Class TextureTensorData

Represents the data storage for a 'Tensor' as a render texture, for backends that use GPU pixel shaders.

Sentis packs the tensor data into the pixels of an RGBA float4 texture.

Sentis chooses a single tensor dimension as the blocked axis, across which data is chunked in float4 blocks.

Tensor dimensions don't map directly to texture dimensions. Sentis creates the texture with dimensions large enough to fit all the data and pixel shaders index the data based on both the tensor and texture dimensions (see example below).

Inheritance

object

TextureTensorData

Implements

ITensorData

IDisposable

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.ReferenceEquals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

Namespace: Unity.Sentis

Assembly: solution.dll

Syntax

public class TextureTensorData : ITensorData, IDisposable

Constructors

Name	Description
TextureTensorData(DataType, TensorShape, int, bool)	Initializes and returns an instance of TextureTensorData with given shape and blocked axis. A RenderTexture is allocated to the correct size.

Properties

Name	Description
blockAxis	Returns the axis of the tensor which is blocked. It is possible to block on negative axes by considering a tensor of shape (d0, d1 ... dn) as one of shape (1, 1, .... 1, d0, d1 ... dn). Thus negative axis values do not count from the back of the shape as elsewhere.
blockedShape	Returns the shape of the tensor with the blocked axis divided by 4.
bufferAsTexture	Returns the backing texture storing the tensor data.
dataType	Returns the data type of the associated tensor.
deviceType	On what device backend are the data elements stored.
dimAxis	The size of the blocked axis in the original tensor shape (when not blocked).
dimAxisDiv4	The size of the blocked axis in the blocked tensor shape, i.e. dimAxisDiv4 = ceil(dimAxis / 4).
maxCapacity	The maximum count of the stored data elements.
shape	Returns the shape of the associated tensor.
strideAxis	The size of the stride of the blocked axis.
widthMask	Returns the width of the texture - 1 for efficient masking in shaders.
widthShift	Returns the power in the power of two width of the backing texture.

Methods

Name	Description
Clone()	Returns a deep copy of the internal storage.
CompleteAllPendingOperations()	Blocking call to make sure that internal data is correctly written to and available for CPU read back.
Dispose()	Disposes of the TextureTensorData and any associated memory.
Download<T>(int, int)	Returns a contiguous block of data from internal storage.
~TextureTensorData()	Finalizes the TextureTensorData.
IsReadbackRequestDone()	Checks if asynchronous readback request is done.
Pin(Tensor, int, bool)	Moves the tensor into GPU memory on the GPUPixel back end device.
ReadbackRequest(Action<bool>)	Schedules asynchronous readback of the internal data.
ReadbackRequestAsync()	Schedules awaitable asynchronous readback of the internal data. See AsyncReadbackRequest for more info
Reserve(int)	Reserves memory for count elements.
ToString()	Returns a string that represents the TextureTensorData.
Upload<T>(NativeArray<T>, int, int)	Uploads a contiguous block of tensor data to internal storage.