Model registry and ML pipeline orchestration for fAIr.
fair-py-ops is the Python package for building ZenML pipelines, validating STAC items, and testing locally. The models/ directory is the single source of truth for base model contributions.
git clone https://github.com/hotosm/fAIr-models.git
cd fAIr-models
just setup
just exampleSee Getting Started for detailed setup, environment options, and running individual examples.
- Getting Started : Installation, local setup, Kubernetes, and running examples
- Architecture : STAC catalog structure, flows, identity model, infrastructure
- Contributing a Model : Guide for adding base models to fAIr
- API Reference : Python package documentation
- Changelog : Release history
Three reference implementations demonstrate the full workflow for each supported task:
| Example | Task | Model | Path |
|---|---|---|---|
| Segmentation | Semantic segmentation | UNet (torchgeo) | examples/segmentation/ |
| Classification | Binary classification | ResNet18 (torchvision) | examples/classification/ |
| Detection | Object detection | YOLOv11n (ultralytics) | examples/detection/ |
Run just to see all recipes. Common commands:
just setup # Install dependencies and set up environment
just example # Run all three example pipelines
just lint # Run Ruff linting and type checking (ty)
just test # Run unit tests
just k8s # Set up Kubernetes dev environment
just local # Switch back to local mode| Concept | Description |
|---|---|
| Base model | Reusable ML blueprint (weights, code, Docker image, STAC item) |
| Local model | Finetuned model produced by ZenML pipeline on user data |
| STAC catalog | Model/dataset registry with MLM and Version extensions |
| ZenML pipeline | Orchestrated training and inference workflows |