functionality, so Microsoft did not recommend it for production workloads.
Not anymore. Yao says the extension is stable, reliable, and production-ready, with additional features such as VNET support.
“After installing this extension, you can run much of this workflow directly from Visual Studio Code,” Yao explains. “The VS Code extension provides a user interface for creating and managing Azure ML resources, such as experiments, compute targets, environments, and deployments. It also supports the Azure ML 2.0 CLI, a new command-line tool that simplifies the specification and execution of machine learning tasks.”
Additionally, we will discuss how Azure ML Studio provides one-click access to VS Code, giving developers the option to get started using VS Code (Web) or VS Code (Desktop).
For more information, see the Azure Machine Learning for Visual Studio Code GitHub repository and the documentation on the Microsoft Learn site, Azure Machine Learning Documentation. The company also ran a survey to gather feedback.
About the Author
David Ramel is an editor and writer at Converge360.