Fast computation of Receiver Operating Characteristic (ROC) curves and Area Under the Curve (AUC) for weighted binary classification problems (weights are example-specific cost values).
new tests compare with geometry and GsymPoint.
simplify vignette/examples (remove AUC from Speed vignette).
always require rather than library the suggested packages.
comparisons with PRROC.
if(require(microbenchmark)) in ROC example for CRAN.
if(require(microbenchmark)) in examples for CRAN.
if(require(microbenchmark)) in vignettes for CRAN.
data(foo, envir=environment()) to pass CRAN checks.
WeightedROC stops with an error when guess has NA values.
WeightedROC stops with an error when there is only one value of label.
WeightedROC returns threshold, FP, FN.
Compare with glmnet::auc.
First version with examples and tests copied from ChIP-seq data analysis project.
Compare with pROC and ROCR packages.