| AbstractPriorMultiProbabilisticMeasure |
The prior measure depends only on the prediction made.
|
| AbstractSignificanceBasedMeasure |
Base class for measures that depend on the significance level.
|
| AggregatedObservedMeasure |
Maintains a running average of an observed measure.
|
| AggregatedObservedMeasures |
Maintains running averages for a set of observed measures.
|
| AggregatedPriorMeasure |
Maintains a running average of a prior measure.
|
| AggregatedPriorMeasures |
Maintains running averages for a set of prior measures.
|
| ExcessCriterion |
The E/Excess criterion is a prior efficiency measure based on how much the
size of the label set exceeds 1.
|
| FuzzinessCriterion |
The F/Fuzziness criterion is a prior efficiency measure based on the
difference between the sum of all p-values and the largest p-value.
|
| MultipleCriterion |
The M/Multiple criterion is a prior efficiency measure based on the size of
the label set.
|
| MultiProbabilisticLowerBound |
This prior measure returns the predicted lower bound probability.
|
| MultiProbabilisticUpperBound |
This prior measure returns the predicted upper bound probability.
|
| NumberCriterion |
The N/Number/AvgC criterion is a prior efficiency measure based on the size
of the label set.
|
| ObservedAccuracy |
The Observed Accuracy is the fraction of predictions that include the
true label in their label set.
|
| ObservedExcessCriterion |
The OE/Observed Excess criterion is an observed efficiency measure based on
the number of false labels in the label set.
|
| ObservedFuzzinessCriterion |
The OF/Observed Fuzziness criterion is a prior efficiency measure based on
the sum of all p-values for false labels.
|
| ObservedMultipleCriterion |
The OM/Observed Multiple criterion is an observed efficiency measure based on
the number of false labels in the label set.
|
| ObservedOneCCriterion |
The Observed OneC criterion is the fraction of predictions that only have
the true label in their label set.
|
| ObservedProbabilityLogLoss |
The observed log loss measure depends on the prediction made and the true label.
|
| ObservedProbabilitySquareLoss |
The observed square loss measure depends on the prediction made and the true label.
|
| ObservedUnconfidenceCriterion |
The UO/Observed Unconfidence criterion is a prior efficiency measure based
on the (un)confidence of the point prediction.
|
| OneCCriterion |
The OneC is the fraction of predictions that only have one label in their
label set.
|
| SumCriterion |
The S/Sum criterion is a prior efficiency measure based on the sum of the
p-values.
|
| UnconfidenceCriterion |
The U/Unconfidence criterion is a prior efficiency measure based on the
(un)confidence of the point prediction.
|