Package com.clearnlp.classification.model

Examples of com.clearnlp.classification.model.StringModel

81
82
83
84
85
86
87
88
89
90
91
   * @param labelCutoff the label count cutoff (exclusive).
   * @param featureCutoff the feature count cutoff (exclusive).
   */
  public StringTrainSpace(boolean hasWeight, int labelCutoff, int featureCutoff)
  {
    super(new StringModel(), hasWeight);
   
    s_model     = (StringModel)m_model;
    l_cutoff    = labelCutoff;
    f_cutoff    = featureCutoff;
    s_instances = Lists.newArrayList();
View Full Code Here
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
  {
    BufferedReader    fin = UTInput.createBufferedFileReader(testFile);
    PrintStream      fout = UTOutput.createPrintBufferedFileStream(outputFile);
    ObjectInputStream  in = new ObjectInputStream(new BufferedInputStream(new FileInputStream(modelFile)));
    SparseModel    pModel = null;
    StringModel    sModel = null;
   
    switch (vectorType)
    {
    case AbstractTrainSpace.VECTOR_SPARSE:
      pModel = (SparseModel)in.readObject(); break;
    case AbstractTrainSpace.VECTOR_STRING:
      sModel = (StringModel)in.readObject(); break;
    }
   
    in.close();
   
    boolean hasWeight = AbstractTrainSpace.hasWeight(vectorType, testFile);
    int correct = 0, total = 0;
    StringPrediction r = null;
    String line, label = null;
   
    System.out.print("Predicting");
   
    while ((line = fin.readLine()) != null)
    {
      if (vectorType == AbstractTrainSpace.VECTOR_SPARSE)
      {
        Pair<String,SparseFeatureVector> sp = SparseTrainSpace.toInstance(line, hasWeight);
        r = pModel.predictBest(sp.o2);
        label = sp.o1;
      }
      else
      {
        StringInstance ss = StringTrainSpace.toInstance(line, hasWeight);
        r = sModel.predictBest(ss.getFeatureVector());
        label = ss.getLabel();
      }
     
      fout.println(r.label+" "+r.score);
      if (r.label.equals(label))  correct++;
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
   * @param labelCutoff the label count cutoff (exclusive).
   * @param featureCutoff the feature count cutoff (exclusive).
   */
  public StringTrainSpace(boolean hasWeight, int labelCutoff, int featureCutoff)
  {
    super(new StringModel(), hasWeight);
   
    s_model     = (StringModel)m_model;
    l_cutoff    = labelCutoff;
    f_cutoff    = featureCutoff;
    s_instances = Lists.newArrayList();
View Full Code Here
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
  {
    BufferedReader    fin = UTInput.createBufferedFileReader(testFile);
    PrintStream      fout = UTOutput.createPrintBufferedFileStream(outputFile);
    ObjectInputStream  in = new ObjectInputStream(new BufferedInputStream(new FileInputStream(modelFile)));
    SparseModel    pModel = null;
    StringModel    sModel = null;
   
    switch (vectorType)
    {
    case AbstractTrainSpace.VECTOR_SPARSE:
      pModel = (SparseModel)in.readObject(); break;
    case AbstractTrainSpace.VECTOR_STRING:
      sModel = (StringModel)in.readObject(); break;
    }
   
    in.close();
   
    boolean hasWeight = AbstractTrainSpace.hasWeight(vectorType, testFile);
    int correct = 0, total = 0;
    StringPrediction r = null;
    String line, label = null;
   
    System.out.print("Predicting");
   
    while ((line = fin.readLine()) != null)
    {
      if (vectorType == AbstractTrainSpace.VECTOR_SPARSE)
      {
        Pair<String,SparseFeatureVector> sp = SparseTrainSpace.toInstance(line, hasWeight);
        r = pModel.predictBest(sp.o2);
        label = sp.o1;
      }
      else
      {
        StringInstance ss = StringTrainSpace.toInstance(line, hasWeight);
        r = sModel.predictBest(ss.getFeatureVector());
        label = ss.getLabel();
      }
     
      fout.println(r.label+" "+r.score);
      if (r.label.equals(label))  correct++;
View Full Code Here
TOP

Related Classes of com.clearnlp.classification.model.StringModel

  • com.clearnlp.classification.train.StringTrainSpace
  • com.clearnlp.classification.vector.SparseFeatureVector
  • com.clearnlp.classification.vector.StringFeatureVector
  • com.clearnlp.run.LiblinearPredict
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.