Package calhoun.analysis.crf.solver

training and inference algorithms used in the CRF engine

See:
Description

Interface Summary

CacheProcessor

interface to implementations of feature caches.

Class Summary

CacheProcessor.FeatureEvaluation	This class holds the feature evaluations for a given position, or position/length combination.
CacheProcessor.LengthFeatureEvaluation
CacheProcessor.SolverSetup
CacheProcessor.StatePotentials
CacheProcessorBasic	basic functionality common to most cache processors.
CacheProcessorDeluxe	A policy based cache processor.
FixedWeightOptimizer	a dummy optimizer that just fixed the weights at values specified in the configuration.
LogFiles
LookbackBuffer	This object holds information about previous positions during the computation of betas and expectations.
MaximumExpectedAccuracySemiMarkovGradient	computes an objective function which is the expected value of a local path similarity score on a semi-Markov model.
MaximumLikelihoodGradient	computes the likelihood of the true path for a Markov CRF.
MaximumLikelihoodSemiMarkovGradient	computes the likelihood of the true path for a semi-Markov CRF.
NoCachingCacheProcessor	a dummy cache processor that fulfills the interface but doesn't cache.
RecyclingBuffer<T>	Utility class that implements a FIFO buffer with constant time adds, removes, and accesses to any position using a circular buffer in an array.
SeededOptimizer	an optimizer that uses the weights from another model as the seed for a new optimization.
SemiMarkovViterbi	viterbi algorithm for semi-Markov CRFs.
SemiMarkovViterbiNoCache	viterbi algorithm for semi-Markov CRFs.
SimplexOptimizer	uses a nelder-mead algorithm (the simplex method) to do a general function optimization objective function.
StandardOptimizer	uses a L-BFGS algorithm to optimize the objective function.
TwoPassOptimizer	a two pass optimizer that does an initial optimization and then uses the weights generated from that as the start of a second optimization.
Viterbi

Package calhoun.analysis.crf.solver Description

training and inference algorithms used in the CRF engine