Modifier and Type | Method and Description |
---|---|
<T extends Vector> |
SVM$.predictVectors()
Provides the operation that makes the predictions for individual examples.
|
static <T extends Vector> |
SVM.predictVectors()
Provides the operation that makes the predictions for individual examples.
|
Modifier and Type | Method and Description |
---|---|
Vector |
LabeledVector.vector() |
Vector |
WeightVector.weights() |
Constructor and Description |
---|
LabeledVector(double label,
Vector vector) |
WeightVector(Vector weights,
double intercept) |
Modifier and Type | Interface and Description |
---|---|
interface |
BreezeVectorConverter<T extends Vector>
Type class which allows the conversion from Breeze vectors to Flink vectors
|
interface |
VectorBuilder<T extends Vector>
Type class to allow the vector construction from different data types
|
Modifier and Type | Class and Description |
---|---|
class |
DenseVector
Dense vector implementation of
Vector . |
class |
SparseVector
Sparse vector implementation storing the data in two arrays.
|
Modifier and Type | Method and Description |
---|---|
<T extends Vector> |
Breeze.Breeze2VectorConverter.fromBreeze(BreezeVectorConverter<T> evidence$1) |
Modifier and Type | Method and Description |
---|---|
Vector |
Vector.copy()
Copies the vector instance
|
Modifier and Type | Method and Description |
---|---|
static void |
BLAS.axpy(double a,
Vector x,
Vector y)
y += a * x
|
void |
BLAS$.axpy(double a,
Vector x,
Vector y)
y += a * x
|
static void |
BLAS.copy(Vector x,
Vector y)
y = x
|
void |
BLAS$.copy(Vector x,
Vector y)
y = x
|
double |
Vector.dot(Vector other)
Returns the dot product of the recipient and the argument
|
double |
SparseVector.dot(Vector other)
Returns the dot product of the recipient and the argument
|
double |
DenseVector.dot(Vector other)
Returns the dot product of the recipient and the argument
|
static double |
BLAS.dot(Vector x,
Vector y)
dot(x, y)
|
double |
BLAS$.dot(Vector x,
Vector y)
dot(x, y)
|
boolean |
Vector.equalsVector(Vector vector) |
static boolean |
SparseVector.equalsVector(Vector vector) |
static boolean |
DenseVector.equalsVector(Vector vector) |
Matrix |
Vector.outer(Vector other)
Returns the outer product of the recipient and the argument
|
SparseMatrix |
SparseVector.outer(Vector other)
Returns the outer product (a.k.a.
|
Matrix |
DenseVector.outer(Vector other)
Returns the outer product (a.k.a.
|
static void |
BLAS.scal(double a,
Vector x)
x = a * x
|
void |
BLAS$.scal(double a,
Vector x)
x = a * x
|
static void |
BLAS.syr(double alpha,
Vector x,
DenseMatrix A)
A := alpha * x * x^T^ + A
|
void |
BLAS$.syr(double alpha,
Vector x,
DenseMatrix A)
A := alpha * x * x^T^ + A
|
Constructor and Description |
---|
Vector2BreezeConverter(Vector vector) |
Modifier and Type | Method and Description |
---|---|
Vector |
IndexedRow.values() |
Modifier and Type | Method and Description |
---|---|
DistributedRowMatrix |
DistributedRowMatrix.byRowOperation(scala.Function2<Vector,Vector,Vector> func,
DistributedRowMatrix other)
Applies a high-order function to couple of rows.
|
DistributedRowMatrix |
DistributedRowMatrix.byRowOperation(scala.Function2<Vector,Vector,Vector> func,
DistributedRowMatrix other)
Applies a high-order function to couple of rows.
|
DistributedRowMatrix |
DistributedRowMatrix.byRowOperation(scala.Function2<Vector,Vector,Vector> func,
DistributedRowMatrix other)
Applies a high-order function to couple of rows.
|
Constructor and Description |
---|
IndexedRow(int rowIndex,
Vector values) |
Modifier and Type | Method and Description |
---|---|
protected static void |
TanimotoDistanceMetric.checkValidArguments(Vector a,
Vector b) |
void |
DistanceMetric.checkValidArguments(Vector a,
Vector b) |
protected static void |
CosineDistanceMetric.checkValidArguments(Vector a,
Vector b) |
protected static void |
ChebyshevDistanceMetric.checkValidArguments(Vector a,
Vector b) |
protected static void |
EuclideanDistanceMetric.checkValidArguments(Vector a,
Vector b) |
protected static void |
MinkowskiDistanceMetric.checkValidArguments(Vector a,
Vector b) |
protected static void |
SquaredEuclideanDistanceMetric.checkValidArguments(Vector a,
Vector b) |
protected static void |
ManhattanDistanceMetric.checkValidArguments(Vector a,
Vector b) |
double |
TanimotoDistanceMetric.distance(Vector a,
Vector b) |
double |
DistanceMetric.distance(Vector a,
Vector b)
Returns the distance between the arguments.
|
double |
CosineDistanceMetric.distance(Vector a,
Vector b) |
double |
ChebyshevDistanceMetric.distance(Vector a,
Vector b) |
double |
EuclideanDistanceMetric.distance(Vector a,
Vector b) |
double |
MinkowskiDistanceMetric.distance(Vector a,
Vector b) |
double |
SquaredEuclideanDistanceMetric.distance(Vector a,
Vector b) |
double |
ManhattanDistanceMetric.distance(Vector a,
Vector b) |
Modifier and Type | Method and Description |
---|---|
static <T extends Vector> |
KNN.fitKNN(TypeInformation<T> evidence$1)
FitOperation which trains a KNN based on the given training data set. |
<T extends Vector> |
KNN$.fitKNN(TypeInformation<T> evidence$1)
FitOperation which trains a KNN based on the given training data set. |
static <T extends Vector> |
KNN.predictValues(scala.reflect.ClassTag<T> evidence$2,
TypeInformation<T> evidence$3)
PredictDataSetOperation which calculates k-nearest neighbors of the given testing data
set. |
<T extends Vector> |
KNN$.predictValues(scala.reflect.ClassTag<T> evidence$2,
TypeInformation<T> evidence$3)
PredictDataSetOperation which calculates k-nearest neighbors of the given testing data
set. |
Modifier and Type | Method and Description |
---|---|
scala.Tuple2<Vector,Vector> |
QuadTree.Node.getCenterWidth()
for testing purposes only; used in QuadTreeSuite.scala
|
scala.Tuple2<Vector,Vector> |
QuadTree.Node.getCenterWidth()
for testing purposes only; used in QuadTreeSuite.scala
|
scala.collection.mutable.ListBuffer<Vector> |
QuadTree.Node.nodeElements() |
scala.collection.Seq<Vector> |
QuadTree.Node.partitionBox(Vector center,
Vector width)
Recursive function that partitions a n-dim box by taking the (n-1) dimensional
plane through the center of the box keeping the n-th coordinate fixed,
then shifting it in the n-th direction up and down
and recursively applying partitionBox to the two shifted (n-1) dimensional planes.
|
scala.collection.mutable.ListBuffer<Vector> |
QuadTree.searchNeighbors(Vector queryPoint,
double radius)
Finds all objects within a neighborhood of queryPoint of a specified radius
scope is modified from original 2D version in:
http://www.cs.trinity.edu/~mlewis/CSCI1321-F11/Code/src/util/Quadtree.scala
|
scala.collection.mutable.ListBuffer<Vector> |
QuadTree.searchNeighborsSiblingQueue(Vector queryPoint)
Used to zoom in on a region near a test point for a fast KNN query.
|
scala.Option<DataSet<Block<Vector>>> |
KNN.trainingSet() |
Modifier and Type | Method and Description |
---|---|
boolean |
QuadTree.Node.contains(Vector queryPoint)
Tests whether the queryPoint is in the node, or a child of that node
|
void |
QuadTree.insert(Vector queryPoint)
Recursively adds an object to the tree
|
boolean |
QuadTree.Node.isNear(Vector queryPoint,
double radius)
Tests if queryPoint is near a node
|
double |
QuadTree.Node.minDist(Vector queryPoint)
minDist is defined so that every point in the box has distance to queryPoint greater
than minDist (minDist adopted from "Nearest Neighbors Queries" by N.
|
boolean |
QuadTree.Node.overlap(Vector queryPoint,
double radius)
Tests if queryPoint is within a radius of the node
|
scala.collection.Seq<Vector> |
QuadTree.Node.partitionBox(Vector center,
Vector width)
Recursive function that partitions a n-dim box by taking the (n-1) dimensional
plane through the center of the box keeping the n-th coordinate fixed,
then shifting it in the n-th direction up and down
and recursively applying partitionBox to the two shifted (n-1) dimensional planes.
|
scala.collection.mutable.ListBuffer<Vector> |
QuadTree.searchNeighbors(Vector queryPoint,
double radius)
Finds all objects within a neighborhood of queryPoint of a specified radius
scope is modified from original 2D version in:
http://www.cs.trinity.edu/~mlewis/CSCI1321-F11/Code/src/util/Quadtree.scala
|
scala.collection.mutable.ListBuffer<Vector> |
QuadTree.searchNeighborsSiblingQueue(Vector queryPoint)
Used to zoom in on a region near a test point for a fast KNN query.
|
int |
QuadTree.Node.whichChild(Vector queryPoint)
Finds which child queryPoint lies in.
|
Constructor and Description |
---|
Node(Vector center,
Vector width,
scala.collection.Seq<QuadTree.Node> children) |
QuadTree(Vector minVec,
Vector maxVec,
DistanceMetric distMetric,
int maxPerBox) |
Modifier and Type | Method and Description |
---|---|
Vector |
NoRegularization$.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient
|
Vector |
RegularizationPenalty.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and regularization penalty
|
Vector |
L1Regularization$.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and L1 regularization penalty
|
static Vector |
L1Regularization.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and L1 regularization penalty
|
static Vector |
L2Regularization.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and L2 regularization penalty
|
Vector |
L2Regularization$.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and L2 regularization penalty
|
static Vector |
NoRegularization.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient
|
Vector |
GradientDescent.takeStep(Vector weightVector,
Vector gradient,
RegularizationPenalty regularizationPenalty,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient
|
Modifier and Type | Method and Description |
---|---|
static WeightVector |
LinearPrediction.gradient(Vector features,
WeightVector weights) |
abstract WeightVector |
PredictionFunction.gradient(Vector features,
WeightVector weights) |
WeightVector |
LinearPrediction$.gradient(Vector features,
WeightVector weights) |
static double |
LinearPrediction.predict(Vector features,
WeightVector weightVector) |
abstract double |
PredictionFunction.predict(Vector features,
WeightVector weights) |
double |
LinearPrediction$.predict(Vector features,
WeightVector weightVector) |
double |
NoRegularization$.regLoss(double oldLoss,
Vector weightVector,
double regularizationParameter)
Returns the unmodified loss value
|
double |
RegularizationPenalty.regLoss(double oldLoss,
Vector weightVector,
double regularizationConstant)
Adds regularization to the loss value
|
double |
L1Regularization$.regLoss(double oldLoss,
Vector weightVector,
double regularizationConstant)
Adds regularization to the loss value
|
static double |
L1Regularization.regLoss(double oldLoss,
Vector weightVector,
double regularizationConstant)
Adds regularization to the loss value
|
static double |
L2Regularization.regLoss(double oldLoss,
Vector weightVector,
double regularizationConstant)
Adds regularization to the loss value
|
double |
L2Regularization$.regLoss(double oldLoss,
Vector weightVector,
double regularizationConstant)
Adds regularization to the loss value
|
static double |
NoRegularization.regLoss(double oldLoss,
Vector weightVector,
double regularizationParameter)
Returns the unmodified loss value
|
Vector |
NoRegularization$.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient
|
Vector |
RegularizationPenalty.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and regularization penalty
|
Vector |
L1Regularization$.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and L1 regularization penalty
|
static Vector |
L1Regularization.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and L1 regularization penalty
|
static Vector |
L2Regularization.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and L2 regularization penalty
|
Vector |
L2Regularization$.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient and L2 regularization penalty
|
static Vector |
NoRegularization.takeStep(Vector weightVector,
Vector gradient,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient
|
Vector |
GradientDescent.takeStep(Vector weightVector,
Vector gradient,
RegularizationPenalty regularizationPenalty,
double regularizationConstant,
double learningRate)
Calculates the new weights based on the gradient
|
Modifier and Type | Method and Description |
---|---|
static <T extends Vector> |
StochasticOutlierSelection.transformVectors(BreezeVectorConverter<T> evidence$1,
TypeInformation<T> evidence$2,
scala.reflect.ClassTag<T> evidence$3)
TransformDataSetOperation applies the stochastic outlier selection algorithm on a
Vector which will transform the high-dimensionaly input to a single Double output. |
<T extends Vector> |
StochasticOutlierSelection$.transformVectors(BreezeVectorConverter<T> evidence$1,
TypeInformation<T> evidence$2,
scala.reflect.ClassTag<T> evidence$3)
TransformDataSetOperation applies the stochastic outlier selection algorithm on a
Vector which will transform the high-dimensionaly input to a single Double output. |
Modifier and Type | Method and Description |
---|---|
<T extends Vector> |
StandardScaler$.fitLabelVectorTupleStandardScaler(BreezeVectorConverter<T> evidence$1,
TypeInformation<T> evidence$2,
scala.reflect.ClassTag<T> evidence$3)
Trains the
StandardScaler by learning the mean and standard deviation of the training
data which is of type (Vector , Double). |
static <T extends Vector> |
StandardScaler.fitLabelVectorTupleStandardScaler(BreezeVectorConverter<T> evidence$1,
TypeInformation<T> evidence$2,
scala.reflect.ClassTag<T> evidence$3)
Trains the
StandardScaler by learning the mean and standard deviation of the training
data which is of type (Vector , Double). |
<T extends Vector> |
MinMaxScaler$.fitVectorMinMaxScaler()
Trains the
MinMaxScaler by learning the minimum and maximum of each feature of the
training data. |
static <T extends Vector> |
MinMaxScaler.fitVectorMinMaxScaler()
Trains the
MinMaxScaler by learning the minimum and maximum of each feature of the
training data. |
<T extends Vector> |
StandardScaler$.fitVectorStandardScaler()
Trains the
StandardScaler by learning the mean and
standard deviation of the training data. |
static <T extends Vector> |
StandardScaler.fitVectorStandardScaler()
Trains the
StandardScaler by learning the mean and
standard deviation of the training data. |
<V extends Vector> |
StandardScaler.StandardScalerTransformOperation.scale(V vector,
scala.Tuple2<breeze.linalg.Vector<Object>,breeze.linalg.Vector<Object>> model,
BreezeVectorConverter<V> evidence$6) |
<T extends Vector> |
StandardScaler$.transformTupleVectorDouble(BreezeVectorConverter<T> evidence$10,
TypeInformation<T> evidence$11,
scala.reflect.ClassTag<T> evidence$12)
|
static <T extends Vector> |
StandardScaler.transformTupleVectorDouble(BreezeVectorConverter<T> evidence$10,
TypeInformation<T> evidence$11,
scala.reflect.ClassTag<T> evidence$12)
|
static <T extends Vector> |
PolynomialFeatures.transformVectorIntoPolynomialBase(VectorBuilder<T> evidence$1,
TypeInformation<T> evidence$2,
scala.reflect.ClassTag<T> evidence$3)
TransformDataSetOperation to map a Vector into the
polynomial feature space. |
<T extends Vector> |
PolynomialFeatures$.transformVectorIntoPolynomialBase(VectorBuilder<T> evidence$1,
TypeInformation<T> evidence$2,
scala.reflect.ClassTag<T> evidence$3)
TransformDataSetOperation to map a Vector into the
polynomial feature space. |
<T extends Vector> |
StandardScaler$.transformVectors(BreezeVectorConverter<T> evidence$7,
TypeInformation<T> evidence$8,
scala.reflect.ClassTag<T> evidence$9)
TransformOperation to transform Vector types |
<T extends Vector> |
MinMaxScaler$.transformVectors(BreezeVectorConverter<T> evidence$1,
TypeInformation<T> evidence$2,
scala.reflect.ClassTag<T> evidence$3)
TransformDataSetOperation which scales input data of subtype of Vector with respect to
the calculated minimum and maximum of the training data. |
static <T extends Vector> |
MinMaxScaler.transformVectors(BreezeVectorConverter<T> evidence$1,
TypeInformation<T> evidence$2,
scala.reflect.ClassTag<T> evidence$3)
TransformDataSetOperation which scales input data of subtype of Vector with respect to
the calculated minimum and maximum of the training data. |
static <T extends Vector> |
StandardScaler.transformVectors(BreezeVectorConverter<T> evidence$7,
TypeInformation<T> evidence$8,
scala.reflect.ClassTag<T> evidence$9)
TransformOperation to transform Vector types |
Modifier and Type | Method and Description |
---|---|
static <T extends Vector> |
MultipleLinearRegression.predictVectors() |
<T extends Vector> |
MultipleLinearRegression$.predictVectors() |
Copyright © 2014–2018 The Apache Software Foundation. All rights reserved.