svm.svm_get_labels(model,labels);
public int[] getLabels() { int res[] = new int[model.nr_class]; svm.svm_get_labels(model, res); return res; }
public AbstractClassifier(File fn, int len) { try{ svmCls = svm.svm_load_model(fn.getAbsolutePath()); int[] labels = new int[2]; svm.svm_get_labels(svmCls, labels); clsIndex = labels[0]==1 ? 0 : 1; }catch(IOException e){ e.printStackTrace(); } }
public AbstractClassifier(File fn, int len) { try{ svmCls = svm.svm_load_model(fn.getAbsolutePath()); int[] labels = new int[2]; svm.svm_get_labels(svmCls, labels); clsIndex = labels[0]==1 ? 0 : 1; }catch(IOException e){ e.printStackTrace(); } }
private double calcAnaphoricity (JCas aJCas, Markable m) { svm_node[] nodes = createAnaphoricityVector(m, aJCas); double[] prob = new double[2]; svm.svm_predict_probability(anaph_model, nodes, prob); int[] labels = new int[2]; svm.svm_get_labels(anaph_model, labels); int anaph_idx = labels[0]==1 ? 0 : 1; return prob[anaph_idx]; }
private double calcAnaphoricity (JCas aJCas, Markable m) { svm_node[] nodes = createAnaphoricityVector(m, aJCas); double[] prob = new double[2]; svm.svm_predict_probability(anaph_model, nodes, prob); int[] labels = new int[2]; svm.svm_get_labels(anaph_model, labels); int anaph_idx = labels[0]==1 ? 0 : 1; return prob[anaph_idx]; }
public double classifyInstance(Observation observation, svm_model model) { List<Double> features = observation.getFeatures(); svm_node[] nodes = new svm_node[observation.getFeatures().size()]; for (int i = 0; i < features.size(); i++) { svm_node node = new svm_node(); node.index = i + 1; node.value = features.get(i); nodes[i] = node; } int[] labels = new int[TOTAL_CLASSES]; svm.svm_get_labels(model, labels); double[] prob_estimates = new double[TOTAL_CLASSES]; return svm.svm_predict_probability(model, nodes, prob_estimates); } }
svm.svm_get_labels(model,labels); prob_estimates = new double[nr_class];
@Override public Map<String, Double> predict(Tuple predict) { double[] feats = predict.vector.getVector(); svm_node[] svmfeats = new svm_node[feats.length]; for (int i = 0; i < feats.length; i++) { svm_node svmfeatI = new svm_node(); svmfeatI.index = i; svmfeatI.value = feats[i]; svmfeats[i] = svmfeatI; } int totalSize = labelIndexer.getLabelSize(); int[] labels = new int[totalSize]; svm.svm_get_labels(model, labels); double[] probs = new double[totalSize]; svm.svm_predict_probability(model, svmfeats, probs); Map<String, Double> result = new HashMap<>(); for (int i = 0; i < labels.length; i++) { result.put(labelIndexer.getLabel(labels[i]), probs[i]); } return result; }
svm.svm_get_labels(model,labels); prob_estimates = new double[nr_class]; output.writeBytes("labels");
svm.svm_get_labels(model, labels); prob_estimates = new double[nr_class]; output.writeBytes("labels");
svm.svm_get_labels(model, labels); prob_estimates = new double[nr_class]; output.writeBytes("labels");
svm.svm_get_labels(model, labels); int k = nr_class-1; if (kBestList.getK() != -1) {
svm.svm_get_labels(model, labels); int k = nr_class-1; if (kBestList.getK() != -1) {
int nr_class = svm.svm_get_nr_class(model); int[] labels = new int[nr_class]; svm.svm_get_labels(model, labels); boolean support_probabilities = svm.svm_check_probability_model(model) == 1; double[] scores = new double[nr_class];
svm.svm_get_labels(model, label); Matrix output = Matrix.Factory.zeros(1, MathUtil.max(label) + 1);
svm.svm_get_labels(m_Model, labels);