org.nd4j.linalg.convolution

Class Convolution

java.lang.Object

org.nd4j.linalg.convolution.Convolution

public class Convolution
extends Object

Convolution is the code for applying the convolution operator.

Author:

Adam Gibson

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	Convolution.Type

Method Summary

Modifier and Type	Method and Description
static INDArray	col2im(INDArray col, int[] stride, int[] padding, int height, int width)
static INDArray	col2im(INDArray col, int sy, int sx, int ph, int pw, int h, int w) Rearrange matrix columns into blocks
static INDArray	conv2d(IComplexNDArray input, IComplexNDArray kernel, Convolution.Type type)
static INDArray	conv2d(INDArray input, INDArray kernel, Convolution.Type type) 2d convolution (aka the last 2 dimensions
static IComplexNDArray	convn(IComplexNDArray input, IComplexNDArray kernel, Convolution.Type type) ND Convolution
static IComplexNDArray	convn(IComplexNDArray input, IComplexNDArray kernel, Convolution.Type type, int[] axes) ND Convolution
static INDArray	convn(INDArray input, INDArray kernel, Convolution.Type type) ND Convolution
static INDArray	convn(INDArray input, INDArray kernel, Convolution.Type type, int[] axes) ND Convolution
static INDArray	im2col(INDArray img, int[] kernel, int[] stride, int[] padding)
static INDArray	im2col(INDArray img, int kh, int kw, int sy, int sx, int ph, int pw, boolean coverAll) Implement column formatted images
static INDArray	im2col(INDArray img, int kh, int kw, int sy, int sx, int ph, int pw, int pval, boolean coverAll) Implement column formatted images
static int	outSize(int size, int k, int s, int p, boolean coverAll) The out size for a convolution

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

col2im

public static INDArray col2im(INDArray col,
                              int[] stride,
                              int[] padding,
                              int height,
                              int width)

Parameters:

col -

stride -

padding -

height -

width -

Returns:

col2im

public static INDArray col2im(INDArray col,
                              int sy,
                              int sx,
                              int ph,
                              int pw,
                              int h,
                              int w)

Rearrange matrix columns into blocks

Parameters:

col - the column transposed image to convert

sy - stride y

sx - stride x

ph - padding height

pw - padding width

h - height

w - width

Returns:

im2col

public static INDArray im2col(INDArray img,
                              int[] kernel,
                              int[] stride,
                              int[] padding)

Parameters:

img -

kernel -

stride -

padding -

Returns:

im2col

public static INDArray im2col(INDArray img,
                              int kh,
                              int kw,
                              int sy,
                              int sx,
                              int ph,
                              int pw,
                              boolean coverAll)

Implement column formatted images

Parameters:

img - the image to process

kh - the kernel height

kw - the kernel width

sy - the stride along y

sx - the stride along x

ph - the padding width

pw - the padding height

coverAll - whether to cover the whole image or not

Returns:

the column formatted image

im2col

public static INDArray im2col(INDArray img,
                              int kh,
                              int kw,
                              int sy,
                              int sx,
                              int ph,
                              int pw,
                              int pval,
                              boolean coverAll)

Implement column formatted images

Parameters:

img - the image to process

kh - the kernel height

kw - the kernel width

sy - the stride along y

sx - the stride along x

ph - the padding width

pw - the padding height

pval - the padding value (not used)

coverAll - whether to cover the whole image or not

Returns:

the column formatted image

outSize

public static int outSize(int size,
                          int k,
                          int s,
                          int p,
                          boolean coverAll)

The out size for a convolution

Parameters:

size -

k -

s -

p -

coverAll -

Returns:

conv2d

public static INDArray conv2d(INDArray input,
                              INDArray kernel,
                              Convolution.Type type)

2d convolution (aka the last 2 dimensions

Parameters:

input - the input to op

kernel - the kernel to convolve with

type -

Returns:

conv2d

public static INDArray conv2d(IComplexNDArray input,
                              IComplexNDArray kernel,
                              Convolution.Type type)

Parameters:

input -

kernel -

type -

Returns:

convn

public static INDArray convn(INDArray input,
                             INDArray kernel,
                             Convolution.Type type,
                             int[] axes)

ND Convolution

Parameters:

input - the input to op

kernel - the kerrnel to op with

type - the type of convolution

axes - the axes to do the convolution along

Returns:

the convolution of the given input and kernel

convn

public static IComplexNDArray convn(IComplexNDArray input,
                                    IComplexNDArray kernel,
                                    Convolution.Type type,
                                    int[] axes)

ND Convolution

Parameters:

input - the input to op

kernel - the kernel to op with

type - the type of convolution

axes - the axes to do the convolution along

Returns:

the convolution of the given input and kernel

convn

public static INDArray convn(INDArray input,
                             INDArray kernel,
                             Convolution.Type type)

ND Convolution

Parameters:

input - the input to op

kernel - the kernel to op with

type - the type of convolution

Returns:

the convolution of the given input and kernel

convn

public static IComplexNDArray convn(IComplexNDArray input,
                                    IComplexNDArray kernel,
                                    Convolution.Type type)

ND Convolution

Parameters:

input - the input to op

kernel - the kernel to op with

type - the type of convolution

Returns:

the convolution of the given input and kernel