What is Kedro-Dagster?

The Kedro-Dagster plugin enables seamless integration between Kedro, a framework for creating reproducible and maintainable data science code, and Dagster, a data orchestrator for machine learning and data pipelines. This plugin makes use of Dagster's orchestration capabilities to automate and monitor Kedro pipelines effectively.

Currently, Kedro-Dagster supports Kedro versions 0.19.x and 1.x, and Dagster versions 1.10.x, 1.11.x, and 1.12.x.

What are the features of Kedro-Dagster?

• Configuration‑driven workflows: Centralize orchestration settings in a dagster.yml file for each Kedro environment. Define jobs from filtered Kedro pipelines, assign executors, schedules.
• Customization: The core integration lives in the auto‑generated Dagster definitions.py. For advanced use cases, you can extend or override these definitions.
• Kedro hooks preservation: Kedro hooks are preserved and called at the appropriate time during pipeline execution, so custom logic (e.g., data validation, logging) continues to work seamlessly.
• MLflow compatibility: Use Kedro-MLflow with Dagster’s MLflow integration to track experiments, log models, and register artifacts.
• Logger integration: Unifies Kedro and Dagster logging so logs from Kedro nodes appear in the Dagster UI and are easy to trace and debug. Additionally, provides configuration to customize Dagster run loggers.
• (Experimental) Dagster partition support: Make use of Dagster's partitions to fan-out Kedro nodes acting on partitioned data.

How to install Kedro-Dagster?

Install the Kedro-Dagster plugin using pip:

pip install kedro-dagster

or using uv:

uv pip install kedro-dagster

or using conda:

conda install -c conda-forge kedro-dagster

or using mamba:

mamba install -c conda-forge kedro-dagster

or alternatively, add kedro-dagster to your requirements.txt or pyproject.toml file.

How to get started with Kedro-Dagster?

• Initialize the plugin in your Kedro project

Use the following command to generate a definitions.py file, where all translated Kedro objects are available as Dagster objects, and a dagster.yml configuration file:

kedro dagster init --env <ENV_NAME>

• Configure jobs, executors, and schedules

Define your job executors and schedules in the dagster.yml configuration file located in your Kedro project's conf/<ENV_NAME> directory. This file allows you to filter Kedro pipelines and assign specific executors and schedules to them.

# conf/local/dagster.yml
schedules:
  daily: # Schedule name
    cron_schedule: "0 0   *" # Schedule parameters
executors: # Executor name   sequential: # Executor parameters     in_process:
multiprocess:     multiprocess:       max_concurrent: 2
jobs:   default: # Job name     pipeline: # Pipeline filter parameters       pipeline_name: default     executor: sequential
paralleldataprocessing:     pipeline:       pipelinename: dataprocessing       node_names:       - preprocesscompaniesnode       - preprocessshuttlesnode     schedule: daily     executor: multiprocess
data_science:     pipeline:       pipelinename: datascience     schedule: daily     executor: sequential

• Launch the Dagster UI

Start the Dagster UI to monitor and manage your pipelines using the following command:

kedro dagster dev --env <ENV_NAME>

The Dagster UI will be available at http://127.0.0.1:3000.

For a concrete use-case, see the Kedro-Dagster example repository.

How do I use Kedro-Dagster?

Full documentation is available at https://kedro-dagster.readthedocs.io/.

Can I contribute?

We welcome contributions, feedback, and questions:

• Report issues or request features: GitHub Issues
• Join the discussion: GitHub Discussions
• Contributing Guide: CONTRIBUTING.md

If you are interested in becoming a maintainer or taking a more active role, please reach out to Guillaume Tauzin on GitHub Discussions.

Where can I learn more?

• Full documentation: https://kedro-dagster.readthedocs.io/
• GitHub Discussions: https://github.com/stateful-y/kedro-dagster/discussions

For questions and discussions, you can also open a discussion.

License

This project is licensed under the terms of the Apache-2.0 License.

Acknowledgements

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.