本文整理汇总了Java中cc.mallet.grmm.learning.ACRF.UnrolledGraph方法的典型用法代码示例。如果您正苦于以下问题：Java ACRF.UnrolledGraph方法的具体用法？Java ACRF.UnrolledGraph怎么用？Java ACRF.UnrolledGraph使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类cc.mallet.grmm.learning.ACRF的用法示例。
在下文中一共展示了ACRF.UnrolledGraph方法的2个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Java代码示例。
示例1: addInstantiatedCliques
import cc.mallet.grmm.learning.ACRF; //导入方法依赖的package包/类
protected void addInstantiatedCliques (ACRF.UnrolledGraph graph, FeatureVectorSequence fvs, LabelsAssignment lblseq)
{
for (int t = 0; t < lblseq.size() - 1; t++) {
Variable var1 = lblseq.varOfIndex (t, lvl1);
Variable var2 = lblseq.varOfIndex (t + 1, lvl2);
assert var1 != null : "Couldn't get label factor "+lvl1+" time "+t;
assert var2 != null : "Couldn't get label factor "+lvl2+" time "+(t+1);
Variable[] vars = new Variable[] { var1, var2 };
FeatureVector fv = fvs.getFeatureVector (t);
ACRF.UnrolledVarSet vs = new ACRF.UnrolledVarSet (graph, this, vars, fv);
graph.addClique (vs);
}
}
示例2: addInstantiatedCliques
import cc.mallet.grmm.learning.ACRF; //导入方法依赖的package包/类
public void addInstantiatedCliques (ACRF.UnrolledGraph graph,
FeatureVectorSequence fvs,
LabelsAssignment lblseq)
{
THashMultiMap fvByWord = constructFvByWord (fvs);
int numSkip = 0;
for (Iterator it = fvByWord.keySet ().iterator (); it.hasNext ();) {
String wordFeature = (String) it.next ();
List infoList = (List) fvByWord.get (wordFeature);
int N = infoList.size ();
if (debug && N > 1) System.err.print ("Processing list of size "+N+" ("+wordFeature+")");
for (int i = 0; i < N; i++) {
for (int j = i + 1; j < N; j++) {
TokenInfo info1 = (TokenInfo) infoList.get (i);
TokenInfo info2 = (TokenInfo) infoList.get (j);
Variable v1 = lblseq.varOfIndex (info1.pos, factor);
Variable v2 = lblseq.varOfIndex (info2.pos, factor);
if (excludeAdjacent && (Math.abs(info1.pos - info2.pos) <= 1)) continue;
Variable[] vars = new Variable[]{v1, v2};
assert v1 != null : "Couldn't get label factor " + factor + " time " + i;
assert v2 != null : "Couldn't get label factor " + factor + " time " + j;
FeatureVector fv = combineFv (wordFeature, info1.fv, info2.fv);
ACRF.UnrolledVarSet clique = new ACRF.UnrolledVarSet (graph, this, vars, fv);
graph.addClique (clique);
numSkip++;
// System.out.println ("Adding "+info1.pos+" --- "+info2.pos);
/* Insanely verbose
if (debug) {
System.err.println ("Combining:\n "+info1.fv+"\n "+info2.fv);
}
*/
}
}
if (debug && N > 1) System.err.println ("...done.");
}
System.err.println ("SimilarTokensTemplate: Total skip edges = "+numSkip);
}