public class DeepML
implements weka.core.Randomizable, weka.core.TechnicalInformationHandler
DeepML.java - Deep Multi-label Classification.
Trains an RBM/DBM on the feature space of the training data; then train on it (with the labels) with whichever multi-label classifier is specified.
See: Jesse Read and Jaakko Hollmen. A Deep Interpretation of Classifier Chains. IDA 2014.
The first RBM will have h = d / 2 hidden units, the second h = h / 2, and so on, where d is the number of original (visible) input feature attributes.