Class Tensor3D

java.lang.Object

deepnetts.tensor.TensorBase

deepnetts.tensor.Tensor3D

All Implemented Interfaces:

Tensor, Serializable

public class Tensor3D
extends TensorBase

A 3D tensor/matrix, with rows, columns and depth.

See Also:

• Serialized Form

Field Summary

Fields inherited from class TensorBase

numDimensions, shape, values, vectorizationImpl

Constructor Summary

Constructors

Constructor	Description
Tensor3D(float[][][] vals)	Creates a 3D tensor from specified 3D array
Tensor3D(int depth, int rows, int cols)	Creates a new 3D tensor with specified number of rows, columns and depth.
Tensor3D(int depth, int rows, int cols, float[] values)

Method Summary

Modifier and Type	Method	Description
final void	add(float value, int z, int row, int col)
float	channelMean(int ch)
final int	cols()
TensorBase	copy()
final void	copy2DSubChannel(int ch, float[] buff)	Copies 2D channel values from this tensor into given float array buffer.
final int	depth()
void	divChannel(float val, int ch)
static Tensor3D	fromFile(String filePath)
final float	get(int ch, int row, int col)
final int	rows()
final void	set(float val, int ch, int row, int col)
void	setRow(float[] rowData, int depthIdx, int rowIdx)
final void	sub(float val, int z, int row, int col)
void	toFile(String filePath)
String	toString()

Methods inherited from class TensorBase

add, add, addInto, apply, apply, clone, containsNaN, copyFrom, copyFrom, copyFromGPU, copyToGPU, createAcceleratorBridge, div, div, div, equals, equals, fill, getAcceleratorBridge, getOrCreateAccBridge, getValues, hashCode, mean, multiply, multiply, multiplyElementWise, numDimensions, numElements, randomize, replace, setShape, setValues, setValuesFromString, setValuesFromStringTransposed, shape, sqr, sqrt, std, sub, sub, sub, subInplace, sum, sumAbs, sumSqr, toTsr, valuesAsString

Methods inherited from class Object

finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Details

Tensor3D

public Tensor3D(int depth,
 int rows,
 int cols)

Creates a new 3D tensor with specified number of rows, columns and depth.

Parameters:

depth - tensor depth

rows - number of rows

cols - number of columns

Tensor3D

public Tensor3D(float[][][] vals)

Creates a 3D tensor from specified 3D array

Parameters:

vals - 2D array of tensor values

Tensor3D

public Tensor3D(int depth,
 int rows,
 int cols,
 float[] values)

Method Details

get

public final float get(int ch,
 int row,
 int col)

set

public final void set(float val,
 int ch,
 int row,
 int col)

add

public final void add(float value,
 int z,
 int row,
 int col)

sub

public final void sub(float val,
 int z,
 int row,
 int col)

copy

public TensorBase copy()

Overrides:

copy in class TensorBase

cols

public final int cols()

rows

public final int rows()

depth

public final int depth()

copy2DSubChannel

public final void copy2DSubChannel(int ch,
 float[] buff)

Copies 2D channel values from this tensor into given float array buffer. Used on 3D tensors.

Parameters:

ch - tensor channel to copy

buff - buffer to copy values into

setRow

public void setRow(float[] rowData,
 int depthIdx,
 int rowIdx)

channelMean

public float channelMean(int ch)

divChannel

public void divChannel(float val,
 int ch)

fromFile

public static Tensor3D fromFile(String filePath)
                         throws IOException

Throws:

IOException

toString

public String toString()

Overrides:

toString in class TensorBase

toFile

public void toFile(String filePath)
            throws IOException

Throws:

IOException