Index

A B C D E F G H I K L M N O P Q R S T U V W X Z
All Classes and Interfaces|All Packages

A

abs() - Method in class mklab.JGNN.core.Tensor

Computes the absolute value of tensor elements.

abs() - Method in class mklab.JGNN.core.tensor.DenseTensor

accessCol(long) - Method in class mklab.JGNN.core.Matrix

Retrieves the given column as a tensor.

accessCol(long) - Method in class mklab.JGNN.core.matrix.WrapCols

AccessCol - Class in mklab.JGNN.core.matrix

Accesses a column of a Matrix as if it were a dense Tensor.

AccessCol(Matrix, long) - Constructor for class mklab.JGNN.core.matrix.AccessCol

Instantiates a see-through access of a matrix column.

accessColumns() - Method in class mklab.JGNN.core.Matrix

Organizes specific matrix columns to a list of tensors that share entries.

accessColumns(long...) - Method in class mklab.JGNN.core.Matrix

Organizes specific matrix columns to a list of tensors that share entries.

accessColumns(Iterable<Long>) - Method in class mklab.JGNN.core.Matrix

Organizes some matrix columns to a list of tensors that share entries.

accessColumns(Tensor) - Method in class mklab.JGNN.core.Matrix

Organizes matrix columns to a list of tensors that share entries.

accessDim(long, String) - Method in class mklab.JGNN.core.Matrix

Retrieves either the given row or column as a trensor.

accessRow(long) - Method in class mklab.JGNN.core.Matrix

Retrieves the given row as a tensor.

accessRow(long) - Method in class mklab.JGNN.core.matrix.WrapRows

AccessRow - Class in mklab.JGNN.core.matrix

Accesses a row of a Matrix as if it were a dense Tensor.

AccessRow(Matrix, long) - Constructor for class mklab.JGNN.core.matrix.AccessRow

Instantiates a see-through access of a matrix row.

accessRows() - Method in class mklab.JGNN.core.Matrix

Organizes matrix rows to a list of tensors that share entries.

accessRows(long...) - Method in class mklab.JGNN.core.Matrix

Organizes specific matrix rows to a list of tensors that share entries.

accessRows(Iterable<Long>) - Method in class mklab.JGNN.core.Matrix

Organizes some matrix rows to a list of tensors that share entries.

accessRows(Tensor) - Method in class mklab.JGNN.core.Matrix

Organizes specific matrix rows to a list of tensors that share entries.

accessSubtensor(long) - Method in class mklab.JGNN.core.Tensor

Wraps a range of elements within a tensor without allocating memory anew.

accessSubtensor(long, long) - Method in class mklab.JGNN.core.Tensor

Wraps a range of elements within a tensor without allocating memory anew.

AccessSubtensor - Class in mklab.JGNN.core.tensor

Wraps a base Tensor by traversing only its elements in a specified range (from begin, up to end-1).

AccessSubtensor(Tensor, long) - Constructor for class mklab.JGNN.core.tensor.AccessSubtensor

Instantiates a see-through access of a tensor elements.

AccessSubtensor(Tensor, long, long) - Constructor for class mklab.JGNN.core.tensor.AccessSubtensor

Instantiates a see-through access of a tensor elements.

Accuracy - Class in mklab.JGNN.nn.loss

Implements an accuracy Loss of row-by-row comparisons.

Accuracy() - Constructor for class mklab.JGNN.nn.loss.Accuracy

Instantiates a row-by-row Accuracy loss.

Adam - Class in mklab.JGNN.nn.optimizers

Thic class implements an Adam Optimizer as explained in the paper: Kingma, Diederik P., and Jimmy Ba.

Adam() - Constructor for class mklab.JGNN.nn.optimizers.Adam

Initializes an NDAdam instance of an Adam optimizer with the default parameters recommended by the papers.

Adam(boolean, double) - Constructor for class mklab.JGNN.nn.optimizers.Adam

Initializes an Adam optimizer with the default parameters recommended in the literature, but allows for the specification of the learning rate and whether NDAdam or simple Adam is used.

Adam(boolean, double, double, double) - Constructor for class mklab.JGNN.nn.optimizers.Adam

Initializes an instance of an Adam optimizer with the default parameters while customizing the variation and learning rate.

Adam(boolean, double, double, double, double) - Constructor for class mklab.JGNN.nn.optimizers.Adam

Initializes an Adam optimizer by customizing all arguments.

Adam(double) - Constructor for class mklab.JGNN.nn.optimizers.Adam

Initializes an NDAdam instance of an Adam optimizer with the default parameters recommended by the papers but allows for the specification of the learning rate.

add(double) - Method in class mklab.JGNN.core.Tensor

add(double) - Method in class mklab.JGNN.core.tensor.DenseTensor

add(Tensor) - Method in class mklab.JGNN.core.Tensor

add(Tensor) - Method in class mklab.JGNN.core.tensor.DenseTensor

add(Tensor) - Method in class mklab.JGNN.core.tensor.VectorizedTensor

Add - Class in mklab.JGNN.nn.operations

Implements a NNOperation that adds its two inputs.

Add() - Constructor for class mklab.JGNN.nn.operations.Add

addGraph(Matrix, Matrix, Tensor) - Method in class mklab.JGNN.adhoc.train.AGFTraining

addInput(Variable) - Method in class mklab.JGNN.nn.Model

Adds to the model's inputs the provided Variable.

addInput(NNOperation) - Method in class mklab.JGNN.nn.inputs.Parameter

addInput(NNOperation) - Method in class mklab.JGNN.nn.NNOperation

addOutput(NNOperation) - Method in class mklab.JGNN.nn.Model

Adds to the model's output the output of the provided operation.

AGFTraining - Class in mklab.JGNN.adhoc.train

Extends the ModelTraining class to be able to train Model instances for attributed graph functions (AGFs).

AGFTraining() - Constructor for class mklab.JGNN.adhoc.train.AGFTraining

aggregate(LSTM) - Method in class mklab.JGNN.nn.operations.LSTM

allBaseMatrixTypes(long, long) - Method in class mklab.JGNN.core.MatrixTest

allMatrixTypes(long) - Method in class mklab.JGNN.core.MatrixTest

allocate(int, Object) - Static method in class mklab.JGNN.core.Memory

allocate(long) - Method in class mklab.JGNN.core.empty.EmptyMatrix

allocate(long) - Method in class mklab.JGNN.core.empty.EmptyTensor

allocate(long) - Method in class mklab.JGNN.core.matrix.AccessCol

allocate(long) - Method in class mklab.JGNN.core.matrix.AccessRow

allocate(long) - Method in class mklab.JGNN.core.matrix.ColumnRepetition

allocate(long) - Method in class mklab.JGNN.core.matrix.DenseMatrix

allocate(long) - Method in class mklab.JGNN.core.matrix.Diagonal

allocate(long) - Method in class mklab.JGNN.core.matrix.RepeatMatrix

allocate(long) - Method in class mklab.JGNN.core.matrix.RowRepetition

allocate(long) - Method in class mklab.JGNN.core.matrix.SparseMatrix

allocate(long) - Method in class mklab.JGNN.core.matrix.SparseSymmetric

Deprecated.

allocate(long) - Method in class mklab.JGNN.core.matrix.TransposedMatrix

allocate(long) - Method in class mklab.JGNN.core.matrix.VectorizedMatrix

allocate(long) - Method in class mklab.JGNN.core.matrix.WrapCols

allocate(long) - Method in class mklab.JGNN.core.matrix.WrapRows

allocate(long) - Method in class mklab.JGNN.core.tensor.AccessSubtensor

allocate(long) - Method in class mklab.JGNN.core.Tensor

allocate(long) - Method in class mklab.JGNN.core.tensor.DenseTensor

allocate(long) - Method in class mklab.JGNN.core.tensor.RepeatTensor

allocate(long) - Method in class mklab.JGNN.core.tensor.SparseTensor

allocate(long) - Method in class mklab.JGNN.core.tensor.VectorizedTensor

allSquareMatrixTypes(long) - Method in class mklab.JGNN.core.MatrixTest

allTensorTypes() - Method in class mklab.JGNN.core.TensorTest

allTensorTypes(long) - Method in class mklab.JGNN.core.TensorTest

allTensorTypesRandom(long) - Method in class mklab.JGNN.core.TensorTest

apply(Model) - Method in class mklab.JGNN.nn.Initializer

Applies the initializer to a given model's parameters.

apply(Model) - Method in class mklab.JGNN.nn.initializers.VariancePreservingInitializer

APPNP - Class in nodeClassification

Demonstrates classification with an APPNP GNN.

APPNP() - Constructor for class nodeClassification.APPNP

argmax() - Method in class mklab.JGNN.core.Tensor

Computes the position of the maximum tensor element.

argmin() - Method in class mklab.JGNN.core.Tensor

Computes the position of the minimum tensor element.

asColumn() - Method in class mklab.JGNN.core.Tensor

Accesses the tensor through a single-column matrix with the tensor as the only row.

asRow() - Method in class mklab.JGNN.core.Tensor

Accesses the tensor through a single-row matrix with the tensor as the only column.

assertBackwardValidity() - Method in class mklab.JGNN.adhoc.ModelBuilder

Asserts that all components parsed into a call graph with ModelBuilder.operation(String) are eventually used by at least one ModelBuilder.out(String) component.

assertExists(String) - Method in class mklab.JGNN.adhoc.ModelBuilder

assertFinite() - Method in class mklab.JGNN.core.Tensor

Asserts that the tensor holds only finite values.

assertMatching(Tensor) - Method in class mklab.JGNN.core.Tensor

Asserts that the tensor's dimensions match with another tensor.

assertSize(long) - Method in class mklab.JGNN.core.Tensor

Asserts that the tensor's Tensor.size() matches the given size.

assertValidName(String) - Method in class mklab.JGNN.adhoc.ModelBuilder

assign(Tensor) - Method in class mklab.JGNN.core.Tensor

Performs a sparse assignment.

asTensor() - Method in class mklab.JGNN.core.Slice

Creates a dense tensor holding the slice's identifiers.

asTransposed() - Method in class mklab.JGNN.core.Matrix

Creates a transposed version of the matrix that accesses the same elements (thus, editing one edits the other) without allocating additional memory.

asTransposed() - Method in class mklab.JGNN.core.matrix.TransposedMatrix

Attention - Class in mklab.JGNN.nn.operations

Implements a NNOperation that creates a version of adjacency matrices with column-wise attention involving neighbor similarity.

Attention() - Constructor for class mklab.JGNN.nn.operations.Attention

autosize(ArrayList<Tensor>) - Method in class mklab.JGNN.nn.NNOperation

autosize(ArrayList<Tensor>) - Method in class mklab.JGNN.nn.operations.Add

autosize(ArrayList<Tensor>) - Method in class mklab.JGNN.nn.operations.MatMul

autosize(List<Tensor>) - Method in class mklab.JGNN.adhoc.ModelBuilder

Applies the ModelBuilder.createForwardValidity(List) method for the given inputs to replace zero tensor dimensions (annotated with ? in symbolic definitions) with a valid dimension size and name, and then checks that all computation outcomes are valid with ModelBuilder.assertBackwardValidity().

autosize(Tensor...) - Method in class mklab.JGNN.adhoc.ModelBuilder

Applies the ModelBuilder.createForwardValidity(List) method for the given inputs to replace zero tensor dimensions (annotated with ? in symbolic definitions) with a valid dimension size and name, and then checks that all computation outcomes are valid with ModelBuilder.assertBackwardValidity().

A B C D E F G H I K L M N O P Q R S T U V W X Z
All Classes and Interfaces|All Packages