Package de.lmu.ifi.dbs.elki.distance.distancevalue

Examples of de.lmu.ifi.dbs.elki.distance.distancevalue.DoubleDistance

125
126
127
128
129
130
131
    @Override
    public DoubleDistance distance(DBID id1, DBID id2) {
      DBIDs neighbors1 = index.getNearestNeighborSet(id1);
      DBIDs neighbors2 = index.getNearestNeighborSet(id2);
      return new DoubleDistance(1.0 - jaccardCoefficient(neighbors1, neighbors2));
    }
View Full Code Here
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
    }

    // compute distances between each point in S and m_i
    Map<DBID, DistanceResultPair<DoubleDistance>> distances = new HashMap<DBID, DistanceResultPair<DoubleDistance>>();
    for(DBID id : s) {
      DoubleDistance dist = distFunc.distance(id, m_i);
      distances.put(id, new GenericDistanceResultPair<DoubleDistance>(dist, id));
    }

    for(int i = 1; i < m; i++) {
      // choose medoid m_i to be far from prevois medoids
      List<DistanceResultPair<DoubleDistance>> d = new ArrayList<DistanceResultPair<DoubleDistance>>(distances.values());
      Collections.sort(d);

      m_i = d.get(d.size() - 1).getDBID();
      medoids.add(m_i);
      s.remove(m_i);
      distances.remove(m_i);

      // compute distances of each point to closest medoid
      for(DBID id : s) {
        DoubleDistance dist_new = distFunc.distance(id, m_i);
        DoubleDistance dist_old = distances.get(id).getDistance();

        DoubleDistance dist = dist_new.compareTo(dist_old) < 0 ? dist_new : dist_old;
        distances.put(id, new GenericDistanceResultPair<DoubleDistance>(dist, id));
      }

      if(logger.isDebugging()) {
        logger.debugFiner("medoids " + medoids);
View Full Code Here
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
    Map<DBID, List<DistanceResultPair<DoubleDistance>>> result = new HashMap<DBID, List<DistanceResultPair<DoubleDistance>>>();

    for(DBID m : medoids) {
      // determine minimum distance between current medoid m and any other
      // medoid m_i
      DoubleDistance minDist = null;
      for(DBID m_i : medoids) {
        if(m_i == m) {
          continue;
        }
        DoubleDistance currentDist = distFunc.distance(m, m_i);
        if(minDist == null || currentDist.compareTo(minDist) < 0) {
          minDist = currentDist;
        }
      }

      // determine points in sphere centered at m with radius minDist
View Full Code Here
616
617
618
619
620
621
622
623
624
625
626
627
      Pair<DoubleDistance, Integer> minDist = null;
      for(int i = 0; i < dimensions.size(); i++) {
        Pair<V, Set<Integer>> pair_i = dimensions.get(i);
        V c_i = pair_i.first;
        Set<Integer> dimensions_i = pair_i.second;
        DoubleDistance currentDist = manhattanSegmentalDistance(p, c_i, dimensions_i);
        if(minDist == null || currentDist.compareTo(minDist.first) < 0) {
          minDist = new Pair<DoubleDistance, Integer>(currentDist, i);
        }
      }
      // add p to cluster with mindist
      assert minDist != null;
View Full Code Here
662
663
664
665
666
667
668
    double result = 0;
    for(Integer d : dimensions) {
      result += Math.abs(o1.doubleValue(d) - o2.doubleValue(d));
    }
    result /= dimensions.size();
    return new DoubleDistance(result);
  }
View Full Code Here
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
    Iterator<DBID> it = database.iterDBIDs();
    while(it.hasNext()) {
      DBID id = it.next();
      V o = database.get(id);

      DoubleDistance minDist = null;
      ORCLUSCluster minCluster = null;

      // determine projected distance between o and cluster
      for(int i = 0; i < clusters.size(); i++) {
        ORCLUSCluster c = clusters.get(i);
        V o_proj = projection(c, o, factory);
        DoubleDistance dist = distFunc.distance(o_proj, projectedCentroids.get(i));
        if(minDist == null || minDist.compareTo(dist) > 0) {
          minDist = dist;
          minCluster = c;
        }
      }
View Full Code Here
285
286
287
288
289
290
291
292
293
294
295
    // covariance matrix of cluster
    // Matrix covariance = Util.covarianceMatrix(database, cluster.objectIDs);
    List<DistanceResultPair<DoubleDistance>> results = new ArrayList<DistanceResultPair<DoubleDistance>>(cluster.objectIDs.size());
    for(Iterator<DBID> it = cluster.objectIDs.iterator(); it.hasNext();) {
      DBID id = it.next();
      DoubleDistance distance = distFunc.distance(cluster.centroid, database.get(id));
      DistanceResultPair<DoubleDistance> qr = new GenericDistanceResultPair<DoubleDistance>(distance, id);
      results.add(qr);
    }
    Collections.sort(results);
    PCAResult pcares = pca.processQueryResult(results, database);
View Full Code Here
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
  private ProjectedEnergy projectedEnergy(Relation<V> database, DistanceQuery<V, DoubleDistance> distFunc, ORCLUSCluster c_i, ORCLUSCluster c_j, int i, int j, int dim) {
    // union of cluster c_i and c_j
    ORCLUSCluster c_ij = union(database, distFunc, c_i, c_j, dim);
    V factory = DatabaseUtil.assumeVectorField(database).getFactory();

    DoubleDistance sum = getDistanceFunction().getDistanceFactory().nullDistance();
    V c_proj = projection(c_ij, c_ij.centroid, factory);
    for(DBID id : c_ij.objectIDs) {
      V o = database.get(id);
      V o_proj = projection(c_ij, o, factory);
      DoubleDistance dist = distFunc.distance(o_proj, c_proj);
      sum = sum.plus(dist.times(dist));
    }
    DoubleDistance projectedEnergy = sum.times(1.0 / c_ij.objectIDs.size());

    return new ProjectedEnergy(i, j, c_ij, projectedEnergy);
  }
View Full Code Here
55
56
57
58
59
60
61
    this.id = id;
  }

  @Override
  public DoubleDistance getDistance() {
    return new DoubleDistance(distance);
  }
View Full Code Here
147
148
149
150
151
152
153
154
155
156
157
          // We need to do some ugly casts to be able to run the optimized version, unfortunately.
          @SuppressWarnings("unchecked")
          final ClusterOrderResult<DoubleDistance> doubleClusterOrder = ClusterOrderResult.class.cast(clusterOrder);
          @SuppressWarnings("unchecked")
          final RangeQuery<O, DoubleDistance> doubleRangeQuery = RangeQuery.class.cast(rangeQuery);
          final DoubleDistance depsilon = DoubleDistance.class.cast(epsilon);
          expandClusterOrderDouble(doubleClusterOrder, database, doubleRangeQuery, id, depsilon, progress);
        }
      }
    }
    else {
View Full Code Here
TOP

Related Classes of de.lmu.ifi.dbs.elki.distance.distancevalue.DoubleDistance

  • de.lmu.ifi.dbs.elki.algorithm.clustering.correlation.CASH
  • de.lmu.ifi.dbs.elki.algorithm.clustering.correlation.ORCLUS
  • de.lmu.ifi.dbs.elki.algorithm.clustering.OPTICS
  • de.lmu.ifi.dbs.elki.algorithm.clustering.subspace.PROCLUS
  • de.lmu.ifi.dbs.elki.algorithm.outlier.SOD
  • de.lmu.ifi.dbs.elki.database.query.DoubleDistanceResultPair
  • de.lmu.ifi.dbs.elki.distance.distancefunction.AbstractVectorDoubleDistanceFunction
  • de.lmu.ifi.dbs.elki.distance.distancefunction.AbstractVectorDoubleDistanceNorm
  • de.lmu.ifi.dbs.elki.distance.distancefunction.adapter.AbstractSimilarityAdapter$Instance
  • de.lmu.ifi.dbs.elki.distance.distancefunction.colorhistogram.HistogramIntersectionDistanceFunction
    • 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.