Class Tensor

Multidimensional array-like data storage

Inheritance

object

Tensor

Implements

IDisposable

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

Namespace: Unity.Barracuda

Assembly: Unity.Barracuda.dll

Syntax

public class Tensor : IDisposable

Constructors

Tensor(int, int, int, int, float[,,,], string)

Create a Tensor of shape [1,1,N,1,1,H,W,C], an array of data srcData and an optional debug name. srcData must be of size n*h*w*c.

Declaration

public Tensor(int n, int h, int w, int c, float[,,,] srcData, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels
float[,,,]	srcData	source data
string	name	name

Tensor(int, int, int, int, float[], string)

Create a Tensor of shape [N,H,W,C], an array of data srcData and an optional debug name. srcData must be of size n*h*w*c.

Declaration

public Tensor(int n, int h, int w, int c, float[] srcData, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels
float[]	srcData	source data
string	name	name

Tensor(int, int, int, int, float[][], string)

Create a Tensor of shape [1,1,N,1,1,H,W,C], an array of data srcData and an optional debug name. srcData must be of size n*h*w*c.

Declaration

public Tensor(int n, int h, int w, int c, float[][] srcData, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels
float[][]	srcData	source data
string	name	name

Tensor(int, int, int, int, string)

Create an uninitialized Tensor of shape [1,1,N,1,1,H,W,C] and an optional debug name.

Declaration

public Tensor(int n, int h, int w, int c, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels
string	name	name

Tensor(int, int, int, int, ITensorAllocator)

Create an uninitialized Tensor of shape [1,1,N,1,1,H,W,C] and an ITensorAllocator allocator.

Declaration

public Tensor(int n, int h, int w, int c, ITensorAllocator allocator)

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels
ITensorAllocator	allocator	allocator

Tensor(int, int, int, int, ITensorData, string)

Create a Tensor of shape [1,1,N,1,1,H,W,C], an ITensorData data and an optional debug name. srcData must be of size n*h*w*c.

Declaration

public Tensor(int n, int h, int w, int c, ITensorData data, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels
ITensorData	data	data
string	name	name

Tensor(int, int, int, int, ITensorData, ITensorAllocator)

Create a Tensor of shape [1,1,N,1,1,H,W,C], an ITensorData data and an ITensorAllocator allocator. data must be of size n*h*w*c.

Declaration

public Tensor(int n, int h, int w, int c, ITensorData data, ITensorAllocator allocator)

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels
ITensorData	data	data
ITensorAllocator	allocator	allocator

Tensor(int, int, int, int, ComputeBuffer, string)

Create a Tensor of shape [1,1,N,1,1,H,W,C], associated ComputeBuffer srcBuffer filled with tensor values, and an optional debug name. srcBuffer must be larger than n*h*w*c.

Declaration

public Tensor(int n, int h, int w, int c, ComputeBuffer srcBuffer, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels
ComputeBuffer	srcBuffer	source buffer
string	name	name

Tensor(int, int, float[,], string)

Create a Tensor of shape [1,1,N,1,1,1,1,C], an array of data srcData and an optional debug name. srcData must be of size n*c.

Declaration

public Tensor(int n, int c, float[,] srcData, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	c	channels
float[,]	srcData	source data
string	name	name

Tensor(int, int, float[], string)

Create a Tensor of shape [N,1,1,C], an array of data srcData and an optional debug name. srcData must be of size n*c.

Declaration

public Tensor(int n, int c, float[] srcData, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	c	channels
float[]	srcData	source data
string	name	name

Tensor(int, int, float[][], string)

Create a Tensor of shape [1,1,N,1,1,1,1,C], an array of data srcData and an optional debug name. srcData must be of size n*c.

Declaration

public Tensor(int n, int c, float[][] srcData, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	c	channels
float[][]	srcData	source data
string	name	name

Tensor(int, int, string)

Create an uninitialized Tensor of shape [1,1,N,1,1,1,1,C] and an optional debug name.

Declaration

public Tensor(int n, int c, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	c	channels
string	name	name

Tensor(int, int, ITensorAllocator)

Create an uninitialized Tensor of shape [1,1,N,1,1,1,1,C] and an ITensorAllocator allocator.

Declaration

public Tensor(int n, int c, ITensorAllocator allocator)

Parameters

Type	Name	Description
int	n	batch
int	c	channels
ITensorAllocator	allocator	allocator

Tensor(int, int, ITensorData, string)

Create a Tensor of shape [1,1,N,1,1,1,1,C], an ITensorData data and an optional debug name. srcData must be of size n*c.

Declaration

public Tensor(int n, int c, ITensorData data, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	c	channels
ITensorData	data	data
string	name	name

Tensor(int, int, ITensorData, ITensorAllocator)

Create a Tensor of shape [1,1,N,1,1,1,1,C], an ITensorData data and an ITensorAllocator allocator. srcData must be of size n*c.

Declaration

public Tensor(int n, int c, ITensorData data, ITensorAllocator allocator)

Parameters

Type	Name	Description
int	n	batch
int	c	channels
ITensorData	data	data
ITensorAllocator	allocator	allocator

Tensor(int, int, ComputeBuffer, string)

Create a Tensor of shape [1,1,N,1,1,1,1,C], associated ComputeBuffer srcBuffer filled with tensor values, and an optional debug name. srcBuffer must be larger than n*c.

Declaration

public Tensor(int n, int c, ComputeBuffer srcBuffer, string name = "")

Parameters

Type	Name	Description
int	n	batch
int	c	channels
ComputeBuffer	srcBuffer	source buffer
string	name	name

Tensor(int[], float[,,,], string)

Create a Tensor from a shape, an array of data srcData and an optional name debug name. shape must be of size 8, the order is [S,R,N,T,D,H,W,C]. S and R must be 1. srcData must be of size s[0]*s[1]*s[2]*s[3]*s[4]*s[5]*s[6]*s[7].

Declaration

public Tensor(int[] shape, float[,,,] srcData, string name = "")

Parameters

Type	Name	Description
int[]	shape	shape
float[,,,]	srcData	source data
string	name	name

Tensor(int[], float[,], string)

Create a Tensor from a shape, an array of data srcData and an optional name debug name. shape must be of size 8, the order is [S,R,N,T,D,H,W,C]. S and R must be 1. srcData must be of size s[0]*s[1]*s[2]*s[3]*s[4]*s[5]*s[6]*s[7].

Declaration

public Tensor(int[] shape, float[,] srcData, string name = "")

Parameters

Type	Name	Description
int[]	shape	shape
float[,]	srcData	source data
string	name	name

Tensor(int[], float[], string)

Create a Tensor from a shape, an array of data srcData and an optional debug name. shape must be of size 8, the order is [S,R,N,T,D,H,W,C]. S and R must be 1. srcData must be of size s[0]*s[1]*s[2]*s[3]*s[4]*s[5]*s[6]*s[7].

Declaration

public Tensor(int[] shape, float[] srcData, string name = "")

Parameters

Type	Name	Description
int[]	shape	shape
float[]	srcData	source data
string	name	name

Tensor(int[], float[][], string)

Create a Tensor from a shape, an array of data srcData and an optional name debug name. shape must be of size 8, the order is [S,R,N,T,D,H,W,C]. S and R must be 1. srcData must be of size s[0]*s[1]*s[2]*s[3]*s[4]*s[5]*s[6]*s[7].

Declaration

public Tensor(int[] shape, float[][] srcData, string name = "")

Parameters

Type	Name	Description
int[]	shape	shape
float[][]	srcData	source data
string	name	name

Tensor(int[], string)

Create an uninitialized Tensor from a shape and an optional debug name. shape must be of size 8, the order is [S,R,N,T,D,H,W,C] S and R must be 1.

Declaration

public Tensor(int[] shape, string name = "")

Parameters

Type	Name	Description
int[]	shape	shape
string	name	name

Tensor(int[], ITensorAllocator)

Create an uninitialized Tensor from a shape and an ITensorAllocator allocator. shape must be of size 8, the order is [S,R,N,T,D,H,W,C]. S and R must be 1.

Declaration

public Tensor(int[] shape, ITensorAllocator allocator)

Parameters

Type	Name	Description
int[]	shape	shape
ITensorAllocator	allocator	allocator

Tensor(int[], ITensorData, string)

Create a Tensor from a shape, an ITensorData data and an optional debug name. shape must be of size 8, the order is [S,R,N,T,D,H,W,C]. S and R must be 1.

Declaration

public Tensor(int[] shape, ITensorData data, string name = "")

Parameters

Type	Name	Description
int[]	shape	shape
ITensorData	data	data
string	name	name

Tensor(int[], ITensorData, ITensorAllocator)

Create a Tensor from a shape, an ITensorData data and an ITensorAllocator allocator. shape must be of size 8, the order is [S,R,N,T,D,H,W,C]. S and R must be 1.

Declaration

public Tensor(int[] shape, ITensorData data, ITensorAllocator allocator)

Parameters

Type	Name	Description
int[]	shape	shape
ITensorData	data	data
ITensorAllocator	allocator	allocator

Tensor(int[], ComputeBuffer, string)

Create a Tensor from a shape, associated ComputeBuffer srcBuffer filled with tensor values, and an optional debug name. shape must be of size 8, the order is [S,R,N,T,D,H,W,C]. S and R must be 1. srcBuffer must be larger than s[0]*s[1]*s[2]*s[3]*s[4]*s[5]*s[6]*s[7].

Declaration

public Tensor(int[] shape, ComputeBuffer srcBuffer, string name = "")

Parameters

Type	Name	Description
int[]	shape	shape
ComputeBuffer	srcBuffer	source buffer
string	name	name

Tensor(string)

Create an uninitialized Tensor with a shape of [1,1,1,1,1,1,1,1] and an optional debug name.

Declaration

public Tensor(string name = "")

Parameters

Type	Name	Description
string	name	name

Tensor(ITensorAllocator)

Create an uninitialized Tensor with a shape of [1,1,1,1,1,1,1,1] and an ITensorAllocator allocator.

Declaration

public Tensor(ITensorAllocator allocator)

Parameters

Type	Name	Description
ITensorAllocator	allocator	allocator

Tensor(TensorShape, float[,,,], string)

Create a Tensor with specified shape, an array of data srcData and an optional debug name. srcData must be of size shape.length.

Declaration

public Tensor(TensorShape shape, float[,,,] srcData, string name = "")

Parameters

Type	Name	Description
TensorShape	shape	shape
float[,,,]	srcData	source data
string	name	name

Tensor(TensorShape, float[,], string)

Create a Tensor with specified shape, an array of data srcData and an optional debug name. srcData must be of size shape.length.

Declaration

public Tensor(TensorShape shape, float[,] srcData, string name = "")

Parameters

Type	Name	Description
TensorShape	shape	shape
float[,]	srcData	source data
string	name	name

Tensor(TensorShape, float[], string)

Create a Tensor with specified shape, an array of data srcData and an optional debug name. srcData must be of size shape.length.

Declaration

public Tensor(TensorShape shape, float[] srcData, string name = "")

Parameters

Type	Name	Description
TensorShape	shape	shape
float[]	srcData	source data
string	name	name

Tensor(TensorShape, float[][], string)

Create a Tensor with specified shape, an array of data srcData and an optional debug name. srcData must be of size shape.length.

Declaration

public Tensor(TensorShape shape, float[][] srcData, string name = "")

Parameters

Type	Name	Description
TensorShape	shape	shape
float[][]	srcData	source data
string	name	name

Tensor(TensorShape, string)

Create an uninitialized Tensor with specified shape and an optional debug name.

Declaration

public Tensor(TensorShape shape, string name = "")

Parameters

Type	Name	Description
TensorShape	shape	shape
string	name	name

Tensor(TensorShape, ITensorAllocator)

Create an uninitialized Tensor with specified shape and ITensorAllocator allocator.

Declaration

public Tensor(TensorShape shape, ITensorAllocator allocator)

Parameters

Type	Name	Description
TensorShape	shape	shape
ITensorAllocator	allocator	allocator

Tensor(TensorShape, ITensorData, string)

Create a Tensor with specified shape, an ITensorData data and an optional debug name.

Declaration

public Tensor(TensorShape shape, ITensorData data, string name = "")

Parameters

Type	Name	Description
TensorShape	shape	shape
ITensorData	data	data
string	name	name

Tensor(TensorShape, ITensorData, ITensorAllocator)

Create a Tensor with specified shape, an ITensorData data and an ITensorAllocator allocator

Declaration

public Tensor(TensorShape shape, ITensorData data, ITensorAllocator allocator)

Parameters

Type	Name	Description
TensorShape	shape	shape
ITensorData	data	data
ITensorAllocator	allocator	allocator

Tensor(TensorShape, ComputeBuffer, string)

Create a Tensor with specified shape, associated ComputeBuffer srcBuffer filled with tensor values, and an optional debug name. srcBuffer must be larger than shape.length.

Declaration

public Tensor(TensorShape shape, ComputeBuffer srcBuffer, string name = "")

Parameters

Type	Name	Description
TensorShape	shape	shape
ComputeBuffer	srcBuffer	source buffer
string	name	name

Exceptions

Type	Condition
ArgumentException	thrown if specified buffer is too small or stride is mismatched

Tensor(Texture, int, string)

Create a Tensor from a texture, shape is [1,1,1,1,1, texture.height, texture.width, channels]. If channels is set to -1 (default value), then number of channels in the new Tensor will match the number of channels in the texture. Just like Texture2D.GetPixels when reading from LDR texture (RGBA32, ARGB32, RGB24, Alpha8, RG16, R8, etc) this function will remap pixel values from byte values to the range of [0.0 .. 1.0]. Pixel values from HDR textures (such as ARGBFloat or ARGBHalf) will be left unchanged.

Declaration

public Tensor(Texture srcTexture, int channels = -1, string name = "")

Parameters

Type	Name	Description
Texture	srcTexture	source texture
int	channels	channels
string	name	name

Tensor(Texture[], int, string)

Create a Tensor from multiple texture, shape is [1,1,1,1, srcTextures.length, texture.height, texture.width, channels]. If channels is set to -1 (default value), then number of channels in the new Tensor will match the number of channels in the texture. All textures must be of the same size and dimension. Just like Texture2D.GetPixels when reading from LDR texture (RGBA32, ARGB32, RGB24, Alpha8, RG16, R8, etc) this function will remap pixel values from byte values to the range of [0.0 .. 1.0]. Pixel values from HDR textures (such as ARGBFloat or ARGBHalf) will be left unchanged.

Declaration

public Tensor(Texture[] srcTextures, int channels = -1, string name = "")

Parameters

Type	Name	Description
Texture[]	srcTextures	source textures
int	channels	channels
string	name	name

Fields

name

Return this tensor name.

Declaration

public string name

Field Value

Type	Description
string

shape

Return this tensor shape as [S,R,N,T,D,H,W,C].

Declaration

public readonly TensorShape shape

Field Value

Type	Description
TensorShape

Properties

this[int]

Access element at offset index in this Tensor. This will create a blocking read, if this Tensor is a result of a computation on a different device (GPU).

Declaration

public float this[int index] { get; set; }

Parameters

Type	Name	Description
int	index	flat index

Property Value

Type	Description
float

this[int, int]

Access element at index [0,0,N,0,0,0,0,C] in this Tensor. This will create a blocking read, if this Tensor is a result of a computation on a different device (GPU).

Declaration

public float this[int b, int ch] { get; set; }

Parameters

Type	Name	Description
int	b	batch
int	ch	channels

Property Value

Type	Description
float

this[int, int, int, int]

Access element at index [0,0,N,0,0,H,W,C] in this Tensor. This will create a blocking read, if this Tensor is a result of a computation on a different device (GPU).

Declaration

public float this[int b, int h, int w, int ch] { get; set; }

Parameters

Type	Name	Description
int	b	batch
int	h	height
int	w	width
int	ch	channels

Property Value

Type	Description
float

this[int, int, int, int, int]

Access element at index [0,0,N,0,D,H,W,C] in this Tensor. This will create a blocking read, if this Tensor is a result of a computation on a different device (GPU).

Declaration

public float this[int b, int d, int h, int w, int ch] { get; set; }

Parameters

Type	Name	Description
int	b
int	d
int	h
int	w
int	ch

Property Value

Type	Description
float

this[int, int, int, int, int, int, int, int]

Access element at index [S,R,N,T,D,H,W,C] in this Tensor. This will create a blocking read, if this Tensor is a result of a computation on a different device (GPU).

Declaration

public float this[int s, int r, int n, int t, int d, int h, int w, int c] { get; set; }

Parameters

Type	Name	Description
int	s	sequence
int	r	direction
int	n	batch
int	t	time
int	d	depth
int	h	height
int	w	width
int	c	channels

Property Value

Type	Description
float

allocator

Return this tensor allocator, see interface ITensorAllocator.

Declaration

public ITensorAllocator allocator { get; }

Property Value

Type	Description
ITensorAllocator

batch

Return the number of batches.

Declaration

public int batch { get; }

Property Value

Type	Description
int

channels

Return the number of channels.

Declaration

public int channels { get; }

Property Value

Type	Description
int

data

Upload data to device and return its instance

Declaration

public ITensorData data { get; }

Property Value

Type	Description
ITensorData

depth

Return the spatial depth.

Declaration

public int depth { get; }

Property Value

Type	Description
int

dimensions

Return the count of non-unit dimension of this tensor shape. For example [1,1,N,1,1,1,1,C] dimensions is 2.

Declaration

public int dimensions { get; }

Property Value

Type	Description
int

flatHeight

Return the number of batch.

Declaration

public int flatHeight { get; }

Property Value

Type	Description
int

flatWidth

Return TDHWC.

Declaration

public int flatWidth { get; }

Property Value

Type	Description
int

height

Return the spatial height.

Declaration

public int height { get; }

Property Value

Type	Description
int

kernelCount

Kernel dimension ordering is [D,H,W,C,K] for efficiency purpose. Return kernel count (aka the number of output channels of the associated operator).

Declaration

public int kernelCount { get; }

Property Value

Type	Description
int

kernelDepth

Kernel dimension ordering is [D,H,W,C,K] for efficiency purpose. Return kernel depth (aka the number of input channels of the associated operator).

Declaration

public int kernelDepth { get; }

Property Value

Type	Description
int

kernelHeight

Kernel dimension ordering is [D,H,W,C,K] for efficiency purpose. Return kernel spatial height.

Declaration

public int kernelHeight { get; }

Property Value

Type	Description
int

kernelSpatialDepth

Kernel dimension ordering is [D,H,W,C,K] for efficiency purpose. Return kernel spatial depth.

Declaration

public int kernelSpatialDepth { get; }

Property Value

Type	Description
int

kernelWidth

Kernel dimension ordering is [D,H,W,C,K] for efficiency purpose. Return kernel spatial width.

Declaration

public int kernelWidth { get; }

Property Value

Type	Description
int

length

Return the total number of elements in this tensor.

Declaration

public int length { get; }

Property Value

Type	Description
int

numberOfDirections

Return the number of directions.

Declaration

public int numberOfDirections { get; }

Property Value

Type	Description
int

readonlyArray

Read-only array of Tensor data

Declaration

[Obsolete("Use ToReadOnlyArray instead.", false)]
public float[] readonlyArray { get; }

Property Value

Type	Description
float[]

readonlyArrayOffset

Offset into read-only array of Tensor data

Declaration

[Obsolete("Use ToReadOnlyArray instead.", false)]
public int readonlyArrayOffset { get; }

Property Value

Type	Description
int

sequenceLength

Return the number of sequences.

Declaration

public int sequenceLength { get; }

Property Value

Type	Description
int

tensorOnDevice

Device specific internal representation of Tensor data

Declaration

public ITensorData tensorOnDevice { get; }

Property Value

Type	Description
ITensorData

width

Return the spatial width.

Declaration

public int width { get; }

Property Value

Type	Description
int

Methods

AttachToDevice(ITensorData)

Associates tensor with the block of data residing on a device. Tensor values will be downloaded from the source upon the first access. source should contain initialized and valid data representing tensor values. See also PrepareCacheForAccess() to schedule download as soon as possible.

Declaration

public void AttachToDevice(ITensorData source)

Parameters

Type	Name	Description
ITensorData	source	source

Axis(int)

Allow to use negative axis to access tensorShape backward. axis should be from -rank to rank (exclusive).

Declaration

public int Axis(int axis)

Parameters

Type	Name	Description
int	axis	axis

Returns

Type	Description
int	remapped axis

DeepCopy()

Create a copy of the current Tensor.

Declaration

public Tensor DeepCopy()

Returns

Type	Description
Tensor	new copy of the Tensor

DetachFromDevice(bool)

Remove tensor from device, will first sync the cache with device data.

Declaration

public ITensorData DetachFromDevice(bool disposeDeviceData = true)

Parameters

Type	Name	Description
bool	disposeDeviceData	dispose device data

Returns

Type	Description
ITensorData	Tensor data

Dispose()

Dispose Tensor and associated memories.

Declaration

public virtual void Dispose()

~Tensor()

Destructor will also dispose associated memories.

Declaration

protected ~Tensor()

Flatten(string)

Create a flattened copy of the current Tensor ie of shape [1,1,N,1,1,1,1,TDHWC]

Declaration

public Tensor Flatten(string newName = null)

Parameters

Type	Name	Description
string	newName	new name

Returns

Type	Description
Tensor	shallow copy of the Tensor with new shape

FlushCache()

Upload cache to device memory and delete it.

Declaration

public void FlushCache()

Index(int, int)

Given an element dimensions indices [0,0,N,0,0,0,0,C] return this element offset in memory.

Declaration

public int Index(int y, int x)

Parameters

Type	Name	Description
int	y	y
int	x	x

Returns

Type	Description
int	flat index (offset in memory)

Index(int, int, int, int)

Given an element dimensions indices [0,0,N,0,0,H,W,C] return this element offset in memory.

Declaration

public int Index(int b, int h, int w, int ch)

Parameters

Type	Name	Description
int	b	batch
int	h	height
int	w	width
int	ch	channels

Returns

Type	Description
int	flat index (offset in memory)

Index(int, int, int, int, int)

Given an element dimensions indices [0,0,N,0,D,H,W,C] return this element offset in memory.

Declaration

public int Index(int b, int d, int h, int w, int ch)

Parameters

Type	Name	Description
int	b	batch
int	d	depth
int	h	height
int	w	width
int	ch	channels

Returns

Type	Description
int

Index(int, int, int, int, int, int, int, int)

Given an element dimensions indices [S,R,N,T,D,H,W,C] return this element offset in memory.

Declaration

public int Index(int s, int r, int n, int t, int d, int h, int w, int c)

Parameters

Type	Name	Description
int	s	sequence
int	r	direction
int	n	batch
int	t	time
int	d	depth
int	h	height
int	w	width
int	c	channels

Returns

Type	Description
int	flat index (offset in memory)

IndexWithBroadcast(int, int, int, int)

Given an element dimensions indices[0,0,N,0,0,H,W,C] with broadcast support, return this element offset in memory.

Declaration

public int IndexWithBroadcast(int n, int h, int w, int c)

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels

Returns

Type	Description
int	flat index (offset in memory)

IndexWithBroadcast(int, int, int, int, int, int, int, int)

Given an element dimensions indices [S,R,N,T,D,H,W,C] with broadcast support, return this element offset in memory.

Declaration

public int IndexWithBroadcast(int s, int r, int n, int t, int d, int h, int w, int c)

Parameters

Type	Name	Description
int	s	sequence
int	r	direction
int	n	batch
int	t	time
int	d	depth
int	h	height
int	w	width
int	c	channels

Returns

Type	Description
int	flat index (offset in memory)

IndexWithClamp(int, int, int, int)

Given an element dimensions indices [0,0,N,0,0,H,W,C] return this element offset in memory, clamping indices to tensor dimensions.

Declaration

public int IndexWithClamp(int n, int h, int w, int c)

Parameters

Type	Name	Description
int	n	batch
int	h	height
int	w	width
int	c	channels

Returns

Type	Description
int	flat index (offset in memory)

IndexWithClamp(int, int, int, int, int)

Given an element dimensions indices [0,0,N,0,D,H,W,C] return this element offset in memory, clamping indices to tensor dimensions.

Declaration

public int IndexWithClamp(int n, int d, int h, int w, int c)

Parameters

Type	Name	Description
int	n	batch
int	d	depth
int	h	height
int	w	width
int	c	channels

Returns

Type	Description
int	flat index (offset in memory)

PrepareCacheForAccess(bool)

Populate the cache with on device data. Blocking read if blocking is true (default)

Declaration

public bool PrepareCacheForAccess(bool blocking = true)

Parameters

Type	Name	Description
bool	blocking	blocking read if true

Returns

Type	Description
bool	true if data is ready

Reshape(TensorShape, string)

Create a reshaped copy of the current Tensor. newShape.length must be equal to this.shape.length.

Declaration

public Tensor Reshape(TensorShape newShape, string newName = null)

Parameters

Type	Name	Description
TensorShape	newShape	new shape
string	newName	new name

Returns

Type	Description
Tensor	shallow copy of the Tensor with new shape and name

ShallowCopy(string)

Create a copy of the current Tensor, sharing data storage with original tensor.

Declaration

public Tensor ShallowCopy(string newName = null)

Parameters

Type	Name	Description
string	newName	new name

Returns

Type	Description
Tensor	shallow copy of the Tensor

TakeOwnership()

Remove system reference to this tensor, caller assume ownership.

Declaration

public void TakeOwnership()

ToReadOnlyArray()

Return the cached linear memory representation of this tensor data. This will create a blocking read, if this Tensor is a result of a computation on a different device (GPU). IMPORTANT: Modifying contents of the returned array will have undefined behavior.

Declaration

public float[] ToReadOnlyArray()

Returns

Type	Description
float[]	cached linear memory representation of this tensor data

ToRenderTexture(int, int, float, float, Texture3D)

Create new RenderTexture and fill it with a portion of the tensor applying scale and bias. Portion of the target is specified by batch and fromChannel. batch specifies the tensor batch to read values from. fromChannel specifies the first tensor channel to start reading values from. Number of channels in the target texture specifies how many channels to read from the tensor, starting from index fromChannel. Resolution of the target must match the spatial dimensions of the tensor. scale multiplier and bias addition is applied to the values read from the tensor and clamped to the range from 0.0 to 1.0.

Declaration

public RenderTexture ToRenderTexture(int batch = 0, int fromChannel = 0, float scale = 1, float bias = 0, Texture3D lut = null)

Parameters

Type	Name	Description
int	batch	batch
int	fromChannel	from channel
float	scale	scale
float	bias	bias
Texture3D	lut	lut table

Returns

Type	Description
RenderTexture

ToRenderTexture(RenderTexture, int, int, float, float, Texture3D)

Fill a target RenderTexture with a portion of the tensor applying scale and bias. Portion of the target is specified by batch and fromChannel. batch specifies the tensor batch to read values from. fromChannel specifies the first tensor channel to start reading values from. Number of channels in the target texture specifies how many channels to read from the tensor, starting from index fromChannel. Resolution of the target must match the spatial dimensions of the tensor. scale multiplier and bias addition is applied to the values read from the tensor and, if target is LDR texture (RGBA32, ARGB32, RGB24, Alpha8, RG16, R8, etc), clamped to the range from 0.0 to 1.0.

Declaration

public void ToRenderTexture(RenderTexture target, int batch = 0, int fromChannel = 0, float scale = 1, float bias = 0, Texture3D lut = null)

Parameters

Type	Name	Description
RenderTexture	target	target RenderTexture
int	batch	batch
int	fromChannel	from channel
float	scale	scale
float	bias	bias
Texture3D	lut	lut table

ToRenderTexture(RenderTexture, int, int, Vector4, Vector4, Texture3D)

Fill a target RenderTexture with a portion of the tensor applying scale and bias. Portion of the target is specified by batch and fromChannel. batch specifies the tensor batch to read values from. fromChannel specifies the first tensor channel to start reading values from. Number of channels in the target texture specifies how many channels to read from the tensor, starting from index fromChannel. Resolution of the target must match the spatial dimensions of the tensor. scale multiplier and bias addition is applied to the values read from the tensor and, if target is LDR texture (RGBA32, ARGB32, RGB24, Alpha8, RG16, R8, etc), clamped to the range from 0.0 to 1.0.

Declaration

public void ToRenderTexture(RenderTexture target, int batch, int fromChannel, Vector4 scale, Vector4 bias, Texture3D lut = null)

Parameters

Type	Name	Description
RenderTexture	target	target RenderTexture
int	batch	batch
int	fromChannel	from channel
Vector4	scale	scale
Vector4	bias	bias
Texture3D	lut	lut table

ToRenderTexture(RenderTextureFormat, int, int, float, float, Texture3D)

Create new RenderTexture and fill it with a portion of the tensor applying scale and bias. Portion of the target is specified by batch and fromChannel. format specifies the type of the new RenderTexture. batch specifies the tensor batch to read values from. fromChannel specifies the first tensor channel to start reading values from. Number of channels in the target texture specifies how many channels to read from the tensor, starting from index fromChannel. scale multiplier and bias addition is applied to the values read from the tensor and, if format is LDR (RGBA32, ARGB32, RGB24, Alpha8, RG16, R8, etc), clamped to the range from 0.0 to 1.0.

Declaration

public RenderTexture ToRenderTexture(RenderTextureFormat format, int batch = 0, int fromChannel = 0, float scale = 1, float bias = 0, Texture3D lut = null)

Parameters

Type	Name	Description
RenderTextureFormat	format	RenderTexture format
int	batch	batch
int	fromChannel	from channel
float	scale	scale
float	bias	bias
Texture3D	lut	lut table

Returns

Type	Description
RenderTexture	created RenderTexture

ToString()

Tensor metadata summary

Declaration

public override string ToString()

Returns

Type	Description
string	Tensor metadata summary

Overrides

object.ToString()

UnpinAndDisposeTensor()

Unload tensor data from device and dispose this Tensor

Declaration

[Obsolete("Use Dispose instead.", false)]
public ITensorData UnpinAndDisposeTensor()

Returns

Type	Description
ITensorData	device specific Tensor data

UploadToDevice(ITensorData, bool)

Upload tensor values to the device. This call associates tensor with the uninitialized block of data residing on a device. destination should be allocated on a target device. Previous contents of destination will be overwritten after this call. By default local cache will be discarded after this call, set invalidateCacheAfterUpload to false to keep the cache.

Declaration

public void UploadToDevice(ITensorData destination, bool invalidateCacheAfterUpload = true)

Parameters

Type	Name	Description
ITensorData	destination	destination
bool	invalidateCacheAfterUpload	invalidate cache after upload

Events

tensorDisposed

Declaration

public static event Action<Tensor> tensorDisposed

Event Type

Type	Description
Action<Tensor>

Implements

IDisposable

Extension Methods

DeprecatedTensorExtensions.AdjustPadToPool(Tensor, (int, int), int[], int[])

DeprecatedTensorExtensions.CastOnDevice(Tensor, ITensorData)

DeprecatedTensorExtensions.PinToDeviceAndDownloadFromIt(Tensor, ITensorData)

DeprecatedTensorExtensions.PinToDeviceAndUploadToIt(Tensor, ITensorData, bool)

DeprecatedTensorExtensions.Unpin(Tensor, bool)

TensorExtensions.Approximately(Tensor, Tensor, float, int)

TensorExtensions.ArgMax(Tensor)

TensorExtensions.ArgSort(Tensor)

TensorExtensions.AsFloats(Tensor)

TensorExtensions.AsInts(Tensor)

TensorExtensions.DataToString(Tensor, int)

TensorExtensions.Equals(Tensor, Tensor)

TensorExtensions.Fill(Tensor, float)

TensorExtensions.MaxDifference(Tensor, Tensor)

TensorExtensions.Print(Tensor, string)

TensorExtensions.PrintDataPart(Tensor, int, string)

TensorExtensions.Reshape(Tensor, int[])