benmack/oneClass: One-class classification in the absence of test data

oneClass implements the one-class classifiers one-class SVM, biased SVM, and Maxent, as custom functions for the train function of the package caret. Thus, the extensive infrastructure of the caret package can be used for training and analyzing one-class classification models. The infrastructure is further extended by one-class classification specific tools which should help the which are useful to understand and thus improve one-class classifier outcomes in the absence of a representative and complete test data. The package is developed for one-class land cover classification and also handles raster data.

install.packages("remotes")
remotes::install_github("benmack/oneClass")