Tune prediction threshold.
Optimizes the threshold of predictions based on probabilities. Works for classification and multilabel tasks. Uses BBmisc::optimizeSubInts for normal binary class problems and GenSA::GenSA for multiclass and multilabel problems.
tuneThreshold(pred, measure, task, model, nsub = 20L, control = list())
pred |
(Prediction) |
measure |
(Measure) |
task |
(Task) |
model |
(WrappedModel) |
nsub |
( |
control |
(list) |
(list). A named list with with the following components:
th is the optimal threshold, perf the performance value.
Other tune:
TuneControl,
getNestedTuneResultsOptPathDf(),
getNestedTuneResultsX(),
getResamplingIndices(),
getTuneResult(),
makeModelMultiplexerParamSet(),
makeModelMultiplexer(),
makeTuneControlCMAES(),
makeTuneControlDesign(),
makeTuneControlGenSA(),
makeTuneControlGrid(),
makeTuneControlIrace(),
makeTuneControlMBO(),
makeTuneControlRandom(),
makeTuneWrapper(),
tuneParams()
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.