NOTE: This feature is in private preview. The interface/APIs may change and no formal support is available during the preview. However, you can still create new production-grade ML projects using the stack. If interested in trying it out, please fill out this form, and you’ll be contacted by a Databricks representative.
This repo provides a customizable stack for starting new ML projects on Databricks that follow production best-practices out of the box.
The default stack in this repo includes three modular components:
| Component | Description | Why it's useful |
|---|---|---|
| ML Code | Example ML project structure, with unit tested Python modules and notebooks | Quickly iterate on ML problems, without worrying about refactoring your code into tested modules for productionization later on. |
| ML Resource Configs as Code | ML pipeline resources (training and batch inference jobs, etc) defined through databricks CLI bundles | Govern, audit, and deploy changes to your ML resources (e.g. "use a larger instance type for automated model retraining") through pull requests, rather than adhoc changes made via UI. |
| CI/CD | GitHub Actions or Azure DevOps workflows to test and deploy ML code and resources | Ship ML code faster and with confidence: ensure all production changes are performed through automation and that only tested code is deployed to prod |
Your organization can use the default stack as is or customize it as needed, e.g. to add/remove components or adapt individual components to fit your organization's best practices. See the stack customization guide for more details.
Using Databricks MLOps stacks, data scientists can quickly get started iterating on ML code for new projects while ops engineers set up CI/CD and ML service state management, with an easy transition to production. You can also use MLOps stacks as a building block in automation for creating new data science projects with production-grade CI/CD pre-configured.
See the FAQ for questions on common use cases.
See this page for detailed description and diagrams of the ML pipeline structure defined in the default stack.
- Python 3.8+
- Cookiecutter Python package >= 2.1.0: This can be installed with pip:
$ pip install 'cookiecutter>=2.1.0'To create a new project, run:
cookiecutter https://github.com/databricks/mlops-stack
This will prompt for parameters for project initialization. Some of these parameters are required to get started:
project_name: name of the current projectroot_dir__update_if_you_intend_to_use_monorepo: name of the root directory. The root directory name should be specified if monorepo will be used for the project. Otherwise, user can leave it blank to use the name of the current project as root directory name.cloud: Cloud provider you use with Databricks (AWS, Azure, or GCP)cicd_platform: CI/CD platform of choice (GitHub Actions or GitHub Actions for GitHub Enterprise Servers or Azure DevOps)
Others must be correctly specified for CI/CD to work, and so can be left at their default values until you're
ready to productionize a model. We recommend specifying any known parameters upfront (e.g. if you know
databricks_staging_workspace_host, it's better to specify it upfront):
databricks_staging_workspace_host: URL of staging Databricks workspace, used to run CI tests on PRs and preview config changes before they're deployed to production. We encourage granting data scientists working on the current ML project non-admin (read) access to this workspace, to enable them to view and debug CI test resultsdatabricks_prod_workspace_host: URL of production Databricks workspace. We encourage granting data scientists working on the current ML project non-admin (read) access to this workspace, to enable them to view production job status and see job logs to debug failures.default_branch: Name of the default branch, where the prod and staging ML resources are deployed from and the latest ML code is staged.release_branch: Name of the release branch. The production jobs (model training, batch inference) defined in this repo pull ML code from this branch.read_user_group: User group name to give READ permissions to for project resources (ML jobs, integration test job runs, and machine learning resources). A group with this name must exist in both the staging and prod workspaces. Defaults to "users", which grants read permission to all users in the staging/prod workspaces. You can specify a custom group name e.g. to restrict read permissions to members of the team working on the current ML project.include_feature_store: If selected, will provide Databricks Feature Store stack components including: project structure and sample feature Python modules, feature engineering notebooks, ML resource configs to provision and manage Feature Store jobs, and automated integration tests covering feature engineering and training.
See the generated README.md for next steps!
We recommend using separate dev/staging/prod Databricks workspaces for stronger isolation between environments. For example, Databricks REST API rate limits are applied per-workspace, so if using Databricks Model Serving, using separate workspaces can help prevent high load in staging from DOSing your production model serving endpoints.
However, you can run the stack against just a single workspace, against a dev and
staging/prod workspace, etc. Just supply the same workspace URL for
databricks_staging_host and databricks_prod_host. If you go this route, we
recommend using different service principals to manage staging vs prod resources,
to ensure that CI workloads run in staging cannot interfere with production resources.
Yes. Currently, you can instantiate a new project from the stack and copy relevant components
into your existing project to productionize it. The stack is modularized, so
you can e.g. copy just the GitHub Actions workflows under .github or ML resource configs
under {{cookiecutter.root_dir__update_if_you_intend_to_use_monorepo}}/{{cookiecutter.project_name_alphanumeric_underscore}}/databricks-resources
and {{cookiecutter.root_dir__update_if_you_intend_to_use_monorepo}}/{{cookiecutter.project_name_alphanumeric_underscore}}/bundle.yml into your existing project.
For this use case, we recommend instantiating the full stack via cookiecutter
and copying the relevant stack subdirectories. For example, all ML resource configs
are defined under {{cookiecutter.root_dir__update_if_you_intend_to_use_monorepo}}/{{cookiecutter.project_name_alphanumeric_underscore}}/databricks-resources
and {{cookiecutter.root_dir__update_if_you_intend_to_use_monorepo}}/{{cookiecutter.project_name_alphanumeric_underscore}}/bundle.yml, while CI/CD is defined e.g. under .github
if using GitHub Actions, or under .azure if using Azure DevOps.
Yes. We provide the default stack in this repo as a production-friendly starting point for MLOps. However, in many cases you may need to customize the stack to match your organization's best practices. See the stack customization guide for details on how to do this.
Since MLOps Stacks is based on databricks CLI bundles, it's not limited only to ML workflows and assets - it works for assets across the Databricks Lakehouse. For instance, while the existing ML code samples contain feature engineering, training, model validation, deployment and batch inference workflows, you can use it for Delta Live Tables pipelines as well.
Please provide feedback (bug reports, feature requests, etc) via GitHub issues.
We welcome community contributions. For substantial changes, we ask that you first file a GitHub issue to facilitate discussion, before opening a pull request.
This stack is implemented as a cookiecutter template
that generates new projects given user-supplied parameters. Parametrized project code can be found under
the {{cookiecutter.root_dir__update_if_you_intend_to_use_monorepo}} directory.
To run tests, install actionlint,
databricks CLI, npm, and
act, then install the Python
dependencies listed in dev-requirements.txt:
pip install -r dev-requirements.txt
NOTE: This section is for open-source developers contributing to the default stack
in this repo. If you are working on an ML project using the stack (e.g. if you ran cookiecutter
to start a new project), see the README.md within your generated
project directory for detailed instructions on how to make and test changes.
Run unit tests:
pytest tests
Run all tests (unit and slower integration tests):
pytest tests --large
Run integration tests only:
pytest tests --large-only
When making changes to the stack, it can be convenient to see how those changes affect an actual new ML project created from the stack. To do this, you can create an example project from your local checkout of the stack, and inspect its contents/run tests within the project.
We provide example project configs for Azure (using both GitHub and Azure DevOps) and AWS (using GitHub) under tests/example-project-configs.
To create an example Azure project, using Azure DevOps as the CI/CD platform, run the following from the desired parent directory
of the example project:
# Note: update MLOPS_STACK_PATH to the path to your local checkout of the stack
MLOPS_STACK_PATH=~/mlops-stack
cookiecutter "$MLOPS_STACK_PATH" --config-file "$MLOPS_STACK_PATH/tests/example-project-configs/azure/azure-devops.yaml" --no-input --overwrite-if-exists
To create an example AWS project, using GitHub Actions for CI/CD, run:
# Note: update MLOPS_STACK_PATH to the path to your local checkout of the stack
MLOPS_STACK_PATH=~/mlops-stack
cookiecutter "$MLOPS_STACK_PATH" --config-file "$MLOPS_STACK_PATH/tests/example-project-configs/aws/aws-github.yaml" --no-input --overwrite-if-exists