Welcome to Kedro AzureML Pipeline's documentation¶
Kedro AzureML Pipeline is a Kedro plugin that connects your data science project to Azure ML Pipelines. With a single CLI command you can run, schedule, and monitor Kedro pipelines on Azure ML managed compute without changing any of your existing Kedro code, catalog, or hooks.
-
Get Started in 5 Minutes
Install the plugin, connect a Kedro project to your Azure ML workspace, and submit your first pipeline run to managed compute.
-
How-to Guides
Task-focused recipes for scheduling runs, managing data assets, scaling with distributed training, tracking with MLflow, and deploying from CI/CD.
-
Understand the Design
Learn how Kedro pipelines become Azure ML pipeline jobs and how data flows between steps.
-
Reference
Configuration fields, CLI flags, dataset parameters, and the full Python API.
Key capabilities¶
- No code changes: integrate Azure ML without touching your Kedro datasets, catalog, or pipelines
- Scheduling: configure cron and recurrence schedules directly in
azureml.yml - Distributed training: scale nodes across multiple GPU instances with
@distributed_job - Data asset management: version and track data through Azure ML using
AzureMLAssetDataset - Full hook lifecycle: all Kedro hooks fire during remote execution, including
kedro-mlflow - Multiple workspaces: target dev, staging, and production workspaces from one config
License¶
This project is licensed under the terms of the Apache-2.0 License.
Acknowledgements¶
This project is a fork of kedro-azureml, originally developed by GetInData. We are grateful for their work in creating the initial plugin that bridges Kedro and Azure ML Pipelines. We have continued development to add new features, improve documentation, and maintain the project under the kedro-azureml-pipeline package name.
We would also like to thank Evolta Technologies for their support to the project.
This project is maintained by stateful-y, an ML consultancy specializing in MLOps and data science & engineering. If you're interested in collaborating or learning more about our services, please visit our website.
