cmu.arktweetnlp.impl

Class ModelSentence

java.lang.Object

cmu.arktweetnlp.impl.ModelSentence

public class ModelSentence
extends java.lang.Object

One sequence structure -- typically, for one sentence This is the model's view of a sentence -- only deals with non-textual numberized versions of everything

Field Summary

Fields

Modifier and Type	Field and Description
double[]	confidences
int[]	edgeFeatures Runtime observed, Trainingtime observed (for MEMM).
int[]	labels Runtime inferred, Trainingtime observed.
java.util.ArrayList<java.util.ArrayList<edu.stanford.nlp.util.Pair<java.lang.Integer,java.lang.Double>>>	observationFeatures Runtime observed, Trainingtime observed.
int	T

Constructor Summary

Constructors

Constructor and Description
ModelSentence(int T)

Method Summary

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

T

public int T

labels

public int[] labels

Runtime inferred, Trainingtime observed. dim T

confidences

public double[] confidences

observationFeatures

public java.util.ArrayList<java.util.ArrayList<edu.stanford.nlp.util.Pair<java.lang.Integer,java.lang.Double>>> observationFeatures

Runtime observed, Trainingtime observed. This is an array-of-arrays of (featID, featValue) pairs. dim T, then variable nnz per t.

edgeFeatures

public int[] edgeFeatures

Runtime observed, Trainingtime observed (for MEMM). dim T st: edgeFeatures[t] = ID of label@(t-1). values in 0..(N_labels-1), plus extra higher numbers for markers (see Model)

Constructor Detail

ModelSentence

public ModelSentence(int T)