当前位置: 首页>>代码示例>>Java>>正文


Java Int2ObjectMap.get方法代码示例

本文整理汇总了Java中it.unimi.dsi.fastutil.ints.Int2ObjectMap.get方法的典型用法代码示例。如果您正苦于以下问题:Java Int2ObjectMap.get方法的具体用法?Java Int2ObjectMap.get怎么用?Java Int2ObjectMap.get使用的例子?那么, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在it.unimi.dsi.fastutil.ints.Int2ObjectMap的用法示例。


在下文中一共展示了Int2ObjectMap.get方法的12个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Java代码示例。

示例1: predict

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
private static double predict(final int user, final int itemI,
        @Nonnull final Int2ObjectMap<Int2FloatMap> knnItems, final int excludeIndex,
        @Nonnull final FloatMatrix weightMatrix) {
    final Int2FloatMap kNNu = knnItems.get(user);
    if (kNNu == null) {
        return 0.d;
    }

    double pred = 0.d;
    for (Int2FloatMap.Entry e : Fastutil.fastIterable(kNNu)) {
        final int itemK = e.getIntKey();
        if (itemK == excludeIndex) {
            continue;
        }
        float ruk = e.getFloatValue();
        pred += ruk * weightMatrix.get(itemI, itemK, 0.d);
    }
    return pred;
}
 
开发者ID:apache,项目名称:incubator-hivemall,代码行数:20,代码来源:SlimUDTF.java

示例2: initialize

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
private int initialize(Int2ObjectMap<CustomTimeBucket> idToTimeBucket)
{
  Preconditions.checkNotNull(idToTimeBucket);

  int tempId = Integer.MIN_VALUE;

  for (int timeBucketId : idToTimeBucket.keySet()) {
    tempId = Math.max(tempId, timeBucketId);
    CustomTimeBucket customTimeBucket = idToTimeBucket.get(timeBucketId);
    textToTimeBucket.put(customTimeBucket.getText(), customTimeBucket);
    Preconditions.checkNotNull(customTimeBucket);
    timeBucketToId.put(customTimeBucket, timeBucketId);
  }

  return tempId;
}
 
开发者ID:apache,项目名称:apex-malhar,代码行数:17,代码来源:CustomTimeBucketRegistry.java

示例3: bipartiteMatch

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
private Collection<Alignment> bipartiteMatch(Int2ObjectMap<Entity> schema1, Int2ObjectMap<Entity> schema2, Mat m) {
    MunkRes munkRes = new MunkRes(m);
    double[] sim = m.sim();
    final Collection<Alignment> alignmentSet = new ArrayList<>();
    for(IntPair ip : munkRes.execute()) {
        if(m.contains(ip._1, ip._2)) {
            final String e1id = schema1.get(ip._1).id;
            final String e2id = schema2.get(ip._2).id;
            final double s = sim[m.ij2index(ip._1, ip._2)];
            alignmentSet.add(new Alignment(e1id, e2id, Math.min(s,1.0)));
        }
    }
    System.err.printf("Produced %d matches from %d x %d\n", alignmentSet.size(), schema1.size(), schema2.size());
    return alignmentSet;
}
 
开发者ID:jmccrae,项目名称:naisc,代码行数:16,代码来源:WeightedBipartiteMatching.java

示例4: main

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
public static void main(String[] args) {
    //  create new map,
    Int2ObjectMap<String> map =
            new Int2ObjectOpenHashMap();

    // put/get method has primitive arg, no unboxing!
    map.put(1, "aa");
    map.get(1);

    //but it also implements full mapdb interface
    Map<Integer, String> map2 = map;

}
 
开发者ID:jankotek,项目名称:talk-save-java-memory,代码行数:14,代码来源:SpecializedMaps.java

示例5: initGraph

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
@Override
    protected void initGraph() {
        Int2ObjectMap<Node> nodes = new Int2ObjectOpenHashMap<>();

        for (int i = 0; i < 9; i++) {
            Node n = db.createNode();
            n.setProperty("id", i);
            nodes.put(i, n);
        }

        for (int i = 0; i < 9; i++) {
            Node src = nodes.get(i);
            Node dst = (i + 1) % 3 != 0 ? nodes.get(i + 1) : nodes.get(i - 2);

            src.createRelationshipTo(dst, CommonsRelationshipTypes.KNOWS);
//            dst.createRelationshipTo(src, CommonsRelationshipTypes.KNOWS);
        }

        nodes.get(0).createRelationshipTo(nodes.get(3), CommonsRelationshipTypes.KNOWS);
        nodes.get(3).createRelationshipTo(nodes.get(6), CommonsRelationshipTypes.KNOWS);
        nodes.get(6).createRelationshipTo(nodes.get(0), CommonsRelationshipTypes.KNOWS);

//        for (int i = 0; i < 9; i += 3) {
//            Node src = nodes.get(i);
//            Node dst1 = nodes.get((i + 3) % 9);
//            Node dst2 = nodes.get((i + 6) % 9);
//            src.createRelationshipTo(dst1, CommonsRelationshipTypes.KNOWS);
// //            dst1.createRelationshipTo(src, CommonsRelationshipTypes.KNOWS);
//            src.createRelationshipTo(dst2, CommonsRelationshipTypes.KNOWS);
// //            dst2.createRelationshipTo(src, CommonsRelationshipTypes.KNOWS);
//        }

    }
 
开发者ID:besil,项目名称:Neo4jSNA,代码行数:34,代码来源:LouvainTest.java

示例6: getGroupStats

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
@Override
public Iterator<GroupStats> getGroupStats(final EZImhotepSession session, final Map<Integer, GroupKey> groupKeys, final List<StatReference> statRefs, final long timeoutTS) throws ImhotepOutOfMemoryException {
    if(groupKeys.isEmpty()) {   // we don't have any parent groups probably because all docs were filtered out
        return Collections.<GroupStats>emptyList().iterator();
    }
    final StatReference countStatRef = session.pushStat(countStat);
    final long[] counts = getCounts(countStatRef);

    final Int2ObjectMap<Int2LongMap> groupToPositionToStats = getPercentileStats(session, groupKeys, countStatRef, counts);

    final List<GroupStats> result = Lists.newArrayList();

    final int statCount = statRefs.size();
    final int groupCount = session.getNumGroups();

    // get values for the normal stats
    final TIntObjectHashMap<double[]> statsResults = (statCount > 0) ? getGroupStatsValues(session, statRefs, groupCount) : null;

    // combine normal stats with distinct counts
    for (int groupNum = 1; groupNum < groupCount; groupNum++) {
        final Int2LongMap groupPercentileData = groupToPositionToStats.get(groupNum);
        double[] statsVals = statsResults != null ? statsResults.get(groupNum) : null;

        double[] values = new double[statCount + fields.size()];
        for(int i = 0, statsValsIndex = 0; i < values.length; i++) {
            if(groupPercentileData != null && groupPercentileData.containsKey(i)) {    // percentile value
                values[i] = groupPercentileData.get(i);
            } else if(statsVals != null && statsValsIndex < statsVals.length) {
                values[i] = statsVals[statsValsIndex++];    // normal stat value available
            } else {
                values[i] = 0;  // normal stat not in stats array
            }
        }

        GroupKey groupKey = groupKeys.get(groupNum);
        result.add(new GroupStats(groupKey, values));
    }

    return result.iterator();
}
 
开发者ID:indeedeng,项目名称:iql,代码行数:41,代码来源:PercentileGrouping.java

示例7: testMultiSegmentReverseIterationIncompleteLiveSegment

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
@Test
public void testMultiSegmentReverseIterationIncompleteLiveSegment() throws Exception {
  LeftIndexedPowerLawMultiSegmentBipartiteGraph leftIndexedPowerLawMultiSegmentBipartiteGraph =
    new LeftIndexedPowerLawMultiSegmentBipartiteGraph(
      3, 3, 5, 2, 2.0, 5, new IdentityEdgeTypeMask(), new NullStatsReceiver());

  addEdges(leftIndexedPowerLawMultiSegmentBipartiteGraph);

  /** One segment is dropped so the segments should have the following edges:
   * Segment 0: Dropped
   * Segment 1: (2, 21), (4, 42), (3, 31)
   * Segment 2: (2, 22), (1, 13), (4, 43)
   * Segment 3: (5, 11)
   */
  Int2ObjectMap<LeftIndexedBipartiteGraphSegment> segments =
    leftIndexedPowerLawMultiSegmentBipartiteGraph.getSegments();

  LeftIndexedBipartiteGraphSegment segment1 = segments.get(1);
  assertEquals(new LongArrayList(new long[]{21}),
    new LongArrayList(segment1.getLeftNodeEdges(2)));
  assertEquals(new LongArrayList(new long[]{31}),
    new LongArrayList(segment1.getLeftNodeEdges(3)));
  assertEquals(new LongArrayList(new long[]{42}),
    new LongArrayList(segment1.getLeftNodeEdges(4)));

  LeftIndexedBipartiteGraphSegment segment2 = segments.get(2);
  assertEquals(new LongArrayList(new long[]{13}),
    new LongArrayList(segment2.getLeftNodeEdges(1)));
  assertEquals(new LongArrayList(new long[]{22}),
    new LongArrayList(segment2.getLeftNodeEdges(2)));
  assertEquals(new LongArrayList(new long[]{43}),
    new LongArrayList(segment2.getLeftNodeEdges(4)));

  LeftIndexedBipartiteGraphSegment segment3 = segments.get(3);
  assertEquals(new LongArrayList(new long[]{11}),
    new LongArrayList(segment3.getLeftNodeEdges(5)));

  // Test that the iterator returns the correct edges after the first segment is dropped.
  assertEquals(new LongArrayList(new long[]{13}),
    new LongArrayList(leftIndexedPowerLawMultiSegmentBipartiteGraph.getLeftNodeEdges(1)));

  // Test that the iterator returns the correct edges and in reverse order when the edges are in different segments.
  assertEquals(new LongArrayList(new long[]{43, 42}),
    new LongArrayList(leftIndexedPowerLawMultiSegmentBipartiteGraph.getLeftNodeEdges(4)));
}
 
开发者ID:twitter,项目名称:GraphJet,代码行数:46,代码来源:MultiSegmentPowerLawBipartiteGraphTest.java

示例8: testMultiSegmentReverseIterationCompleteLiveSegment

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
@Test
public void testMultiSegmentReverseIterationCompleteLiveSegment() throws Exception {
  LeftIndexedPowerLawMultiSegmentBipartiteGraph leftIndexedPowerLawMultiSegmentBipartiteGraph =
    new LeftIndexedPowerLawMultiSegmentBipartiteGraph(
      4, 4, 2, 10, 2.0, 20, new IdentityEdgeTypeMask(), new NullStatsReceiver());

  for (long leftNode = 1; leftNode <= 2; leftNode++) {
    for (long i = 0; i < 10; i++) {
      leftIndexedPowerLawMultiSegmentBipartiteGraph.addEdge(leftNode, leftNode*10 + i, (byte) 0);
    }
  }

  /** One segment is dropped so the segments should have the following edges:
   * Segment 0: Dropped
   * Segment 1: (1, 14), (1, 15), (1, 16), (1, 17)
   * Segment 2: (1, 18), (1, 19), (2, 20), (2, 21)
   * Segment 3: (2, 22), (2, 23), (2, 24), (2, 25)
   * Segment 4: (2, 26), (2, 27), (2, 28), (2, 29)
   */
  Int2ObjectMap<LeftIndexedBipartiteGraphSegment> segments =
    leftIndexedPowerLawMultiSegmentBipartiteGraph.getSegments();

  LeftIndexedBipartiteGraphSegment segment1 = segments.get(1);
  assertEquals(new LongArrayList(new long[]{14, 15, 16, 17}),
    new LongArrayList(segment1.getLeftNodeEdges(1)));

  LeftIndexedBipartiteGraphSegment segment2 = segments.get(2);
  assertEquals(new LongArrayList(new long[]{18, 19}),
    new LongArrayList(segment2.getLeftNodeEdges(1)));
  assertEquals(new LongArrayList(new long[]{20, 21}),
    new LongArrayList(segment2.getLeftNodeEdges(2)));

  LeftIndexedBipartiteGraphSegment segment3 = segments.get(3);
  assertEquals(new LongArrayList(new long[]{22, 23, 24, 25}),
    new LongArrayList(segment3.getLeftNodeEdges(2)));

  LeftIndexedBipartiteGraphSegment segment4 = segments.get(4);
  assertEquals(new LongArrayList(new long[]{26, 27, 28, 29}),
    new LongArrayList(segment4.getLeftNodeEdges(2)));

  // Test that the iterator returns the correct edges and in reverse order when the edges are in different segments.
  assertEquals(new LongArrayList(new long[]{26, 27, 28, 29, 22, 23, 24, 25, 20, 21}),
    new LongArrayList(leftIndexedPowerLawMultiSegmentBipartiteGraph.getLeftNodeEdges(2)));
}
 
开发者ID:twitter,项目名称:GraphJet,代码行数:45,代码来源:MultiSegmentPowerLawBipartiteGraphTest.java

示例9: setup

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
@Override
public void setup(Context.OperatorContext context)
{
  super.setup(context);

  LOG.info("Done setting up super");
  aggregatorRegistry.setup();

  //Create prepared statements
  schema = new DimensionalConfigurationSchema(eventSchema, aggregatorRegistry);

  List<FieldsDescriptor> keyFDs = schema.getDimensionsDescriptorIDToKeyDescriptor();

  for (int ddID = 0; ddID < keyFDs.size(); ddID++) {

    LOG.info("ddID {}", ddID);
    FieldsDescriptor keyFD = keyFDs.get(ddID);
    Int2ObjectMap<FieldsDescriptor> aggIDToAggFD = schema
        .getDimensionsDescriptorIDToAggregatorIDToOutputAggregatorDescriptor().get(ddID);

    Map<Integer, PreparedStatement> aggIDToStatement = ddIDToAggIDToStatement.get(ddID);

    if (aggIDToStatement == null) {
      aggIDToStatement = Maps.newHashMap();
      ddIDToAggIDToStatement.put(ddID, aggIDToStatement);
    }

    for (Map.Entry<String, String> aggTable : tableNames.get(ddID).entrySet()) {
      int aggID = aggregatorRegistry.getIncrementalAggregatorNameToID().get(aggTable.getKey());

      LOG.info("aggID {}", aggID);
      FieldsDescriptor aggFD = aggIDToAggFD.get(aggID);

      List<String> keyNames = keyFD.getFieldList();
      keyNames.remove(DimensionsDescriptor.DIMENSION_TIME_BUCKET);

      LOG.info("List fields {}", keyNames);
      List<String> aggregateNames = aggFD.getFieldList();
      LOG.info("List fields {}", aggregateNames);
      String tableName = aggTable.getValue();

      String statementString = buildStatement(tableName, keyNames, aggregateNames);

      try {
        aggIDToStatement.put(aggID, store.getConnection().prepareStatement(statementString));
      } catch (SQLException ex) {
        throw new RuntimeException(ex);
      }
    }
  }
}
 
开发者ID:apache,项目名称:apex-malhar,代码行数:52,代码来源:JDBCDimensionalOutputOperator.java

示例10: setup

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
@Override
public void setup(Context.OperatorContext context)
{
  super.setup(context);

  LOG.info("Done setting up super");
  aggregatorRegistry.setup();

  //Create prepared statements
  schema = new DimensionalConfigurationSchema(eventSchema,
                                              aggregatorRegistry);

  List<FieldsDescriptor> keyFDs = schema.getDimensionsDescriptorIDToKeyDescriptor();

  for(int ddID = 0; ddID < keyFDs.size(); ddID ++) {

    LOG.info("ddID {}", ddID);
    FieldsDescriptor keyFD = keyFDs.get(ddID);
    Int2ObjectMap<FieldsDescriptor> aggIDToAggFD = schema.getDimensionsDescriptorIDToAggregatorIDToOutputAggregatorDescriptor().get(ddID);

    Map<Integer, PreparedStatement> aggIDToStatement = ddIDToAggIDToStatement.get(ddID);

    if(aggIDToStatement == null) {
      aggIDToStatement = Maps.newHashMap();
      ddIDToAggIDToStatement.put(ddID, aggIDToStatement);
    }

    for(Map.Entry<String, String> aggTable: tableNames.get(ddID).entrySet()) {
      int aggID = aggregatorRegistry.getIncrementalAggregatorNameToID().get(aggTable.getKey());

      LOG.info("aggID {}", aggID);
      FieldsDescriptor aggFD = aggIDToAggFD.get(aggID);

      List<String> keyNames = keyFD.getFieldList();
      keyNames.remove(DimensionsDescriptor.DIMENSION_TIME_BUCKET);
      List<String> aggregateNames = aggFD.getFieldList();
      String tableName = aggTable.getValue();

      String statementString = buildStatement(tableName, keyNames, aggregateNames);

      try {
        aggIDToStatement.put(aggID, store.getConnection().prepareStatement(statementString));
      } catch (SQLException ex) {
        throw new RuntimeException(ex);
      }
    }
  }
}
 
开发者ID:tweise,项目名称:apex-samples,代码行数:49,代码来源:JDBCDimensionalOutputOperator.java

示例11: removeDuplicatedNodes

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
public void removeDuplicatedNodes(Int2ObjectMap<List<Node>> borderPixelMap, int imageWidth) {
    ObjectIterator<Int2ObjectMap.Entry<List<Node>>> it = borderPixelMap.int2ObjectEntrySet().iterator();
    while (it.hasNext()) {
        Int2ObjectMap.Entry<List<Node>> entry = it.next();
        List<Node> nodes = entry.getValue();
        if (nodes.size() > 1) {
            Node refNode = nodes.get(0); // refNode
            // explore duplicated nodes
            for (int i=1; i<nodes.size(); i++) {
                Node currentNode = nodes.get(i);
                int edgeCount = currentNode.getEdgeCount();
                for (int k=0; k<edgeCount; k++) {
                    Edge edge = currentNode.getEdgeAt(k);
                    Node neighNit = edge.getTarget();
                    int removedEdgeIndex = neighNit.removeEdge(currentNode);
                    assert(removedEdgeIndex >= 0);

                    Edge edgeToFirstNode = neighNit.findEdge(refNode);
                    if (edgeToFirstNode == null) {
                        // create an edge neighNit -> refNode
                        neighNit.addEdge(refNode, edge.getBoundary());
                        // create an edge refNode -> neighNit
                        refNode.addEdge(neighNit, edge.getBoundary());
                    }
                }
                currentNode.setExpired(true);
            }

            IntSet borderCells = AbstractSegmenter.generateBorderCells(refNode.getContour(), refNode.getId(), imageWidth);
            IntIterator itCells = borderCells.iterator();
            while (itCells.hasNext()) {
                int gridId = itCells.nextInt();
                List<Node> resultNodes = borderPixelMap.get(gridId);
                if (resultNodes != null) {
                    resultNodes.clear();
                    resultNodes.add(refNode);
                }
            }
        }
    }
    removeExpiredNodes();
}
 
开发者ID:senbox-org,项目名称:s2tbx,代码行数:43,代码来源:Graph.java

示例12: constructStarTree

import it.unimi.dsi.fastutil.ints.Int2ObjectMap; //导入方法依赖的package包/类
private void constructStarTree(TreeNode node, int startDocId, int endDocId, int level) throws IOException {
  if (level == _dimensionsSplitOrder.size()) {
    return;
  }

  int splitDimensionId = _dimensionsSplitOrder.get(level);
  LOGGER.debug("Building tree at level: {} from startDoc: {} endDocId: {} splitting on dimension id: {}", level,
      startDocId, endDocId, splitDimensionId);

  int numDocs = endDocId - startDocId;
  Int2ObjectMap<IntPair> dimensionRangeMap;
  try (StarTreeDataTable dataTable = new StarTreeDataTable(
      new MMapBuffer(_dataFile, startDocId * _docSize, numDocs * _docSize, MMapMode.READ_ONLY), _dimensionSize,
      _metricSize, startDocId, endDocId)) {
    dimensionRangeMap = dataTable.groupOnDimension(startDocId, endDocId, splitDimensionId);
  }
  LOGGER.debug("Group stats:{}", dimensionRangeMap);

  node._childDimensionId = splitDimensionId;
  // Reserve one space for star node
  Map<Integer, TreeNode> children = new HashMap<>(dimensionRangeMap.size() + 1);
  node._children = children;
  for (Int2ObjectMap.Entry<IntPair> entry : dimensionRangeMap.int2ObjectEntrySet()) {
    int childDimensionValue = entry.getIntKey();
    TreeNode child = new TreeNode();
    _numNodes++;
    children.put(childDimensionValue, child);

    // The range pair value is the relative value to the start document id
    IntPair range = dimensionRangeMap.get(childDimensionValue);
    int childStartDocId = range.getLeft();
    child._startDocId = childStartDocId;
    int childEndDocId = range.getRight();
    child._endDocId = childEndDocId;

    if (childEndDocId - childStartDocId > _maxNumLeafRecords) {
      constructStarTree(child, childStartDocId, childEndDocId, level + 1);
    }
  }

  // Directly return if we don't need to create star-node
  if (_skipStarNodeCreationDimensions != null && _skipStarNodeCreationDimensions.contains(splitDimensionId)) {
    return;
  }

  // Create star node
  TreeNode starChild = new TreeNode();
  _numNodes++;
  children.put(StarTreeNode.ALL, starChild);
  starChild._dimensionId = splitDimensionId;
  starChild._dimensionValue = StarTreeNode.ALL;

  Iterator<Pair<DimensionBuffer, MetricBuffer>> iterator =
      getUniqueCombinations(startDocId, endDocId, splitDimensionId);
  int starChildStartDocId = _numRawDocs + _numAggregatedDocs;
  while (iterator.hasNext()) {
    Pair<DimensionBuffer, MetricBuffer> next = iterator.next();
    DimensionBuffer dimensions = next.getLeft();
    MetricBuffer metrics = next.getRight();
    appendToAggBuffer(dimensions, metrics);
  }
  _dataOutputStream.flush();
  int starChildEndDocId = _numRawDocs + _numAggregatedDocs;

  starChild._startDocId = starChildStartDocId;
  starChild._endDocId = starChildEndDocId;
  if (starChildEndDocId - starChildStartDocId > _maxNumLeafRecords) {
    constructStarTree(starChild, starChildStartDocId, starChildEndDocId, level + 1);
  }
}
 
开发者ID:linkedin,项目名称:pinot,代码行数:71,代码来源:OffHeapStarTreeBuilder.java


注:本文中的it.unimi.dsi.fastutil.ints.Int2ObjectMap.get方法示例由纯净天空整理自Github/MSDocs等开源代码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。