mlr (version 2.3)

plotROCRCurves: Visualize binary classification predictions via ROCR ROC curves.

Description

Plot is generated by calling asROCRPrediction, ROCR's performance, then ROCR's plot function.

See these methods in ROCR for further info.

Usage

plotROCRCurves(obj, meas1 = "tpr", meas2 = "fpr", avg = "threshold",
  cols = NULL, ltys = NULL, add.legend = NULL, add.diag = TRUE,
  perf.args = list(), legend.args = list(), task.id = NULL)

Value

[invisible(NULL)].

See Also

Other predict: asROCRPrediction; getProbabilities; plotViperCharts; predict.WrappedModel; setPredictThreshold; setPredictType

Other roc: asROCRPrediction; plotViperCharts

Examples

lrn1 = makeLearner("classif.logreg", predict.type = "prob")
lrn2 = makeLearner("classif.rpart", predict.type = "prob")
b = benchmark(list(lrn1, lrn2), pid.task)
z = plotROCRCurves(b)