dependencypipe2o.java

MSTParser是以最大生成树理论为基础的判别式依存句法分析器。它将一科依存树的得分看作是 所有依存关系的得分的总和

		}
	    }

	    out.writeInt(-3);
						
	    for(FeatureVector curr = inst.fv; curr.next != null; curr = curr.next)
		out.writeInt(curr.index);

	    out.writeInt(-4);
	    out.writeObject(inst.sentence);
	    out.writeInt(-5);
	    out.writeObject(inst.pos);
	    out.writeInt(-6);
	    out.writeObject(inst.labs);
	    out.writeInt(-7);
	    out.writeObject(inst.actParseTree);
			
	    out.writeInt(-1);
	    out.reset();

	} catch (IOException e) {}
		
    }

    public DependencyInstance getFeatureVector(ObjectInputStream in,
					       DependencyInstance inst,
					       FeatureVector[][][] fvs,
					       double[][][] probs,
					       FeatureVector[][][] fvs_trips,
					       double[][][] probs_trips,
					       FeatureVector[][][] fvs_sibs,
					       double[][][] probs_sibs,
					       FeatureVector[][][][] nt_fvs,
					       double[][][][] nt_probs,
					       Parameters params) throws IOException {
	int length = inst.length;
		
	// Get production crap.		
	for(int w1 = 0; w1 < length; w1++) {
	    for(int w2 = w1+1; w2 < length; w2++) {
				
		for(int ph = 0; ph < 2; ph++) {

		    FeatureVector prodFV = new FeatureVector(-1,-1.0,null);
					
		    int indx = in.readInt();
		    while(indx != -2) {
			prodFV = new FeatureVector(indx,1.0,prodFV);
			indx = in.readInt();
		    }
					
		    double prodProb = params.getScore(prodFV);
		    fvs[w1][w2][ph] = prodFV;
		    probs[w1][w2][ph] = prodProb;
		}
	    }
			
	}
	int last = in.readInt();
	if(last != -3) { System.out.println("Error reading file."); System.exit(0); }

	if(labeled) {
	    for(int w1 = 0; w1 < length; w1++) {
		
		for(int t = 0; t < types.length; t++) {
		    String type = types[t];
		    
		    for(int ph = 0; ph < 2; ph++) {						
			
			for(int ch = 0; ch < 2; ch++) {						
			    
			    FeatureVector prodFV = new FeatureVector(-1,-1.0,null);
			    
			    int indx = in.readInt();
			    while(indx != -2) {
				prodFV = new FeatureVector(indx,1.0,prodFV);
				indx = in.readInt();
			    }
			    
			    double nt_prob = params.getScore(prodFV);
			    nt_fvs[w1][t][ph][ch] = prodFV;
			    nt_probs[w1][t][ph][ch] = nt_prob;
			    
			}
		    }
		}
		
	    }
	    last = in.readInt();
	    if(last != -3) { System.out.println("Error reading file."); System.exit(0); }
	}

	for(int w1 = 0; w1 < length; w1++) {
	    for(int w2 = w1; w2 < length; w2++) {
		for(int w3 = w2+1; w3 < length; w3++) {
		    FeatureVector prodFV = new FeatureVector(-1,-1.0,null);
		    
		    int indx = in.readInt();
		    while(indx != -2) {
			prodFV = new FeatureVector(indx,1.0,prodFV);
			indx = in.readInt();
		    }

		    double prodProb = params.getScore(prodFV);
		    fvs_trips[w1][w2][w3] = prodFV;
		    probs_trips[w1][w2][w3] = prodProb;
		    

		}
	    }
	    for(int w2 = w1; w2 >= 0; w2--) {
		for(int w3 = w2-1; w3 >= 0; w3--) {
		    FeatureVector prodFV = new FeatureVector(-1,-1.0,null);
		    
		    int indx = in.readInt();
		    while(indx != -2) {
			prodFV = new FeatureVector(indx,1.0,prodFV);
			indx = in.readInt();
		    }

		    double prodProb = params.getScore(prodFV);
		    fvs_trips[w1][w2][w3] = prodFV;
		    probs_trips[w1][w2][w3] = prodProb;

		}
	    }
	}
			
	last = in.readInt();
	if(last != -3) { System.out.println("Error reading file."); System.exit(0); }
			
	for(int w1 = 0; w1 < length; w1++) {
	    for(int w2 = 0; w2 < length; w2++) {
		for(int wh = 0; wh < 2; wh++) {
		    if(w1 != w2) {
			FeatureVector prodFV = new FeatureVector(-1,-1.0,null);
			
			int indx = in.readInt();
			while(indx != -2) {
			    prodFV = new FeatureVector(indx,1.0,prodFV);
			    indx = in.readInt();
			}

			double prodProb = params.getScore(prodFV);
			fvs_sibs[w1][w2][wh] = prodFV;
			probs_sibs[w1][w2][wh] = prodProb;

		    }
		}
	    }
	}

	last = in.readInt();
	if(last != -3) { System.out.println("Error reading file."); System.exit(0); }

	FeatureVector nfv = new FeatureVector(-1,-1.0,null);
	int next = in.readInt();
	while(next != -4) {
	    nfv = new FeatureVector(next,1.0,nfv);
	    next = in.readInt();
	}

	String[] toks = null;
	String[] pos = null;
	String[] labs = null;
	String actParseTree = null;
	try {
	    toks = (String[])in.readObject();
	    next = in.readInt();
	    pos = (String[])in.readObject();
	    next = in.readInt();
	    labs = (String[])in.readObject();
	    next = in.readInt();
	    actParseTree = (String)in.readObject();
	    next = in.readInt();
	}
	catch(ClassNotFoundException e) { System.out.println("Error reading file."); System.exit(0); }
		
	if(next != -1) { System.out.println("Error reading file."); System.exit(0); }

	DependencyInstance pti = new DependencyInstance(toks,pos,labs,nfv);
	pti.actParseTree = actParseTree;
	return pti;
		
    }
		
    public void getFeatureVector(DependencyInstance inst,
				 FeatureVector[][][] fvs,
				 double[][][] probs,
				 FeatureVector[][][] fvs_trips,
				 double[][][] probs_trips,
				 FeatureVector[][][] fvs_sibs,
				 double[][][] probs_sibs,
				 FeatureVector[][][][] nt_fvs,
				 double[][][][] nt_probs, Parameters params) {

	String[] toks = inst.sentence;
	String[] pos = inst.pos;
	String[] labs = inst.labs;
		
	String[] posA = new String[pos.length];
	for(int i = 0; i < pos.length; i++) {
	    posA[i] = pos[i].substring(0,1);
	}

	// Get production crap.		
	for(int w1 = 0; w1 < toks.length; w1++) {
	    for(int w2 = w1+1; w2 < toks.length; w2++) {
				
		for(int ph = 0; ph < 2; ph++) {
		    boolean attR = ph == 0 ? true : false;
		    
		    int childInt = attR ? w2 : w1;
		    int parInt = attR ? w1 : w2;
		    
		    FeatureVector prodFV = createFeatureVector(toks,pos,posA,w1,w2,attR,
							       new FeatureVector(-1,-1.0,null));
										
		    double prodProb = params.getScore(prodFV);
		    fvs[w1][w2][ph] = prodFV;
		    probs[w1][w2][ph] = prodProb;
		}
	    }
			
	}

	if(labeled) {
	    for(int w1 = 0; w1 < toks.length; w1++) {
		
		for(int t = 0; t < types.length; t++) {
		    String type = types[t];
		    
		    for(int ph = 0; ph < 2; ph++) {						
			boolean attR = ph == 0 ? true : false;
			
			for(int ch = 0; ch < 2; ch++) {						
			    boolean child = ch == 0 ? true : false;						
			    
			    FeatureVector prodFV = createFeatureVector(toks,pos,posA,w1,
								       type,attR,child,
								       new FeatureVector(-1,-1.0,null));
			    
			    double nt_prob = params.getScore(prodFV);
			    nt_fvs[w1][t][ph][ch] = prodFV;
			    nt_probs[w1][t][ph][ch] = nt_prob;
			    
			}
		    }
		}
		
	    }
	}

		
	for(int w1 = 0; w1 < toks.length; w1++) {
	    for(int w2 = w1; w2 < toks.length; w2++) {
		for(int w3 = w2+1; w3 < toks.length; w3++) {
		    FeatureVector prodFV = createFeatureVector(toks,pos,posA,w1,w2,w3,
							       new FeatureVector(-1,-1.0,null));
		    double prodProb = params.getScore(prodFV);
		    fvs_trips[w1][w2][w3] = prodFV;
		    probs_trips[w1][w2][w3] = prodProb;
		}
	    }
	    for(int w2 = w1; w2 >= 0; w2--) {
		for(int w3 = w2-1; w3 >= 0; w3--) {
		    FeatureVector prodFV = createFeatureVector(toks,pos,posA,w1,w2,w3,
							       new FeatureVector(-1,-1.0,null));
		    double prodProb = params.getScore(prodFV);
		    fvs_trips[w1][w2][w3] = prodFV;
		    probs_trips[w1][w2][w3] = prodProb;
		}
	    }
	}
			
	for(int w1 = 0; w1 < toks.length; w1++) {
	    for(int w2 = 0; w2 < toks.length; w2++) {
		for(int wh = 0; wh < 2; wh++) {
		    if(w1 != w2) {
			FeatureVector prodFV = createFeatureVectorSib(toks,pos,w1,w2,wh == 0,
								      new FeatureVector(-1,-1.0,null));
			double prodProb = params.getScore(prodFV);
			fvs_sibs[w1][w2][wh] = prodFV;
			probs_sibs[w1][w2][wh] = prodProb;
		    }
		}
	    }
	}
    }
		
}