org.bytedeco.tvm

Class DLTensor

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.tvm.DLTensor

All Implemented Interfaces:

AutoCloseable

@Properties(inherit=tvm_runtime.class)
public class DLTensor
extends Pointer

\brief Plain C Tensor object, does not manage memory.

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
DLTensor() Default native constructor.
DLTensor(long size) Native array allocator.
DLTensor(Pointer p) Pointer cast constructor.

Method Summary

Modifier and Type	Method and Description
long	byte_offset() \brief The offset in bytes to the beginning pointer to data
DLTensor	byte_offset(long setter)
Pointer	data() \brief The data pointer points to the allocated data.
DLTensor	data(Pointer setter)
DLDevice	device() \brief The device of the tensor
DLTensor	device(DLDevice setter)
DLDataType	dtype() \brief The data type of the pointer
DLTensor	dtype(DLDataType setter)
DLTensor	getPointer(long i)
int	ndim() \brief Number of dimensions
DLTensor	ndim(int setter)
DLTensor	position(long position)
LongPointer	shape() \brief The shape of the tensor
DLTensor	shape(LongPointer setter)
LongPointer	strides() \brief strides of the tensor (in number of elements, not bytes) can be NULL, indicating tensor is compact and row-majored.
DLTensor	strides(LongPointer 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, getDirectBufferAddress, getPointer, getPointer, getPointer, hashCode, interruptDeallocatorThread, isNull, isNull, limit, limit, malloc, maxBytes, maxPhysicalBytes, memchr, memcmp, memcpy, memmove, memset, offsetAddress, offsetof, offsetof, parseBytes, physicalBytes, physicalBytesInaccurate, position, put, realloc, referenceCount, releaseReference, retainReference, setNull, sizeof, sizeof, toString, totalBytes, totalCount, totalPhysicalBytes, withDeallocator, zero

Methods inherited from class java.lang.Object

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

Constructor Detail

DLTensor

public DLTensor()

Default native constructor.

DLTensor

public DLTensor(long size)

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

DLTensor

public DLTensor(Pointer p)

Pointer cast constructor. Invokes Pointer(Pointer).

Method Detail

position

public DLTensor position(long position)

Overrides:

position in class Pointer

getPointer

public DLTensor getPointer(long i)

Overrides:

getPointer in class Pointer

data

public Pointer data()

\brief The data pointer points to the allocated data. This will be CUDA device pointer or cl_mem handle in OpenCL. It may be opaque on some device types. This pointer is always aligned to 256 bytes as in CUDA. The byte_offset field should be used to point to the beginning of the data. Note that as of Nov 2021, multiply libraries (CuPy, PyTorch, TensorFlow, TVM, perhaps others) do not adhere to this 256 byte aligment requirement on CPU/CUDA/ROCm, and always use byte_offset=0. This must be fixed (after which this note will be updated); at the moment it is recommended to not rely on the data pointer being correctly aligned. For given DLTensor, the size of memory required to store the contents of data is calculated as follows:

 {.c}
 static inline size_t GetDataSize(const DLTensor* t) {
   size_t size = 1;
   for (tvm_index_t i = 0; i < t->ndim; ++i) {
     size *= t->shape[i];
   }
   size *= (t->dtype.bits * t->dtype.lanes + 7) / 8;
   return size;
 }
 

data

public DLTensor data(Pointer setter)

device

@ByRef
public DLDevice device()

\brief The device of the tensor

device

public DLTensor device(DLDevice setter)

ndim

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

\brief Number of dimensions

ndim

public DLTensor ndim(int setter)

dtype

@ByRef
public DLDataType dtype()

\brief The data type of the pointer

dtype

public DLTensor dtype(DLDataType setter)

shape

@Cast(value="int64_t*")
public LongPointer shape()

\brief The shape of the tensor

shape

public DLTensor shape(LongPointer setter)

strides

@Cast(value="int64_t*")
public LongPointer strides()

\brief strides of the tensor (in number of elements, not bytes) can be NULL, indicating tensor is compact and row-majored.

strides

public DLTensor strides(LongPointer setter)

byte_offset

@Cast(value="uint64_t")
public long byte_offset()

\brief The offset in bytes to the beginning pointer to data

byte_offset

public DLTensor byte_offset(long setter)