Uses of Interface
opennlp.tools.ml.maxent.quasinewton.Function
Packages that use Function
Package
Description
Package related to ML by means of the Quasi Newton (QN) algorithm.
-
Uses of Function in opennlp.tools.ml.maxent.quasinewton
Classes in opennlp.tools.ml.maxent.quasinewton that implement FunctionModifier and TypeClassDescriptionclass
Evaluate negative log-likelihood and its gradient fromDataIndexer
.class
Evaluate negative log-likelihood and its gradient in parallelstatic class
L2-regularized objectiveFunction
.Methods in opennlp.tools.ml.maxent.quasinewton with parameters of type FunctionModifier and TypeMethodDescriptionstatic void
LineSearch.doConstrainedLineSearch
(Function function, double[] direction, LineSearch.LineSearchResult lsr, double l1Cost, double initialStepSize) Conducts a constrained line search (see section 3.2 in the paper "Scalable Training of L1-Regularized Log-Linear Models", Andrew et al. 2007)static void
LineSearch.doLineSearch
(Function function, double[] direction, LineSearch.LineSearchResult lsr, double initialStepSize) Conducts a backtracking line search (see Nocedal & Wright 2006, Numerical Optimization, p. 37).double[]
Finds the parameters that minimize the objective function.Constructors in opennlp.tools.ml.maxent.quasinewton with parameters of type Function