Package opennlp.tools.ml.model
package opennlp.tools.ml.model
Package related to ML models and feature selection techniques.
-
ClassDescriptionAbstract
DataIndexerimplementation for collecting event and context counts used in training.A basicMaxentModelimplementation.An abstract, basic implementation of a model reader.An abstract, basic implementation of a model writer.ADataReaderthat reads files from a binary format.A maxent event representation which we can use to sort based on the predicates indexes contained in the events.A maxent predicate representation which we can use to sort based on the outcomes.Class which associates a real valued parameter or expected value with a particular contextual predicate or feature.Represents an indexer which compresses events in memory and performs feature selection.A factory that producesDataIndexerinstances.Describes generic ways to read data from aDataInputStream.This class encapsulates the variables used in producing probabilities from a model and facilitates passing these variables to the eval method.The context of a decision point during training.Class for using a file ofeventsas anevent stream.An genericAbstractModelReaderimplementation.An genericAbstractModelWriterimplementation.A hash sum basedAbstractObjectStreamimplementation.Interface for maximum entropy models.A helper class that handles Strings with more than 64k (65535 bytes) in length.An extension ofContextused to store parameters or expected values associated with this context which can be updated or assigned.ADataReaderimplementation based onObjectInputStream.ADataIndexerfor maxent model data which handles cutoffs for uncommon contextual predicates and provides a unique integer index for each of the predicates.ADataIndexerfor maxent model data which handles cutoffs for uncommon contextual predicates and provides a unique integer index for each of the predicates and maintains event values.A genericDataReaderimplementation for plain text files.This interface allows one to implement a prior distribution for use in maximum entropy model training.Class for using a file of real-valuedeventsas anevent stream.Sequence<T>Class which models a sequence.A classification model that can label an inputSequence.Interface for streams ofsequencesused to train sequence models.Class which turns aSequenceStreaminto an event stream.Collecting event and context counts by making two passes over the events.Provide a maximum entropy model with a uniformPrior.