Map<K,V> mapping
boolean readOnly
true when the BasisMapping should not create new
dimensions for unseen keys.WorkQueue workQueue
LinkClustering instances to perform
multi-threaded operations.List<E> mergeOrder
List<E> edgeList
int numRows
PermutationFunction<T extends Vector> permFunc
PermutationFunction to use for permuting vectors.PermutationFunction<T extends Vector> permFunc
PermutationFunction to use for permuting vectors.PermutationFunction<T extends Vector> permFunc
PermutationFunction to use for permuting vectors.int numEdges
gnu.trove.map.TIntObjectMap<V> vertexToEdges
Edge instances
that connect it to other members of the graph.Collection<E> subgraphs
WeakReference in order to keep track
of the canonical Subgraph instance while ensuring that it is
garbage collected when it is no longer referred to (which would never
happen if this list contained strong references).gnu.trove.map.TObjectIntMap<K> typeCounts
gnu.trove.map.TIntObjectMap<V> vertexToEdges
Collection<E> subgraphs
WeakReference in order to keep track
of the canonical Subgraph instance while ensuring that it is
garbage collected when it is no longer referred to (which would never
happen if this list contained strong references).int size
Graph<E extends Edge> g
int subgraphSize
Iterator<E> vertexIter
g.Queue<E> nextSubgraphs
double[] traversalProbabilitiesAtDepth
i corresponds to the
probability of expanding the subgraph at depth i. This array is
the same size as the the number of desired vertices in the subgraph.int from
int to
Object type
int from
int to
Object type
int from
int to
Object type
double weight
int from
int to
Object type
double weight
private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException
IOExceptionClassNotFoundExceptionprivate void writeObject(ObjectOutputStream out) throws IOException
IOExceptionint rootVertex
gnu.trove.map.hash.TIntObjectHashMap<V> inEdges
gnu.trove.map.hash.TIntObjectHashMap<V> outEdges
gnu.trove.set.hash.TIntHashSet connected
inEdges and outEdges, representing it as its own data
structure offers significantly faster iteration over the set returned by
SparseDirectedTypedEdgeSet.connected().int size
connected, which doesn't include bidirectional
connections.private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException
IOExceptionClassNotFoundExceptionprivate void writeObject(ObjectOutputStream out) throws IOException
IOExceptionint rootVertex
gnu.trove.map.hash.TIntObjectHashMap<V> edges
int size
BitSet setTypes
Graph<E extends Edge> g
int subgraphSize
IntIterator vertexIter
g, which are each used to a
seed to generate sequences of subgraphsedu.ucla.sspace.graph.SubgraphIterator.Extension ext
Graph<E extends Edge> next
null if no further subgraphs
remaingnu.trove.map.TObjectIntMap<K> typeCounts
gnu.trove.map.TIntObjectMap<V> vertexToEdges
Collection<E> subgraphs
WeakReference in order to keep track
of the canonical Subgraph instance while ensuring that it is
garbage collected when it is no longer referred to (which would never
happen if this list contained strong references).int size
gnu.trove.map.TObjectIntMap<K> typeCounts
gnu.trove.map.TIntObjectMap<V> vertexToEdges
Collection<E> subgraphs
WeakReference in order to keep track
of the canonical Subgraph instance while ensuring that it is
garbage collected when it is no longer referred to (which would never
happen if this list contained strong references).int size
IsomorphismTester isoTest
Map<K,V> orderAndSizeToGraphs
int sum
boolean allowNewMotifs
List<E> graphs
IsomorphismTester tester
IsomorphismTester isoTest
Map<K,V> typesToGraphs
int sum
boolean allowNewMotifs
IsomorphismTester isoTest
Map<K,V> graphIndices
Map<K,V> wordToSemantics
int windowSize
Set<E> semanticFilter
BasisMapping<T,E> basisMapping
double stdev
int indexVectorLength
double mean
Vector.Random randomGenerator
int numVectorValues
TernaryVector.int variance
TernaryVector.int indexVectorLength
double mean
double std
DoubleVectorint vectorLength
List<E> generatedVectors
SparseMatrix backing
SparseDoubleVector lemmaVector
Map<K,V> selPreferences
Map<K,V> inverseSelPreferences
VectorCombinor combinor
StringBasisMapping termBasis
Map<K,V> preferenceVectors
VectorCombinor combinor
VectorCombinor responsible for merging features between two
SparseDoubleVectors. This is used when computing the relational
preference vectors for each word and for computing the contextualized
vectors.Object[] array
int size
Map or -1 if the number of
mappings has yet to be calculated. The latter condition occurs when an
array is initially passed in and hasn't been checked for null
elements, which are considered non-existant mappings.int bound
boolean isKeyBound
true if the bound should apply to the number of keys, or false if the number of valueboolean isFair
true if the map should pick a random
mapping to remove, which ensures fairness, rather than an arbitrary
mapping.char[] keyIndices
Object[] values
private void readObject(ObjectInputStream in) throws IOException, ClassNotFoundException
IOExceptionClassNotFoundExceptionprivate void writeObject(ObjectOutputStream out) throws IOException
IOExceptionIndexer<T> objectIndices
gnu.trove.map.TIntIntMap indexToCount
boolean allowNewIndices
objectIndices Indexer.int sum
Iterator<E> iter
Object cur
null if no elements have
been returned yet or if the current element was removed via IteratorDecorator.remove().int[] buckets
boolean isEmptyMarkerValuePresent
EMPTY_MARKER has been stored in this
set.boolean isDeletedMarkerValuePresent
DELETED_MARKER has been stored in
this set.int size
gnu.trove.map.TLongIntMap counts
Indexer<T> elementIndices
Pair instance to an index,
and instead maps the concatenation of the pair's elements' IDs to a
count, which minizes the space required.int sum
private void readObject(ObjectInputStream ois) throws ClassNotFoundException, IOException
ClassNotFoundExceptionIOExceptionprivate void writeObject(ObjectOutputStream out) throws IOException
SemanticSpace is not serializable.IOExceptionint maxLength
Map<K,V> indexToValue
int[] indices
SparseHashArray.getElementIndices() and is set back to null if any
mutating operation occur subsequently.SortedMap<K,V> map
int range
boolean recalculateRange
TreeMultiMap.range() call can return the correct result. An invalid range is the
result of a submap being created.TreeMultiMap<K,V> parent
null if
this is the original map.BitSet bitSet
List<E> values
edu.ucla.sspace.vector.AmortizedSparseVector.CellComparator comp
IndexValue.int maxLength
int knownLength
SparseDoubleVector vector
SparseDoubleVector that this AtomicSparseVector
decorates.Lock readLock
vector.Lock writeLock
vector.DoubleVector vector
Vector that this AtomicVector decorates.Lock readLock
vector.Lock writeLock
vector.SparseIntArray intArray
double magnitude
SparseDoubleArray vector
SparseDoubleArray which provides most of the functionality in
this class.double magnitude
DoubleVector doubleVector
DoubleVector reference to the backing vectorint[] columnMask
boolean updated
SparseVector<T extends Number> sparseVector
SparseVector reference to the DoubleVector data backing
this view.Map<K,V> reverseColumnMask
int[] nonZeroIndices
MaskedSparseDoubleVectorView.getNonZeroIndices() is called and some value
has been modified in the base vector.SparseHashArray<T> vector
double magnitude
Copyright © 2012. All Rights Reserved.