Class CompareOps

Compares output of two different implementations of IOps. Useful for debugging purposes

Inheritance
Object
CompareOps
Namespace: Unity.Barracuda
Syntax
public class CompareOps : object, IOps, IOpsStatistics, IModelCompiler

Constructors

CompareOps(IOps, IOps, CompareOpsUtils.LogLevel, Single)

Create CompareOps

Declaration
public CompareOps(IOps ops1, IOps ops2, CompareOpsUtils.LogLevel differenceLogLevel, float epsilon)
Parameters
Type Name Description
IOps ops1

first IOps implementation
IOps ops2

second IOps implementation
CompareOpsUtils.LogLevel differenceLogLevel

difference log level
Single epsilon

error threshold

Methods

LSTM(Tensor, Tensor[], Tensor[], Tensor[], Tensor[], Tensor, Tensor)

LSTM

Declaration
public Tensor[] LSTM(Tensor X, Tensor[] W, Tensor[] R, Tensor[] Wb, Tensor[] Rb, Tensor hidden, Tensor cell)
Parameters
Type Name Description
Tensor X

The input sequences packed into one 3-D tensor.
Tensor[] W

W parameter weight matrix for input, output, forget, and cell gates - W[iofc]
Tensor[] R

R recurrence weight matrix for input, output, forget, and cell gates - R[iofc]
Tensor[] Wb

W bias vectors for input, output, forget, and cell gates - Wb[iofc]
Tensor[] Rb

R bias vectors for input, output, forget, and cell gates - Rb[iofc]
Tensor hidden

Initial value of the hidden
Tensor cell

Initial value of the cell
Returns
Type Description
Tensor[]

[Y (concatenated intermediate values of the hidden), Y_h (final hidden), Y_c (final cell)]
Implements

NonZero(Tensor)

Indices for non zero values

Declaration
public Tensor NonZero(Tensor X)
Parameters
Type Name Description
Tensor X

input
Returns
Type Description
Tensor

output Tensor
Implements

PostLayerCleanup()

Called after every layer execution. It allows IOps to run cleanup operations such as clearing temporary buffers only used in the scope of the last layer executed.

Declaration
public virtual void PostLayerCleanup()
Implements

PreExecuteLayer(Layer, Tensor[])

Declaration
public virtual void PreExecuteLayer(Layer layer, Tensor[] inputs)
Parameters
Type Name Description
Layer layer
Tensor[] inputs

PrepareModel(Model, IDictionary<String, TensorShape>, IVars)

Declaration
public virtual void PrepareModel(Model model, IDictionary<string, TensorShape> inputShapes, IVars vars)
Parameters
Type Name Description
Model model
IDictionary<String, TensorShape> inputShapes
IVars vars

TopKValues(Tensor, Tensor, Int32)

Top K values

Declaration
public Tensor TopKValues(Tensor X, Tensor I, int axis)
Parameters
Type Name Description
Tensor X

input
Tensor I

indices
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

Explicit Interface Implementations

IOps.Abs(Tensor)

Abs

Declaration
Tensor IOps.Abs(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Acos(Tensor)

Acos

Declaration
Tensor IOps.Acos(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Acosh(Tensor)

Acosh

Declaration
Tensor IOps.Acosh(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Add(Tensor[])

Add tensors together

Declaration
Tensor IOps.Add(Tensor[] tensors)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ArgMax(Tensor, Int32)

ArgMax

Declaration
Tensor IOps.ArgMax(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ArgMin(Tensor, Int32)

ArgMax

Declaration
Tensor IOps.ArgMin(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Asin(Tensor)

Asin

Declaration
Tensor IOps.Asin(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Asinh(Tensor)

Asinh

Declaration
Tensor IOps.Asinh(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Atan(Tensor)

Atan

Declaration
Tensor IOps.Atan(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Atanh(Tensor)

Atanh

Declaration
Tensor IOps.Atanh(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.AvgPool2D(Tensor, Int32[], Int32[], Int32[])

2D average pooling

Declaration
Tensor IOps.AvgPool2D(Tensor X, int[] pool, int[] stride, int[] pad)
Parameters
Type Name Description
Tensor X
Int32[] pool

pooling
Int32[] stride

stride
Int32[] pad

padding
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Border2D(Tensor, Int32[], Single)

2D border padding

Declaration
Tensor IOps.Border2D(Tensor x, int[] pad, float value)
Parameters
Type Name Description
Tensor x

input
Int32[] pad

padding
Single value
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Border3D(Tensor, Int32[], Single)

3D border padding

Declaration
Tensor IOps.Border3D(Tensor x, int[] pad, float value)
Parameters
Type Name Description
Tensor x

input
Int32[] pad

padding
Single value
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Ceil(Tensor)

Ceil

Declaration
Tensor IOps.Ceil(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Clip(Tensor, Single, Single)

Clip

Declaration
Tensor IOps.Clip(Tensor X, float min, float max)
Parameters
Type Name Description
Tensor X
Single min

min value
Single max

max value
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Concat(Tensor[], Int32)

Concatenate tensors across axis

Declaration
Tensor IOps.Concat(Tensor[] tensors, int axis)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ConstantOfShape(TensorShape, Single)

Creates a constant of shape input

Declaration
Tensor IOps.ConstantOfShape(TensorShape X, float value)
Parameters
Type Name Description
TensorShape X

input shape
Single value

value
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Conv2D(Tensor, Tensor, Tensor, Int32[], Int32[], Layer.FusedActivation)

2D convolution

Declaration
Tensor IOps.Conv2D(Tensor X, Tensor K, Tensor B, int[] stride, int[] pad, Layer.FusedActivation fusedActivation)
Parameters
Type Name Description
Tensor X
Tensor K
Tensor B
Int32[] stride

stride
Int32[] pad

padding
Layer.FusedActivation fusedActivation

fused activation type
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Conv2DTrans(Tensor, Tensor, Tensor, Int32[], Int32[], Int32[], Layer.FusedActivation)

Transpose 2D convolution

Declaration
Tensor IOps.Conv2DTrans(Tensor X, Tensor K, Tensor B, int[] stride, int[] pad, int[] outputAdjustment, Layer.FusedActivation fusedActivation)
Parameters
Type Name Description
Tensor X
Tensor K
Tensor B
Int32[] stride

stride
Int32[] pad

padding
Int32[] outputAdjustment

output adjustments
Layer.FusedActivation fusedActivation

fused activation type
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Conv3D(Tensor, Tensor, Tensor, Int32[], Int32[], Layer.FusedActivation)

3D convolution

Declaration
Tensor IOps.Conv3D(Tensor X, Tensor K, Tensor B, int[] stride, int[] pad, Layer.FusedActivation fusedActivation)
Parameters
Type Name Description
Tensor X
Tensor K
Tensor B
Int32[] stride

stride
Int32[] pad

padding
Layer.FusedActivation fusedActivation

fused activation type
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Copy(Tensor)

Copy

Declaration
Tensor IOps.Copy(Tensor x)
Parameters
Type Name Description
Tensor x

input
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Cos(Tensor)

Cos

Declaration
Tensor IOps.Cos(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Cosh(Tensor)

Cosh

Declaration
Tensor IOps.Cosh(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Dense(Tensor, Tensor, Tensor, Layer.FusedActivation)

Dense layer (matrix multiplication) o = x ⨯ w + b

Declaration
Tensor IOps.Dense(Tensor X, Tensor W, Tensor B, Layer.FusedActivation fusedActivation)
Parameters
Type Name Description
Tensor X
Tensor W
Tensor B
Layer.FusedActivation fusedActivation

fused activation type
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Dense3(Tensor, Tensor, Tensor)

rank3 Dense layer (matrix multiplication) o = x ⨯ w + b O: N,,W,C / X: N,,W,C / W:N,,,C / B:N,,,_

Declaration
Tensor IOps.Dense3(Tensor X, Tensor W, Tensor B)
Parameters
Type Name Description
Tensor X
Tensor W
Tensor B
Returns
Type Description
Tensor

output Tensor
Implements

IOps.DepthToSpace(Tensor, Int32[], Layer.DepthToSpaceMode)

Depth to space

Declaration
Tensor IOps.DepthToSpace(Tensor X, int[] scale, Layer.DepthToSpaceMode mode)
Parameters
Type Name Description
Tensor X
Int32[] scale

scale
Layer.DepthToSpaceMode mode

mode
Returns
Type Description
Tensor

output Tensor
Implements

IOps.DepthwiseConv2D(Tensor, Tensor, Tensor, Int32[], Int32[], Layer.FusedActivation)

Depthwise 2D convolution

Declaration
Tensor IOps.DepthwiseConv2D(Tensor X, Tensor K, Tensor B, int[] stride, int[] pad, Layer.FusedActivation fusedActivation)
Parameters
Type Name Description
Tensor X
Tensor K
Tensor B
Int32[] stride

stride
Int32[] pad

padding
Layer.FusedActivation fusedActivation

fused activation type
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Div(Tensor[])

Divide tensors o = tensors[0] / tensors[1] / ... / tensors[N-1]

Declaration
Tensor IOps.Div(Tensor[] tensors)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Dropout(Tensor, Single)

Dropout

Declaration
Tensor IOps.Dropout(Tensor X, float alpha)
Parameters
Type Name Description
Tensor X
Single alpha

alpha
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Elu(Tensor, Single)

ELU

Declaration
Tensor IOps.Elu(Tensor X, float alpha)
Parameters
Type Name Description
Tensor X
Single alpha

alpha
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Equal(Tensor, Tensor)

Equal

Declaration
Tensor IOps.Equal(Tensor a, Tensor b)
Parameters
Type Name Description
Tensor a

left Tensor
Tensor b

right Tensor
Returns
Type Description
Tensor

Tensor with true where a == b
Implements

IOps.Erf(Tensor)

Erf

Declaration
Tensor IOps.Erf(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Exp(Tensor)

Exponent e^x

Declaration
Tensor IOps.Exp(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Expand(Tensor, TensorShape)

Expand

Declaration
Tensor IOps.Expand(Tensor X, TensorShape shape)
Parameters
Type Name Description
Tensor X
TensorShape shape

new shape
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Flatten(Tensor)

Flatten

Declaration
Tensor IOps.Flatten(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Floor(Tensor)

Floor

Declaration
Tensor IOps.Floor(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Gather(Tensor[], Int32)

Gather

Declaration
Tensor IOps.Gather(Tensor[] tensors, int axis)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.GetModelExecutionsReporter()

Get model executions reporter

Declaration
IModelExecutionsReporter IOps.GetModelExecutionsReporter()
Returns
Type Description
IModelExecutionsReporter

model executions reporter
Implements

IOps.GlobalAvgPool2D(Tensor)

2D global average pooling

Declaration
Tensor IOps.GlobalAvgPool2D(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.GlobalAvgVariancePool2D(Tensor)

2D global average variance pooling

Declaration
Tensor IOps.GlobalAvgVariancePool2D(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.GlobalMaxPool2D(Tensor)

2D global max pooling

Declaration
Tensor IOps.GlobalMaxPool2D(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Greater(Tensor, Tensor)

Greater

Declaration
Tensor IOps.Greater(Tensor a, Tensor b)
Parameters
Type Name Description
Tensor a

left Tensor
Tensor b

right Tensor
Returns
Type Description
Tensor

Tensor with true where a > b
Implements

IOps.GreaterEqual(Tensor, Tensor)

Greater or equal

Declaration
Tensor IOps.GreaterEqual(Tensor a, Tensor b)
Parameters
Type Name Description
Tensor a

left Tensor
Tensor b

right Tensor
Returns
Type Description
Tensor

Tensor with true where a >= b
Implements

IOps.HardSigmoid(Tensor, Single, Single)

HardSigmoid

Declaration
Tensor IOps.HardSigmoid(Tensor X, float alpha, float beta)
Parameters
Type Name Description
Tensor X
Single alpha

alpha
Single beta
Returns
Type Description
Tensor

output Tensor
Implements

IOps.LeakyRelu(Tensor, Single)

Leaky ReLU

Declaration
Tensor IOps.LeakyRelu(Tensor X, float alpha)
Parameters
Type Name Description
Tensor X
Single alpha

alpha
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Less(Tensor, Tensor)

Less

Declaration
Tensor IOps.Less(Tensor a, Tensor b)
Parameters
Type Name Description
Tensor a

left Tensor
Tensor b

right Tensor
Returns
Type Description
Tensor

Tensor with true where a < b
Implements

IOps.LessEqual(Tensor, Tensor)

Less or equal

Declaration
Tensor IOps.LessEqual(Tensor a, Tensor b)
Parameters
Type Name Description
Tensor a

left Tensor
Tensor b

right Tensor
Returns
Type Description
Tensor

Tensor with true where a < b
Implements

IOps.Log(Tensor)

Log

Declaration
Tensor IOps.Log(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.LogicalAnd(Tensor, Tensor)

And

Declaration
Tensor IOps.LogicalAnd(Tensor a, Tensor b)
Parameters
Type Name Description
Tensor a

left Tensor
Tensor b

right Tensor
Returns
Type Description
Tensor

Tensor with true where a && b
Implements

IOps.LogicalNot(Tensor)

Not

Declaration
Tensor IOps.LogicalNot(Tensor x)
Parameters
Type Name Description
Tensor x

input
Returns
Type Description
Tensor

Tensor with !x values
Implements

IOps.LogicalOr(Tensor, Tensor)

Or

Declaration
Tensor IOps.LogicalOr(Tensor a, Tensor b)
Parameters
Type Name Description
Tensor a

left Tensor
Tensor b

right Tensor
Returns
Type Description
Tensor

Tensor with true where a || b
Implements

IOps.LogicalXor(Tensor, Tensor)

Xor

Declaration
Tensor IOps.LogicalXor(Tensor a, Tensor b)
Parameters
Type Name Description
Tensor a

left Tensor
Tensor b

right Tensor
Returns
Type Description
Tensor

Tensor with true where a xor b
Implements

IOps.LogSoftmax(Tensor, Int32)

LogSoftmax

Declaration
Tensor IOps.LogSoftmax(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.LRN(Tensor, Single, Single, Single, Int32)

LRN (Local Response Normalization)

Declaration
Tensor IOps.LRN(Tensor X, float alpha, float beta, float bias, int size)
Parameters
Type Name Description
Tensor X
Single alpha

alpha
Single beta

beta
Single bias

bias
Int32 size

size
Returns
Type Description
Tensor

output Tensor
Implements

IOps.MatMul(Tensor, Boolean, Tensor, Boolean)

Matrix multiplication o = x ⨯ y

Declaration
Tensor IOps.MatMul(Tensor X, bool xTranspose, Tensor Y, bool yTranspose)
Parameters
Type Name Description
Tensor X
Boolean xTranspose

transposed x flag
Tensor Y
Boolean yTranspose

transposed y flag
Returns
Type Description
Tensor

output Tensor
Implements

IOps.MatMul(Tensor, Int32, Tensor, Int32)

Multidimensional Matrix multiplication o = x ⨯ y

Declaration
Tensor IOps.MatMul(Tensor X, int rankX, Tensor Y, int rankY)
Parameters
Type Name Description
Tensor X
Int32 rankX

rank of x
Tensor Y
Int32 rankY

rank of y
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Max(Tensor[])

Max

Declaration
Tensor IOps.Max(Tensor[] tensors)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Returns
Type Description
Tensor

output Tensor
Implements

IOps.MaxPool2D(Tensor, Int32[], Int32[], Int32[])

2D max pooling

Declaration
Tensor IOps.MaxPool2D(Tensor X, int[] pool, int[] stride, int[] pad)
Parameters
Type Name Description
Tensor X
Int32[] pool

pooling
Int32[] stride

stride
Int32[] pad

padding
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Mean(Tensor[])

Mean

Declaration
Tensor IOps.Mean(Tensor[] tensors)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Min(Tensor[])

Min

Declaration
Tensor IOps.Min(Tensor[] tensors)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Mul(Tensor[])

Multiply tensors together

Declaration
Tensor IOps.Mul(Tensor[] tensors)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Multinomial(Tensor, Int32, Int32)

Multinomial random distribution

Declaration
Tensor IOps.Multinomial(Tensor X, int count, int seed)
Parameters
Type Name Description
Tensor X
Int32 count

count
Int32 seed

seed
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Neg(Tensor)

Neg

Declaration
Tensor IOps.Neg(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.NonMaxSuppression(Tensor[], Int32, Single, Single, Int32)

Non max suppression tensors[0] - boxes, tensors[1] - scores

Declaration
Tensor IOps.NonMaxSuppression(Tensor[] tensors, int maxOutputBoxesPerClass, float iouThreshold, float scoreThreshold, int centerPointBox)
Parameters
Type Name Description
Tensor[] tensors
Int32 maxOutputBoxesPerClass

max output boxes per class
Single iouThreshold

IOU (Intersection Over Union) threshold
Single scoreThreshold

score threshold
Int32 centerPointBox

center point box
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Normalization(Tensor, Tensor, Tensor, Int32, Int32, Single, Layer.FusedActivation)

Normalization

Declaration
Tensor IOps.Normalization(Tensor X, Tensor S, Tensor B, int pool, int axis, float epsilon, Layer.FusedActivation fusedActivation)
Parameters
Type Name Description
Tensor X
Tensor S
Tensor B
Int32 pool

pooling
Int32 axis

axis
Single epsilon

threshold
Layer.FusedActivation fusedActivation

fused activation type
Returns
Type Description
Tensor

output Tensor
Implements

IOps.OneHot(Tensor, Int32, Single, Single)

One hot

Declaration
Tensor IOps.OneHot(Tensor X, int depth, float onValue, float offValue)
Parameters
Type Name Description
Tensor X
Int32 depth

output depth
Single onValue

on value
Single offValue

off value
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Pad2DEdge(Tensor, Int32[])

Edge padding

Declaration
Tensor IOps.Pad2DEdge(Tensor x, int[] pad)
Parameters
Type Name Description
Tensor x

input
Int32[] pad

padding
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Pad2DReflect(Tensor, Int32[])

Reflection padding

Declaration
Tensor IOps.Pad2DReflect(Tensor x, int[] pad)
Parameters
Type Name Description
Tensor x

input
Int32[] pad

padding
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Pad2DSymmetric(Tensor, Int32[])

Symmetric padding

Declaration
Tensor IOps.Pad2DSymmetric(Tensor x, int[] pad)
Parameters
Type Name Description
Tensor x

input
Int32[] pad

padding
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Pow(Tensor, Single)

Power

Declaration
Tensor IOps.Pow(Tensor X, float alpha)
Parameters
Type Name Description
Tensor X
Single alpha

alpha
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Pow(Tensor[])

Raise tensors to the power o =tensors[0] ^ tensors[1] ^ ... ^ tensors[N-1]

Declaration
Tensor IOps.Pow(Tensor[] tensors)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Returns
Type Description
Tensor

output Tensor
Implements

IOps.PRelu(Tensor, Tensor)

PReLU

Declaration
Tensor IOps.PRelu(Tensor X, Tensor S)
Parameters
Type Name Description
Tensor X
Tensor S
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Prepare(Tensor)

Prepares tensor for use

Declaration
Tensor IOps.Prepare(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

Tensor
Implements

IOps.PrepareNoAlloc(Tensor)

Prepares tensor for use without uploading internal data to device

Declaration
Tensor IOps.PrepareNoAlloc(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

Tensor
Implements

IOps.RandomNormal(TensorShape, Single, Single, Int32)

Normal random distribution

Declaration
Tensor IOps.RandomNormal(TensorShape s, float mean, float scale, int seed)
Parameters
Type Name Description
TensorShape s

shape
Single mean

mean
Single scale

scale
Int32 seed

seed
Returns
Type Description
Tensor

output Tensor
Implements

IOps.RandomUniform(TensorShape, Single, Single, Int32)

Uniform random distribution

Declaration
Tensor IOps.RandomUniform(TensorShape s, float mean, float scale, int seed)
Parameters
Type Name Description
TensorShape s

shape
Single mean

mean
Single scale

scale
Int32 seed

seed
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Reciprocal(Tensor)

Reciprocal (1/x)

Declaration
Tensor IOps.Reciprocal(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ReduceMax(Tensor, Int32)

Reduce with max

Declaration
Tensor IOps.ReduceMax(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ReduceMean(Tensor, Int32)

Reduce with mean

Declaration
Tensor IOps.ReduceMean(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ReduceMin(Tensor, Int32)

Reduce with min

Declaration
Tensor IOps.ReduceMin(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ReduceProd(Tensor, Int32)

Reduce with product

Declaration
Tensor IOps.ReduceProd(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ReduceSum(Tensor, Int32)

Reduce with sum

Declaration
Tensor IOps.ReduceSum(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Relu(Tensor)

ReLU

Declaration
Tensor IOps.Relu(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Relu6(Tensor)

ReLU capped to 6

Declaration
Tensor IOps.Relu6(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Resample2D(Tensor, Int32[], Boolean)

Resample 2D

Declaration
Tensor IOps.Resample2D(Tensor X, int[] size, bool bilinear)
Parameters
Type Name Description
Tensor X
Int32[] size

size
Boolean bilinear

bilinear flag
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ResetAllocator(Boolean)

Reset internal allocator

Declaration
void IOps.ResetAllocator(bool keepCachedMemory)
Parameters
Type Name Description
Boolean keepCachedMemory

keep cached memory flag
Implements

IOps.Reshape(Tensor, TensorShape)

Reshape

Declaration
Tensor IOps.Reshape(Tensor X, TensorShape shape)
Parameters
Type Name Description
Tensor X
TensorShape shape

new shape
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Round(Tensor)

Round to nearest integer. In case of halfs, round to nearest even integer

Declaration
Tensor IOps.Round(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.ScaleBias(Tensor, Tensor, Tensor)

Scale bias o = s * x + b, element wise

Declaration
Tensor IOps.ScaleBias(Tensor X, Tensor S, Tensor B)
Parameters
Type Name Description
Tensor X
Tensor S
Tensor B
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Selu(Tensor, Single, Single)

SELU

Declaration
Tensor IOps.Selu(Tensor X, float alpha, float gamma)
Parameters
Type Name Description
Tensor X
Single alpha

alpha
Single gamma

gamma
Returns
Type Description
Tensor

output Tensor
Implements

IOps.SetModelExecutionsReporter(IModelExecutionsReporter)

Set model executions reporter model executions reporter

Declaration
void IOps.SetModelExecutionsReporter(IModelExecutionsReporter executionsReporter)
Parameters
Type Name Description
IModelExecutionsReporter executionsReporter
Implements

IOps.Shape(Tensor, Int32)

Shape of the input

Declaration
Tensor IOps.Shape(Tensor X, int axis)
Parameters
Type Name Description
Tensor X

input
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Sigmoid(Tensor)

Sigmoid

Declaration
Tensor IOps.Sigmoid(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Sign(Tensor)

Sign

Declaration
Tensor IOps.Sign(Tensor x)
Parameters
Type Name Description
Tensor x

input
Returns
Type Description
Tensor

Tensor with 1 if x > 0 -1 if < 0 and 0 if == 0 values
Implements

IOps.Sin(Tensor)

Sin

Declaration
Tensor IOps.Sin(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Sinh(Tensor)

Sinh

Declaration
Tensor IOps.Sinh(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Softmax(Tensor, Int32)

Softmax

Declaration
Tensor IOps.Softmax(Tensor X, int axis)
Parameters
Type Name Description
Tensor X
Int32 axis

axis
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Softplus(Tensor)

Softplus

Declaration
Tensor IOps.Softplus(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.SpaceToDepth(Tensor, Int32[])

Space to depth

Declaration
Tensor IOps.SpaceToDepth(Tensor X, int[] scale)
Parameters
Type Name Description
Tensor X
Int32[] scale

scale
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Sqrt(Tensor)

Sqrt

Declaration
Tensor IOps.Sqrt(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.StridedSlice(Tensor, Int32[], Int32[], Int32[])

Strided slice

Declaration
Tensor IOps.StridedSlice(Tensor X, int[] starts, int[] ends, int[] strides)
Parameters
Type Name Description
Tensor X
Int32[] starts
Int32[] ends
Int32[] strides
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Sub(Tensor[])

Subtract tensors o = tensors[0] - tensors[1] - ... - tensors[N-1]

Declaration
Tensor IOps.Sub(Tensor[] tensors)
Parameters
Type Name Description
Tensor[] tensors

input tensors
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Swish(Tensor)

Swish

Declaration
Tensor IOps.Swish(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Tan(Tensor)

Tan

Declaration
Tensor IOps.Tan(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Tanh(Tensor)

Tanh

Declaration
Tensor IOps.Tanh(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Tile(Tensor, Int32[])

Tile

Declaration
Tensor IOps.Tile(Tensor X, int[] repeats)
Parameters
Type Name Description
Tensor X
Int32[] repeats

repetition counts
Returns
Type Description
Tensor

output Tensor
Implements

IOps.TopKIndices(Tensor, Int32, Int32, Boolean, Boolean)

Top K indices

Declaration
Tensor IOps.TopKIndices(Tensor X, int k, int axis, bool largest, bool sorted)
Parameters
Type Name Description
Tensor X
Int32 k

k
Int32 axis

axis
Boolean largest

largest flag
Boolean sorted

sorted flag
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Transpose(Tensor)

Transpose matrix

Declaration
Tensor IOps.Transpose(Tensor X)
Parameters
Type Name Description
Tensor X
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Transpose(Tensor, Int32[])

Transpose according to permutations

Declaration
Tensor IOps.Transpose(Tensor X, int[] permutations)
Parameters
Type Name Description
Tensor X
Int32[] permutations

new axis order
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Upsample2D(Tensor, Int32[], Boolean)

Upsample 2D

Declaration
Tensor IOps.Upsample2D(Tensor X, int[] scale, bool bilinear)
Parameters
Type Name Description
Tensor X
Int32[] scale

scale
Boolean bilinear

bilinear flag
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Upsample3D(Tensor, Int32[], Boolean)

Upsample 3D

Declaration
Tensor IOps.Upsample3D(Tensor X, int[] scale, bool trilinear)
Parameters
Type Name Description
Tensor X
Int32[] scale

scale
Boolean trilinear

trilinear flag
Returns
Type Description
Tensor

output Tensor
Implements

IOps.Where(Tensor, Tensor, Tensor)

Where

Declaration
Tensor IOps.Where(Tensor c, Tensor a, Tensor b)
Parameters
Type Name Description
Tensor c

Tensor c
Tensor a

Tensor a
Tensor b

Tensor b
Returns
Type Description
Tensor

Tensor with values c ? a : b
Implements

Did you find this page useful? Please give it a rating: