The predict framework is designed to work with dataset where rows correspond to observations and columns to descriptors. PAMR wants it the other way, and also to have the fitting set response vector supplied in a list with the descriptors. This function applies a standard pre-processing function and then reformats the result to satisfy PAMR.
pre_pamr(data)Fitting and testing data sets, as returned by
pre_split.
A list with fitting and testing sets, formatted the way pamr wants them.
pre_pamr must be run last if chained with other pre-processing
functions, since it substantially reshapes the data.
Robert Tibshirani, Trevor Hastie, Balasubramanian Narasimhan and Gilbert Chu (2002) Diagnosis of multiple cancer types by shrunken centroids of gene expression. URL www.pnas.org/cgi/doi/10.1073/pnas.082099299