|
Class Summary |
| ExampleBasedFMeasureOptimizer |
This class takes the marginal probabilities estimated for each label by a multi-label learner and transforms them into a bipartition which is approximately optimal for example-based FMeasure. |
| Meta |
Base class for instance-based prediction of a bipartition from the labels'
scores |
| MetaLabeler |
Class implementing the MetaLabeler algorithm. |
| MLPTO |
Class that implements the Multi Label Probabilistic Threshold Optimizer (MLTPTO). |
| OneThreshold |
Class that estimates a single threshold for all labels and examples. |
| RCut |
Classs that implements RCut(Rank-based cut). |
| SCut |
Class that implements the SCut method (Score-based local optimization). |
| ThresholdPrediction |
Class that learns to predict a different threshold per exampleFor more information, see
Elisseeff, Andre, Weston, Jason: A kernel method for multi-labelled classification. |