org.bytedeco.ngraph

Class CPU_Backend

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.ngraph.Backend

org.bytedeco.ngraph.CPU_Backend

All Implemented Interfaces:

AutoCloseable

@Namespace(value="ngraph::runtime::cpu")
 @NoOffset
 @Properties(inherit=ngraph.class)
public class CPU_Backend
extends Backend

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.ngraph.Backend

memory_attach

Fields inherited from class org.bytedeco.javacpp.Pointer

address, capacity, limit, position

Constructor Summary

Constructors

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

Method Summary

Modifier and Type	Method and Description
Executable	compile(Function func)
Executable	compile(Function func, boolean enable_performance_counters) \brief Compiles a Function.
Executable	compile(Function func, PassConfig pass_config)
Executable	compile(Function func, PassConfig pass_config, boolean enable_performance_counters) \brief Compiles a Function.
Tensor	create_tensor(Type element_type, Shape shape) \brief Create a tensor specific to this backend
Tensor	create_tensor(Type element_type, Shape shape, Pointer memory_pointer) \brief Create a tensor specific to this backend
Allocator	get_host_memory_allocator() \brief Returns memory allocator used by backend for host allocations
CPU_Backend	getPointer(long i)
boolean	is_supported_property(int prop) \brief Test if a backend particular property is supported
boolean	is_supported(Node node) \brief Test if a backend is capable of supporting an op
CPU_Backend	position(long position)
void	remove_compiled_function(Executable exec)
void	set_host_memory_allocator(Allocator allocator) \brief Set the host memory allocator to be used by the backend

Methods inherited from class org.bytedeco.ngraph.Backend

create_dynamic_tensor, create, create, create, create, executable_can_create_tensors, get_backend_op, get_backend_op, get_backend_shared_library_search_directory, get_device_memory_allocator, get_registered_devices, get_version, is_device_memory, load, set_backend_shared_library_search_directory, set_backend_shared_library_search_directory, set_config, supports_dynamic_tensors

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

CPU_Backend

public CPU_Backend()

Default native constructor.

CPU_Backend

public CPU_Backend(long size)

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

CPU_Backend

public CPU_Backend(Pointer p)

Pointer cast constructor. Invokes Pointer(Pointer).

Method Detail

position

public CPU_Backend position(long position)

Overrides:

position in class Pointer

getPointer

public CPU_Backend getPointer(long i)

Overrides:

getPointer in class Pointer

create_tensor

@SharedPtr
 @ByVal
public Tensor create_tensor(@Const @ByRef
                                               Type element_type,
                                               @Const @ByRef
                                               Shape shape,
                                               Pointer memory_pointer)

Description copied from class: Backend

\brief Create a tensor specific to this backend

Overrides:

create_tensor in class Backend

Parameters:

element_type - The type of the tensor element

shape - The shape of the tensor

memory_pointer - A pointer to a buffer used for this tensor. The size of the buffer must be sufficient to contain the tensor. The lifetime of the buffer is the responsibility of the caller.

Returns:

shared_ptr to a new backend-specific tensor

create_tensor

@SharedPtr
 @ByVal
public Tensor create_tensor(@Const @ByRef
                                               Type element_type,
                                               @Const @ByRef
                                               Shape shape)

Description copied from class: Backend

\brief Create a tensor specific to this backend

Overrides:

create_tensor in class Backend

Parameters:

element_type - The type of the tensor element

shape - The shape of the tensor

Returns:

shared_ptr to a new backend-specific tensor

compile

@SharedPtr
public Executable compile(@SharedPtr @ByVal
                                     Function func,
                                     @Cast(value="bool")
                                     boolean enable_performance_counters)

Description copied from class: Backend

\brief Compiles a Function.

Overrides:

compile in class Backend

Parameters:

func - The function to compile

Returns:

compiled function or nullptr on failure

compile

@SharedPtr
public Executable compile(@SharedPtr @ByVal
                                     Function func)

Overrides:

compile in class Backend

compile

@SharedPtr
public Executable compile(@SharedPtr @ByVal
                                     Function func,
                                     @ByRef
                                     PassConfig pass_config,
                                     @Cast(value="bool")
                                     boolean enable_performance_counters)

Description copied from class: Backend

\brief Compiles a Function.

Overrides:

compile in class Backend

Parameters:

func - The function to compile

pass_config - Configuration object for defining compilation options

Returns:

compiled function or nullptr on failure

compile

@SharedPtr
public Executable compile(@SharedPtr @ByVal
                                     Function func,
                                     @ByRef
                                     PassConfig pass_config)

Overrides:

compile in class Backend

remove_compiled_function

public void remove_compiled_function(@SharedPtr
                                     Executable exec)

Overrides:

remove_compiled_function in class Backend

get_host_memory_allocator

public Allocator get_host_memory_allocator()

Description copied from class: Backend

\brief Returns memory allocator used by backend for host allocations

Overrides:

get_host_memory_allocator in class Backend

set_host_memory_allocator

public void set_host_memory_allocator(Allocator allocator)

Description copied from class: Backend

\brief Set the host memory allocator to be used by the backend

Overrides:

set_host_memory_allocator in class Backend

Parameters:

allocator - is pointer to host memory allocator object

is_supported

@Cast(value="bool")
public boolean is_supported(@Const @ByRef
                                                Node node)

Description copied from class: Backend

\brief Test if a backend is capable of supporting an op

Overrides:

is_supported in class Backend

Parameters:

node - is the op to test.

Returns:

true if the op is supported, false otherwise.

is_supported_property

@Cast(value="bool")
public boolean is_supported_property(@Cast(value="const ngraph::runtime::cpu::CPU_Backend::Property")
                                                         int prop)

Description copied from class: Backend

\brief Test if a backend particular property is supported

Overrides:

is_supported_property in class Backend

Parameters:

prop - is the feature to test.

Returns:

true if the property is supported, false otherwise.