Skip to contents

R6 base class defining the protocol that all multivariate analysis methods must implement. Methods declare their capabilities, implement fit(), and optionally provide resampling hooks and UI metadata.

Details

This is the core abstraction for method-neutral multivariate neuroimaging analysis. PLS, CPCA, CCA, ICA, and other methods all implement this protocol. The framework uses capabilities to adapt the builder API, Shiny UI, and result accessors to each method automatically.