This page displays usability scores for each of the methods based on the information presented in associated papers, package documentation and GitHub activity.

Note about usability scores

The usability data was collected prior to 17 December 2020 and may have changed since. Please refer to the various packages, documentation and publications for the most up-to-date information.

1 Paper scores

These scores assess how well publications associated with a method explain how it works and demonstrate its effectiveness.

  • Peer-reviewed - The method is published in a peer-reviewed journal
  • Evaluation of accuracy
    • Datasets - The paper shows the method’s usefulness on more than two (1.0), two (0.5) or one (0.25) real datasets
    • Simulations - The paper quantifies the accuracy of the method using a simulated dataset
  • Robustness - The paper assessed method robustness (to eg. noise, feature selection, down-sampling, parameter changes, stability) in one (0.25) , two (0.5), or more than two (1.0) ways
  • Benchmarking - The paper compares the performance of the method against other published methods

2 Package scores

These scores assess the software packages that implement a method.

  • Open source
    • Code - The packages code is freely available
    • Platform - The code can be run using a freely available platform or programming language
  • Version control - The code is available on a public version controlled repository, such as GitHub
  • Unit tests - The package includes unit tests
  • Tutorial
    • Has tutorial - The package has a vignette or tutorial
    • Errors - The steps in the tutorial can be followed without any issues or errors
    • Scenarios - The tutorial shows the method’s usefulness on more than two (1.0), two (0.5), one (0.25) or no different scenarios
    • Non-native - The tutorial shows how to run the method in a non-native language (i.e. R to Python or vice versa)
  • Function documentation
    • Purpose - The purpose and usage of functions/commands is documented
    • Parameters - The parameters of functions/commands are documented
    • Output - The output of functions/commands is documented
  • GitHub issues
    • Activity - The amount of issues the GitHub repository receives
    • Response - How quickly new GitHub issues are responsed to