General guidelines

Thank you for your interest in Anemoi Inference! Please follow the general Anemoi contributing guidelines.

These include general guidelines for contributions to Anemoi, instructions on setting up a development environment, and guidelines on collaboration on GitHub, writing documentation, testing, and code style.

Unit tests

Anemoi-inference includes unit tests that can be executed locally using pytest. For more information on testing, please refer to the general Anemoi testing guidelines.

Classes and code structure

This code base follows a registry pattern. (For more details about code layout, read Adding to Configuration options). New features should be contained in small classes, as a subclass to a larger, more general case.

Always be aware when accepting user input (in our case, from the configuration file) and ensure it is properly validated. If you are enforcing specific expectations on structure, those should be clearly documented in docstrings and in user documentation, with clear error messages if the user does not meet them. Consider validating it using a Pydantic model, which provides specific tooling for validation. These are used in some places in the code base.

All code should include type-hints. These type hints should be reasonably restrictive, rather than using Any.

If you need more complex type hints, these can be created in src/anemoi/inference/types.py and shared between different classes. For example, if you can accept any of a string, list, or dictionary, specify this by creating a union type in that class (or using an existing one, if possible).