NEPooling3dLayer Class Reference

Basic function to simulate a pooling 3d layer with the specified pooling operation. More...

#include <NEPooling3dLayer.h>

Collaboration diagram for NEPooling3dLayer:

Public Member Functions
	NEPooling3dLayer (std::shared_ptr< IMemoryManager > memory_manager=nullptr)
	Constructor. More...
	NEPooling3dLayer (const NEPooling3dLayer &)=delete
	Prevent instances of this class from being copied (As this class contains pointers) More...
NEPooling3dLayer &	operator= (const NEPooling3dLayer &)=delete
	Prevent instances of this class from being copied (As this class contains pointers) More...
	NEPooling3dLayer (NEPooling3dLayer &&)=delete
	Prevent instances of this class from being moved (As this class contains non movable objects) More...
NEPooling3dLayer &	operator= (NEPooling3dLayer &&)=delete
	Prevent instances of this class from being moved (As this class contains non movable objects) More...
	~NEPooling3dLayer ()
	Default destructor. More...
void	configure (const ITensor *input, ITensor *output, const Pooling3dLayerInfo &pool_info)
	Set the input and output tensors. More...
void	run () override
	Run the kernels contained in the function. More...
Public Member Functions inherited from IFunction
virtual	~IFunction ()=default
	Destructor. More...
virtual void	prepare ()
	Prepare the function for executing. More...

Static Public Member Functions
static Status	validate (const ITensorInfo *input, const ITensorInfo *output, const Pooling3dLayerInfo &pool_info)
	Static function to check if given info will lead to a valid configuration of NEPooling3dLayer. More...

Detailed Description

Basic function to simulate a pooling 3d layer with the specified pooling operation.

This function calls the following kernels:

1. cpu::CpuPool3d

Definition at line 42 of file NEPooling3dLayer.h.

Constructor & Destructor Documentation

NEPooling3dLayer() [1/3]

NEPooling3dLayer

(

std::shared_ptr< IMemoryManager >

memory_manager = nullptr

)

Constructor.

Definition at line 47 of file NEPooling3dLayer.cpp.

                                                                               : _impl(std::make_unique<Impl>())
{
    _impl->memory_group = MemoryGroup(std::move(memory_manager));
}

NEPooling3dLayer() [2/3]

NEPooling3dLayer ( const NEPooling3dLayer &  )

delete

Prevent instances of this class from being copied (As this class contains pointers)

NEPooling3dLayer() [3/3]

NEPooling3dLayer ( NEPooling3dLayer &&  )

delete

Prevent instances of this class from being moved (As this class contains non movable objects)

~NEPooling3dLayer()

~NEPooling3dLayer ( )

default

Default destructor.

Member Function Documentation

configure()

void configure	(	const ITensor *	input,
		ITensor *	output,
		const Pooling3dLayerInfo &	pool_info
	)

Set the input and output tensors.

Valid data layouts:

• NDHWC

Valid data type configurations:

src	dst
F16	F16
F32	F32
QASYMM8	QASYMM8
QASYMM8_SIGNED	QASYMM8_SIGNED

Note

Source tensor is padded with -inf for MAX pooling and 0 otherwise

Parameters

[in]	input	Source tensor. Data types supported: F16/F32/QASYMM8/QASYMM8_SIGNED.
[out]	output	Destination tensor.
[in]	pool_info	Contains pooling operation information described in Pooling3dLayerInfo.

Definition at line 52 of file NEPooling3dLayer.cpp.

{
    _impl->src = input;
    _impl->dst = output;
    _impl->op  = std::make_unique<cpu::CpuPool3d>();
    _impl->op->configure(input->info(), output->info(), pool_info);
 
    _impl->run_pack          = {{TensorType::ACL_SRC, _impl->src}, {TensorType::ACL_DST_0, _impl->dst}};
    _impl->workspace_tensors = manage_workspace<Tensor>(_impl->op->workspace(), _impl->memory_group, _impl->run_pack);
}

References arm_compute::ACL_DST_0, arm_compute::ACL_SRC, ITensor::info(), and arm_compute::test::validation::input.

operator=() [1/2]

NEPooling3dLayer& operator= ( const NEPooling3dLayer &  )

delete

Prevent instances of this class from being copied (As this class contains pointers)

operator=() [2/2]

NEPooling3dLayer& operator= ( NEPooling3dLayer &&  )

delete

Prevent instances of this class from being moved (As this class contains non movable objects)

run()

void run ( )

overridevirtual

Run the kernels contained in the function.

For CPU kernels:

• Multi-threading is used for the kernels which are parallelisable.
• By default std::thread::hardware_concurrency() threads are used.

Note

CPPScheduler::set_num_threads() can be used to manually set the number of threads

For OpenCL kernels:

• All the kernels are enqueued on the queue associated with CLScheduler.
• The queue is then flushed.

Note

The function will not block until the kernels are executed. It is the user's responsibility to wait.

Will call prepare() on first run if hasn't been done

Implements IFunction.

Definition at line 69 of file NEPooling3dLayer.cpp.

{
    MemoryGroupResourceScope scope_mg(_impl->memory_group);
    ARM_COMPUTE_ERROR_ON_NULLPTR(_impl->src, _impl->dst);
    _impl->op->run(_impl->run_pack);
}

References ARM_COMPUTE_ERROR_ON_NULLPTR.

validate()

Status validate ( const ITensorInfo *  input, const ITensorInfo *  output, const Pooling3dLayerInfo &  pool_info  )

static

Static function to check if given info will lead to a valid configuration of NEPooling3dLayer.

Parameters

[in]	input	Source tensor info. Data types supported: F16/F32/QASYMM8/QASYMM8_SIGNED.
[in]	output	Destination tensor info.
[in]	pool_info	Contains pooling operation information described in Pooling3dLayerInfo.

Returns

a status

Definition at line 64 of file NEPooling3dLayer.cpp.

{
    return cpu::CpuPool3d::validate(input, output, pool_info);
}

References arm_compute::test::validation::input, and CpuPool3d::validate().

---

The documentation for this class was generated from the following files:

• arm_compute/runtime/NEON/functions/NEPooling3dLayer.h
• src/runtime/NEON/functions/NEPooling3dLayer.cpp