org.bytedeco.onnx

Class onnxTensorDescriptorV1

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.onnx.onnxTensorDescriptorV1

All Implemented Interfaces:

AutoCloseable

@Properties(inherit=onnx.class)
public class onnxTensorDescriptorV1
extends Pointer

Nested Class Summary

Nested classes/interfaces inherited from class org.bytedeco.javacpp.Pointer

Pointer.CustomDeallocator, Pointer.Deallocator, Pointer.NativeDeallocator, Pointer.ReferenceCounter

Field Summary

Fields inherited from class org.bytedeco.javacpp.Pointer

address, capacity, limit, position

Constructor Summary

Constructors

Constructor and Description
onnxTensorDescriptorV1() Default native constructor.
onnxTensorDescriptorV1(long size) Native array allocator.
onnxTensorDescriptorV1(Pointer p) Pointer cast constructor.

Method Summary

Modifier and Type	Method and Description
int	buffer() Pointers to tensor data.
onnxTensorDescriptorV1	buffer(int setter)
int	dataType() Data type of the elements in the tensor.
onnxTensorDescriptorV1	dataType(int setter)
int	dimensions() Number of dimensions in the tensor.
onnxTensorDescriptorV1	dimensions(int setter)
int	memoryType() Type of memory that stores the tensor.
onnxTensorDescriptorV1	memoryType(int setter)
BytePointer	name() Name of the blob corresponding to this tensor in the ONNX model.
onnxTensorDescriptorV1	name(BytePointer setter)
onnxTensorDescriptorV1	position(long position)
IntPointer	shape() Dimensions of the tensor.
onnxTensorDescriptorV1	shape(IntPointer setter)
int	tag() 32-bit tag needed to distinguish different versions of a tensor descriptor structure.
onnxTensorDescriptorV1	tag(int setter)

Methods inherited from class org.bytedeco.javacpp.Pointer

address, asBuffer, asByteBuffer, availablePhysicalBytes, calloc, capacity, capacity, close, deallocate, deallocate, deallocateReferences, deallocator, deallocator, equals, fill, formatBytes, free, hashCode, isNull, isNull, limit, limit, malloc, maxBytes, maxPhysicalBytes, memchr, memcmp, memcpy, memmove, memset, offsetof, parseBytes, physicalBytes, position, put, realloc, referenceCount, releaseReference, retainReference, setNull, sizeof, toString, totalBytes, totalPhysicalBytes, withDeallocator, zero

Methods inherited from class java.lang.Object

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

Constructor Detail

onnxTensorDescriptorV1

public onnxTensorDescriptorV1()

Default native constructor.

onnxTensorDescriptorV1

public onnxTensorDescriptorV1(long size)

Native array allocator. Access with Pointer.position(long).

onnxTensorDescriptorV1

public onnxTensorDescriptorV1(Pointer p)

Pointer cast constructor. Invokes Pointer.Pointer(Pointer).

Method Detail

position

public onnxTensorDescriptorV1 position(long position)

Overrides:

position in class Pointer

tag

@Cast(value="int32_t")
public int tag()

32-bit tag needed to distinguish different versions of a tensor descriptor structure. In the onnxTensorDescriptorV1 structure, the tag MUST be set to ONNXIFI_TAG_TENSOR_DESCRIPTOR_V1. If ONNXIFI introduce a new version of the tensor descriptor structure in the future, it WILL have 32-bit tag with a different value as the first member of the structure. ONNXIFI implementations MUST validate tag before accessing any other members of the structure.

tag

public onnxTensorDescriptorV1 tag(int setter)

name

@Cast(value="const char*")
public BytePointer name()

Name of the blob corresponding to this tensor in the ONNX model. The name must exactly match the ValueInfoProto.name of one of the ModelProto.graph.input or ModelProto.graph.output

name

public onnxTensorDescriptorV1 name(BytePointer setter)

dataType

@Cast(value="onnxEnum")
public int dataType()

Data type of the elements in the tensor. Possible values: ONNXIFI_DATATYPE_FLOAT16 ONNXIFI_DATATYPE_FLOAT32 ONNXIFI_DATATYPE_FLOAT64 ONNXIFI_DATATYPE_INT8 ONNXIFI_DATATYPE_INT16 ONNXIFI_DATATYPE_INT32 ONNXIFI_DATATYPE_INT64 ONNXIFI_DATATYPE_UINT8 ONNXIFI_DATATYPE_UINT16 ONNXIFI_DATATYPE_UINT32 ONNXIFI_DATATYPE_UINT64 ONNXIFI_DATATYPE_COMPLEX64 ONNXIFI_DATATYPE_COMPLEX128

dataType

public onnxTensorDescriptorV1 dataType(int setter)

memoryType

@Cast(value="onnxEnum")
public int memoryType()

Type of memory that stores the tensor. ONNXIFI_MEMORY_TYPE_CPU memory type is always supported by the backend, but other memory types are optional. The use MUST call onnxGetBackendInfo with ONNXIFI_BACKEND_MEMORY_TYPES to check if a particular memory type is supported before using it. If the memory type is different than ONNXIFI_MEMORY_TYPE_CPU, it must be allocated on the same device as the backend. Possible values: ONNXIFI_MEMORY_TYPE_CPU (always supported) ONNXIFI_MEMORY_TYPE_CUDA_BUFFER (support is optional) ONNXIFI_MEMORY_TYPE_OPENCL_BUFFER (support is optional) ONNXIFI_MEMORY_TYPE_OPENGLES_TEXTURE_2D (support is optional) ONNXIFI_MEMORY_TYPE_D3D_RESOURCE (support is optional)

memoryType

public onnxTensorDescriptorV1 memoryType(int setter)

dimensions

@Cast(value="uint32_t")
public int dimensions()

Number of dimensions in the tensor. For a scalar, the number of dimensions is 0.

dimensions

public onnxTensorDescriptorV1 dimensions(int setter)

shape

@Cast(value="const uint64_t*")
public IntPointer shape()

Dimensions of the tensor. For a scalar, this pointer can be NULL.

shape

public onnxTensorDescriptorV1 shape(IntPointer setter)

buffer

@Cast(value="onnxPointer")
public int buffer()

Pointers to tensor data. Interpretation depends on memoryType: - ONNXIFI_MEMORY_TYPE_CPU: buffer is a valid pointer to CPU memory. - ONNXIFI_MEMORY_TYPE_CUDA_BUFFER: buffer is a valid pointer to CUDA device memory, allocated via cudaMalloc or cuMalloc. CUDA device memory must be allocated on the same device as the backend. - ONNXIFI_MEMORY_TYPE_OPENCL_BUFFER: buffer is a cl_mem handle for an OpenCL buffer or a sub-buffer. cl_mem object must be allocated on the same OpenCL context as the backend. The context must be specified via ONNXIFI_BACKEND_OPENCL_CONTEXT initialization property to onnxInitBackend. - ONNXIFI_MEMORY_TYPE_OPENGLES_TEXTURE_2D: buffer is a name of a 2D texture created by glGenTextures. The texture must be allocated on the same device as the backend. - ONNXIFI_MEMORY_TYPE_D3D_RESOURCE: TBD

buffer

public onnxTensorDescriptorV1 buffer(int setter)