Welcome to correctipy’s documentation!

correctipy is a python package for calculating corrected test statistics for comparing machine learning models on correlated samples. This set of documentation contains code examples and mathematical details on the test statistics that have been implemented.

Contents

Contents:

• Quickstart
  • Installation
  • Usage
• Test statistic mathematics
  • Random subsampling correction
  • k-fold cross-validation correction
  • Repeated k-fold cross-validation correction