Package de.lmu.ifi.dbs.elki.data

Examples of de.lmu.ifi.dbs.elki.data.DoubleVector

128
129
130
131
132
133
134
135
136
137
138
139
140
141
    // also test partial queries, forward only
    testKNNQueries(rep, lin_knn_query, preproc_knn_query, k / 2);

    // insert new objects
    List<DoubleVector> insertions = new ArrayList<DoubleVector>();
    DoubleVector o = DatabaseUtil.assumeVectorField(rep).getFactory();
    Random random = new Random(seed);
    for(int i = 0; i < updatesize; i++) {
      DoubleVector obj = VectorUtil.randomVector(o, random);
      insertions.add(obj);
    }
    System.out.println("Insert " + insertions);
    System.out.println();
    DBIDs deletions = db.insert(MultipleObjectsBundle.makeSimple(rep.getDataTypeInformation(), insertions));
View Full Code Here
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
    // verify data set size.
    assertTrue(rep.size() == shoulds);

    {
      // get the 10 next neighbors
      DoubleVector dv = new DoubleVector(querypoint);
      KNNQuery<DoubleVector, DoubleDistance> knnq = db.getKNNQuery(dist, k);
      assertTrue("Returned knn query is not of expected class.", expectKNNQuery.isAssignableFrom(knnq.getClass()));
      KNNResult<DoubleDistance> ids = knnq.getKNNForObject(dv, k);
      assertEquals("Result size does not match expectation!", shouldd.length, ids.size());

      // verify that the neighbors match.
      int i = 0;
      for(DistanceResultPair<DoubleDistance> res : ids) {
        // Verify distance
        assertEquals("Expected distance doesn't match.", shouldd[i], res.getDistance().doubleValue());
        // verify vector
        DBID id = res.getDBID();
        DoubleVector c = rep.get(id);
        DoubleVector c2 = new DoubleVector(shouldc[i]);
        assertEquals("Expected vector doesn't match: " + c.toString(), 0.0, dist.distance(c, c2).doubleValue(), 0.00001);

        i++;
      }
    }
    {
      // Do a range query
      DoubleVector dv = new DoubleVector(querypoint);
      RangeQuery<DoubleVector, DoubleDistance> rangeq = db.getRangeQuery(dist, eps);
      assertTrue("Returned range query is not of expected class.", expectRangeQuery.isAssignableFrom(rangeq.getClass()));
      List<DistanceResultPair<DoubleDistance>> ids = rangeq.getRangeForObject(dv, eps);
      assertEquals("Result size does not match expectation!", shouldd.length, ids.size());

      // verify that the neighbors match.
      int i = 0;
      for(DistanceResultPair<DoubleDistance> res : ids) {
        // Verify distance
        assertEquals("Expected distance doesn't match.", shouldd[i], res.getDistance().doubleValue());
        // verify vector
        DBID id = res.getDBID();
        DoubleVector c = rep.get(id);
        DoubleVector c2 = new DoubleVector(shouldc[i]);
        assertEquals("Expected vector doesn't match: " + c.toString(), 0.0, dist.distance(c, c2).doubleValue(), 0.00001);

        i++;
      }
    }
View Full Code Here
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
    // verify data set size.
    assertTrue(rep.size() == shoulds);

    {
      // get the 10 next neighbors
      DoubleVector dv = new DoubleVector(querypoint);
      KNNQuery<DoubleVector, DoubleDistance> knnq = db.getKNNQuery(dist, k);
      assertTrue("Returned knn query is not of expected class.", expectKNNQuery.isAssignableFrom(knnq.getClass()));
      List<DistanceResultPair<DoubleDistance>> ids = knnq.getKNNForObject(dv, k);
      assertEquals("Result size does not match expectation!", shouldd.length, ids.size());

      // verify that the neighbors match.
      int i = 0;
      for(DistanceResultPair<DoubleDistance> res : ids) {
        // Verify distance
        assertEquals("Expected distance doesn't match.", shouldd[i], res.getDistance().doubleValue());
        // verify vector
        DBID id = res.getDBID();
        DoubleVector c = rep.get(id);
        DoubleVector c2 = new DoubleVector(shouldc[i]);
        assertEquals("Expected vector doesn't match: " + c.toString(), 0.0, dist.distance(c, c2).doubleValue(), 0.00001);

        i++;
      }
    }
    {
      // Do a range query
      DoubleVector dv = new DoubleVector(querypoint);
      RangeQuery<DoubleVector, DoubleDistance> rangeq = db.getRangeQuery(dist, eps);
      assertTrue("Returned range query is not of expected class.", expectRangeQuery.isAssignableFrom(rangeq.getClass()));
      List<DistanceResultPair<DoubleDistance>> ids = rangeq.getRangeForObject(dv, eps);
      assertEquals("Result size does not match expectation!", shouldd.length, ids.size());

      // verify that the neighbors match.
      int i = 0;
      for(DistanceResultPair<DoubleDistance> res : ids) {
        // Verify distance
        assertEquals("Expected distance doesn't match.", shouldd[i], res.getDistance().doubleValue());
        // verify vector
        DBID id = res.getDBID();
        DoubleVector c = rep.get(id);
        DoubleVector c2 = new DoubleVector(shouldc[i]);
        assertEquals("Expected vector doesn't match: " + c.toString(), 0.0, dist.distance(c, c2).doubleValue(), 0.00001);

        i++;
      }
    }
View Full Code Here
324
325
326
327
328
329
330
331
332
333
334
            throw new AbortException("Expected word token, got: " + tokenizer.toString());
          }
          cur[k] = tokenizer.nval;
          nextToken(tokenizer);
        }
        data[out] = new DoubleVector(cur);
      }
      else if(etyp[out] == TypeUtil.LABELLIST) {
        // Build a label list out of successive labels
        LabelList ll = new LabelList(dimsize[out]);
        for(int k = 0; k < dimsize[out]; k++) {
View Full Code Here
409
410
411
412
413
414
415
416
417
418
419
        // Collect labels:
        for(int i = 0; i < dimsize[out]; i++) {
          labels[i] = names.get(out + i);
        }
        if(!sparse) {
          VectorFieldTypeInformation<DoubleVector> type = new VectorFieldTypeInformation<DoubleVector>(DoubleVector.class, DoubleVector.STATIC, dimsize[out], labels, new DoubleVector(new double[dimsize[out]]));
          bundle.appendColumn(type, new ArrayList<DoubleVector>());
        }
        else {
          VectorFieldTypeInformation<SparseFloatVector> type = new VectorFieldTypeInformation<SparseFloatVector>(SparseFloatVector.class, dimsize[out], labels, new SparseFloatVector(SparseFloatVector.EMPTYMAP, dimsize[out]));
          bundle.appendColumn(type, new ArrayList<SparseFloatVector>());
View Full Code Here
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
    derivator = parameters.tryInstantiate(cls);

    CorrelationAnalysisSolution<DoubleVector> model = derivator.run(derivatorDB);

    Matrix weightMatrix = model.getSimilarityMatrix();
    DoubleVector centroid = new DoubleVector(model.getCentroid());
    DistanceQuery<DoubleVector, DoubleDistance> df = QueryUtil.getDistanceQuery(derivatorDB, new WeightedDistanceFunction(weightMatrix));
    DoubleDistance eps = df.getDistanceFactory().parseString("0.25");

    ids.addDBIDs(interval.getIDs());
    // Search for nearby vectors in original database
    for(DBID id : relation.iterDBIDs()) {
      DoubleVector v = new DoubleVector(relation.get(id).getColumnVector().getArrayRef());
      DoubleDistance d = df.distance(v, centroid);
      if(d.compareTo(eps) < 0) {
        ids.add(id);
      }
    }
View Full Code Here
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
   */
  private Database buildDerivatorDB(Relation<ParameterizationFunction> relation, CASHInterval interval) throws UnableToComplyException {
    DBIDs ids = interval.getIDs();
    ProxyDatabase proxy = new ProxyDatabase(ids);
    int dim = relation.get(ids.iterator().next()).getDimensionality();
    SimpleTypeInformation<DoubleVector> type = new VectorFieldTypeInformation<DoubleVector>(DoubleVector.class, dim, new DoubleVector(new double[dim]));
    MaterializedRelation<DoubleVector> prep = new MaterializedRelation<DoubleVector>(proxy, type, ids);
    proxy.addRelation(prep);

    // Project
    for(DBID id : ids) {
      DoubleVector v = new DoubleVector(relation.get(id).getColumnVector().getArrayRef());
      prep.set(id, v);
    }

    if(logger.isDebugging()) {
      logger.debugFine("db fuer derivator : " + prep.size());
View Full Code Here
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
   *         possible
   */
  private Database buildDerivatorDB(Relation<ParameterizationFunction> relation, DBIDs ids) throws UnableToComplyException {
    ProxyDatabase proxy = new ProxyDatabase(ids);
    int dim = relation.get(ids.iterator().next()).getDimensionality();
    SimpleTypeInformation<DoubleVector> type = new VectorFieldTypeInformation<DoubleVector>(DoubleVector.class, dim, new DoubleVector(new double[dim]));
    MaterializedRelation<DoubleVector> prep = new MaterializedRelation<DoubleVector>(proxy, type, ids);
    proxy.addRelation(prep);

    // Project
    for(DBID id : ids) {
      DoubleVector v = new DoubleVector(relation.get(id).getColumnVector().getArrayRef());
      prep.set(id, v);
    }

    return proxy;
  }
View Full Code Here
TOP

Related Classes of de.lmu.ifi.dbs.elki.data.DoubleVector

  • de.lmu.ifi.dbs.elki.algorithm.clustering.correlation.CASH
  • de.lmu.ifi.dbs.elki.algorithm.outlier.TestOnlineLOF
  • de.lmu.ifi.dbs.elki.algorithm.statistics.DistanceStatisticsWithClasses
  • de.lmu.ifi.dbs.elki.algorithm.statistics.EvaluateRankingQuality
  • de.lmu.ifi.dbs.elki.algorithm.statistics.RankingQualityHistogram
  • de.lmu.ifi.dbs.elki.data.synthetic.bymodel.GeneratorMain
  • de.lmu.ifi.dbs.elki.datasource.filter.DoubleVectorProjectionFilter
  • de.lmu.ifi.dbs.elki.datasource.filter.DoubleVectorRandomProjectionFilter
  • de.lmu.ifi.dbs.elki.datasource.parser.ArffParser
  • de.lmu.ifi.dbs.elki.datasource.parser.DoubleVectorLabelParser
    • 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.