How do I authenticate with Azure using this skill?

The skill utilizes DefaultAzureCredential from the azure-identity package, which automatically handles various authentication methods including environment variables, managed identities, and Azure CLI sessions.

What is the difference between this and the legacy Azure ML SDK?

This skill focuses on the v2 SDK (azure-ai-ml), which offers a more streamlined experience, improved CLI compatibility, and a more consistent object model compared to the v1 SDK.

Can I use this to track and monitor training progress?

Yes, by creating 'Jobs' (such as command jobs), you can stream logs in real-time and access the Azure Machine Learning Studio URL to monitor metrics and outputs.

Does this skill support automated resource cleanup?

It provides the logic for managing compute clusters, including setting up idle time-before-scale-down to automatically reduce costs when resources aren't in use.

Azure Machine Learning SDK (Python)

Name: Azure Machine Learning SDK (Python)
Author: microsoft

bymicrosoft

•

1,777

•

Data Science & ML

Manages Azure Machine Learning resources including workspaces, compute clusters, data assets, and training jobs using the Python SDK v2.

The Azure Machine Learning SDK v2 (azure-ai-ml) skill empowers developers to programmatically control the entire ML lifecycle within the Azure ecosystem. It simplifies complex tasks such as setting up workspaces, provisioning scalable compute clusters, registering datasets and models, and orchestrating multi-step pipelines. Whether you are automating training jobs, deploying models to the registry, or managing cloud infrastructure for AI experiments, this skill provides the implementation patterns and best practices needed to build robust MLOps workflows directly within Claude.

Key Features

01Data asset registration and versioning for URI files and folders

02Full workspace and compute cluster lifecycle management

031,777 GitHub stars

04Simplified job submission for training and experimentation

05Model registry operations for versioning and archiving ML models

06DSL-based pipeline orchestration for multi-step ML workflows

Use Cases

01Automating the provisioning of CPU/GPU compute clusters for distributed training

02Managing centralized model registries and data assets for team collaboration

03Building reproducible MLOps pipelines that connect data preprocessing to model training

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add microsoft/skills azure-ai-ml-py

For use in Claude.ai and ChatGPT