Contributing to polyview

We welcome contributions to polyview! If you have an idea for a new feature, have found a bug, or want to improve the documentation, please feel free to submit a pull request directly on the github repository.

Submitting an issue

If planning on submitting an issue, please do your best to follow the following guidelines:

  • Provide a minimal, reproducible example that demonstrates the issue. Additional information can be found here.

  • If the previous point is not applicable, please provide a detailed description of the issue, including the expected behavior and the actual behavior. Include which functions or classes are involved, and any relevant error messages or stack traces.

  • Please include your environment details, such as the version of polyview you are using, your Python version, and any other relevant libraries or dependencies.

Contributing code

If planning on contributing code, here are some additional guidelines to follow.

Coding guidelines

  • Follow the existing code style as closely as possible. This includes naming conventions, formatting, and documentation style. We use black for code formatting and numpy style for docstrings.

  • Format docstrings properly, as they are used for documentation generation by Sphinx. Make sure to include descriptions of parameters, return values, and any relevant reference.

API of polyview objects

Please ensure that all public methods and attributes of the proposed change follows the API of polyview. The API of polyview objects is designed to be consistent with sklearn’s API, which includes methods such as fit, predict, and fit_predict.

Initialization. The initialization of method through __init__ may accept some values that changes the behavior of the method, but should not take as input the data itself. The data should be passed to the fit method. All parameters passed to the __init__ method should have default values, so that the method can be initialized without any arguments.

Fitting. The fit method should be used to train the model on the provided data, and should return the instance itself. Generally, the fit method should take as input a list of views and an optional array of labels.