A B C D E F G I L M N O R S T

A

actdev(double) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Applies first derivative of layer's activation function to activation on unit (not net input).

actfun(double) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Applies layer's activation function to net input.

activate(BPLayer, double[]) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Activates this layer by clamping activations on another and running a forward pass.

add(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Adds a scalar to this vector.

add(SNMatrix) - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Adds a matrix to this matrix.

add(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Adds a vector to this vector.

add(SNVector, double) - Static method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the sum of a vector and a scalar.

add(SNVector, SNVector) - Static method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the sum of two vectors.

and(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the logical AND of this vector with another.

arrayMultiply(SNMatrix, SNMatrix) - Static method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns a matrix containing the products of the elements of two matrices.

arrayMultiply(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Multiplies the elements of this vector by those of another.

arrayMultiply(SNVector, SNVector) - Static method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a vector containing the products of the elements of two vectors.

attach(double[][]) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Attaches pattern to layer as input or target.

B

batch(double, double) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Runs one step of back-propagation in batch mode.

batch(int, double, double) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Same as batch(nep, eta, mu, report), but with error reporting every second.

batch(int, double, double, int) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Trains all layer's in this layer's network, using back-propagation in batch mode.

BPLayer - class edu.wlu.cs.levy.SNARLI.BPLayer.

BPLayer is a class supporting creation and training of layered neural networks through back-propagation.

BPLayer(int) - Constructor for class edu.wlu.cs.levy.SNARLI.BPLayer

Creates a neural network layer.

bpttPattern() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Steps through one pattern for Back-Prop Thru Time.

bpttPattern(int) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Steps through one pattern for Back-Prop Thru Time.

bpttResetEta() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Resets current weight- and bias-changes for Back-Prop-Through-Time.

bpttUpdate(double, double, int) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Updates weights and bias on layer using the Delta Rule, for Back-Prop-Through-Time.

C

connect(BPLayer) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Makes a normal (full, Sigma) connection to this layer from another layer.

connect(BPLayer, BPLayer) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Makes a Sigma-Pi connection to this layer from two other layers.

copy() - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a deep copy of this vector.

copy() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns a deep copy of this matrix.

D

delay(BPLayer) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Same as delay(from, weight), with weight = 1.0.

delay(BPLayer, double) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Makes one-to-one time-delay connection to this layer from the specified layer, using the specified connection strength.

divide(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Divides this vector by a scalar.

divide(double) - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Divides this matrix by a scalar.

divide(SNMatrix, double) - Static method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the quotient of a matrix and a scalar.

divide(SNVector, double) - Static method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the quotient of a vector and a scalar.

dontCare() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Returns out-of-bounds value for don't-care condition.

E

edu.wlu.cs.levy.SNARLI - package edu.wlu.cs.levy.SNARLI

eucDist(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the Euclidean distance between this vector and another.

F

find(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns indices where vector equals scalar argument.

G

gaussianNoise(int, int, Random) - Static method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns a matrix of normally distributed random values.

gaussianNoise(int, Random) - Static method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a vector of normally distributed random values.

ge(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a vector containing ones where this vector's elements are greater than or equal to a scalar, and zeros elsewhere.

getBias() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Returns the bias on this layer.

getCols() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the number of columns of this Matrix.

getColumnDimension() - Method in class edu.wlu.cs.levy.SNARLI.Map2D

Returns the number of columns in this map.

getMaxError() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Returns error of maximum magnitude on layer over all training patterns.

getRMSError() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Returns Root-Mean-Squared error on layer over all training patterns.

getRow(int) - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Gets the row vector at a certain index of this matrix.

getRowDimension() - Method in class edu.wlu.cs.levy.SNARLI.Map2D

Returns the number of rows in this map.

getRows() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the number of rows of this Matrix.

getSquaredErrors() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Returns squared errors on layer over all training patterns.

getUnit(double[]) - Method in class edu.wlu.cs.levy.SNARLI.Map2D

Returns indices of grid point whose reference vector is closest to a given vector.

getUnits() - Method in class edu.wlu.cs.levy.SNARLI.Map2D

Returns indices of grid points from reference vectors.

getValue(int) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns an indexed value from this vector.

getValues() - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a 1D array of the values contained in this vector.

getValues() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns a 2D array of the values contained in this matrix.

getWeights() - Method in class edu.wlu.cs.levy.SNARLI.Map2D

Returns current reference vector weights.

getWeights(BPLayer) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Returns the weights on this layer from another layer (Sigma connection), as a 2D array.

getWeights(BPLayer, BPLayer) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Returns the weights on this layer from two other layers (Sigma-Pi connection), as an array of 2D arrays.

grow(int, double, double, int) - Method in class edu.wlu.cs.levy.SNARLI.IGG

Runs the growth phase.

I

IGG - class edu.wlu.cs.levy.SNARLI.IGG.

IGG implements Fritzke's Incremental Growing Grid network.

IGG(double[][]) - Constructor for class edu.wlu.cs.levy.SNARLI.IGG

Builds IGG with random initial weights and uniform probability density function.

IGG(double[][], double[]) - Constructor for class edu.wlu.cs.levy.SNARLI.IGG

Builds IGG with random initial weights.

IGG(double[][], double[], long) - Constructor for class edu.wlu.cs.levy.SNARLI.IGG

Builds IGG with random initial weights.

IGG(double[][], double[], Random) - Constructor for class edu.wlu.cs.levy.SNARLI.IGG

Builds IGG with random initial weights.

IGG(double[][], long) - Constructor for class edu.wlu.cs.levy.SNARLI.IGG

Builds IGG with random initial weights and uniform probability density function.

IGG(double[][], Random) - Constructor for class edu.wlu.cs.levy.SNARLI.IGG

Builds IGG with random initial weights and uniform probability density function.

L

le(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a vector containing ones where this vector's elements are less than or equal to a scalar, and zeros elsewhere.

learn(double, double, double, double, int) - Method in class edu.wlu.cs.levy.SNARLI.SOM

Runs learning iterations.

length() - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the length of this vector.

M

Map2D - class edu.wlu.cs.levy.SNARLI.Map2D.

Map2D is a general class for two-dimensional maps (SOM, IGG, ...).

max() - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the largest value of any element in this vector.

max() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the value of the largest element of this matrix.

meanRows() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns a vector repsenting the mean over the rows of this Matrix.

min() - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the value of the smallest element of this vector.

multiply(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Multiplies this vector by a scalar.

multiply(SNMatrix) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the product of this Vector and a Matrix.

multiply(SNMatrix, double) - Static method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the product of a matrix and a scalar.

multiply(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Multiplies the elements of this vector by those of another vector.

multiply(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the product of this matrix and a vector.

multiply(SNVector, double) - Static method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the product of a vector and a scalar.

N

ne(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a vector containing ones where this vector's elements are not equal to a scalar, and zeros elsewhere.

ne(double) - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns a matrix containing ones where this matrix's elements are not equal to a scalar, and zeros elsewhere.

nycDist(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the Manhattan distance (L1 norm of differences) between this vector and another.

O

online(int, double, double) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Same as online(nep, eta, mu, report), but with error report every second.

online(int, double, double, int) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Trains all layers in this layer's network, using back-propagation in on-line mode.

outer(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the outer product of this vector and another.

R

randomize() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Same as randomize(seed), with arbitrary seed.

randomize(long) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Randomizes biases and weights on this layer.

randomize(Random) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Randomizes biases and weights on this layer to values normally distributed around zero.

repmat(int) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Replicates this vector as a matrix.

reportValue(int, int, int, double, PrintStream) - Static method in class edu.wlu.cs.levy.SNARLI.BPLayer

Reports a value in a friendly way.

resetMu() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Resets previous weight- and bias- changes.

S

setAll(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Sets all entries to a scalar.

setBias(double[]) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Sets bias on this layer to the values in vector.

setRow(int, SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Sets the row vector at a certain index of this matrix.

setValue(int, double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Sets an entry to a scalar at a certain index.

setWeights(BPLayer, BPLayer, double[][][]) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Sets weights on this layer from two others (Sigma-Pi connection) to values in array of 2D arrays.

setWeights(BPLayer, double[][]) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Sets weights on this layer from another (Sigma connection) to values in 2D array.

SNMatrix - class edu.wlu.cs.levy.SNARLI.SNMatrix.

This class provides glue code for matrix operations in SNARLI.

SNMatrix(double[][]) - Constructor for class edu.wlu.cs.levy.SNARLI.SNMatrix

Constructs a matrix using a 2D array of double-precision floating-point values.

SNMatrix(int, int) - Constructor for class edu.wlu.cs.levy.SNARLI.SNMatrix

Constructs a matrix of all zeros.

SNVector - class edu.wlu.cs.levy.SNARLI.SNVector.

This class provides glue code for vector operations in SNARLI.

SNVector(double[]) - Constructor for class edu.wlu.cs.levy.SNARLI.SNVector

Constructs a vector using a 1D array of double-precision floating-point values.

SNVector(int) - Constructor for class edu.wlu.cs.levy.SNARLI.SNVector

Constructs a vector of all zeros.

SOM - class edu.wlu.cs.levy.SNARLI.SOM.

SOM implements Kohonen's Self-Organizing Map.

SOM(double[][], int, int) - Constructor for class edu.wlu.cs.levy.SNARLI.SOM

Builds SOM with random initial weights and uniform probability density function.

SOM(double[][], int, int, double[]) - Constructor for class edu.wlu.cs.levy.SNARLI.SOM

Builds SOM with random initial weights.

SOM(double[][], int, int, double[], long) - Constructor for class edu.wlu.cs.levy.SNARLI.SOM

Builds SOM with random initial weights.

SOM(double[][], int, int, double[], Random) - Constructor for class edu.wlu.cs.levy.SNARLI.SOM

Builds SOM with random initial weights.

SOM(double[][], int, int, long) - Constructor for class edu.wlu.cs.levy.SNARLI.SOM

Builds SOM with random initial weights and uniform probability density function.

SOM(double[][], int, int, Random) - Constructor for class edu.wlu.cs.levy.SNARLI.SOM

Builds SOM with random initial weights and uniform probability density function.

sort() - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a copy of this vector, sorted in ascending order.

subtract(double) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Subtracts a scalar from this vector.

subtract(SNMatrix, SNMatrix) - Static method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the difference between two matrices.

subtract(SNVector) - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Subtracts a vector from this vector.

subtract(SNVector, double) - Static method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the difference between a vector and a scalar.

subtract(SNVector, SNVector) - Static method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the difference between two vectors.

sum() - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns the sum of the elements of this vector.

sum() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the sum over all elements of this Matrix.

sumRows() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns a vector representing the sum over the rows of this Matrix.

T

test() - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Tests a (trained) network layer.

test(int, double) - Method in class edu.wlu.cs.levy.SNARLI.BPLayer

Tests a (recurrent) network layer without any input.

toString() - Method in class edu.wlu.cs.levy.SNARLI.SNVector

Returns a String representation of this vector.

toString() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns a String representation of this matrix.

transpose() - Method in class edu.wlu.cs.levy.SNARLI.SNMatrix

Returns the transpose of this matrix.

tune(double, double, double, int) - Method in class edu.wlu.cs.levy.SNARLI.IGG

Runs the fine-tuning phase.

A B C D E F G I L M N O R S T