Skip to content

Contributing to kima

If you are interested in contributing to the development of kima, thank you! 🙏
Feedback, bug reports, help with documentation, or code contributions are very welcome.

Issues and Feature Requests

If you find a bug or run into some unexpected behaviour, feel free to open an issue on GitHub.

Sometimes, your specific analysis and setup will be hard to reproduce, so if you can provide a minimal example, or representative data where the problem occurs, it really helps! Feel free to also include a saved version of the model in a pickle file, for example.

Writing documentation

kima, like other software created by researchers and for research, suffers from a lack of up-to-date comprehensive documentation. This is a shame, but it's almost unavoidable. First, we are not software engineers, and therefore lack the skills and knowledge necessary to produce and document the code at a professional level. Second, none of the incentives in our professional life as an astronomer align with writing documentation, because the system forces a mindless focus on writing papers and securing the next position. (rant over 🗣)

If you find an undocumented piece of code, or a mistake in the documentation, please open an issue or create a pull request to correct it.

Adding a new feature

If kima doesn't have the feature you need, feel free to propose an implementation. For example, the BINARIESmodel and the ExponentialRayleighMixture distribution, among other things, were added by the community (see this paper and this paper; thank you Tom and Adam!).

Please don't be scared by the C++ code. You can probably go a long way just by mimicking one of the existing models or data types.