|File Search||Catalog||Content Search|
ROCR features: ROC curves, precision/recall plots, lift charts, cost curves, custom curves by freely selecting one performance measure for the x axis and one for the y axis, handling of data from cross-validation or bootstrapping, curve averaging (vertically, horizontally, or by threshold), standard error bars, box plots, curves that are color-coded by cutoff, printing threshold values on the curve, tight integration with Rs plotting facilities (making it easy to adjust plots or to combine multiple plots), fully customizable, easy to use (only 3 commands).
Performance measures that ROCR knows: Accuracy, error rate, true positive rate, false positive rate, true negative rate, false negative rate, sensitivity, specificity, recall, positive predictive value, negative predictive value, precision, fallout, miss, phi correlation coefficient, Matthews correlation coefficient, mutual information, chi square statistic, odds ratio, lift value, precision/recall F measure, ROC convex hull, area under the ROC curve, precision/recall break-even point, calibration error, mean cross-entropy, root mean squared error, SAR measure, expected cost, explicit cost.
Getting started with ROCR: -------------------------- * After installation (cf. file 'INSTALL'), and starting R, load the package with 'library(ROCR)'. * For a short overview of ROCR: demo(ROCR) * For an overview of ROCR's online help: help(package=ROCR) * ROCR help pages: help(prediction) help(performance) help(plot.performance) help('prediction-class') help('performance-clas more»
You need the R package 'RUnit' to run the unit tests for ROCR. You can start the test suite by typing: R --vanilla < unittests/runit.ROCR.R from _this_ directory.
This file documents changes and updates to the ROCR package. Version 1.0-4 (Dec 08, 2009) - fixes bug with 1.0-3 that prevented plot arguments getting passed through Version 1.0-3 - adapted to more strict R CMD CHECK rules in R > 2.9 Version 1.0-2 (Jan 27, 2007) - fixed minor bug in 'prediction' function concerning the optional parameter 'label.ordering' (thanks to Robert Perdisci for notify more»
r-cran-rocr (1.0-4-3) unstable; urgency=low * Support depending on R pre-releases (debian/rules, more»