blob: af03e7960d1d291f1a6e07105c86a41a07e0ddf2 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
|
//
// Copyright © 2017 Arm Ltd. All rights reserved.
// SPDX-License-Identifier: MIT
//
#include "NeonPreluWorkload.hpp"
#include "NeonWorkloadUtils.hpp"
#include <aclCommon/ArmComputeUtils.hpp>
#include <armnn/utility/PolymorphicDowncast.hpp>
#include <arm_compute/runtime/NEON/functions/NEPReluLayer.h>
namespace armnn
{
arm_compute::Status NeonPreluWorkloadValidate(const TensorInfo& input,
const TensorInfo& alpha,
const TensorInfo& output)
{
const arm_compute::TensorInfo aclInput = armcomputetensorutils::BuildArmComputeTensorInfo(input);
const arm_compute::TensorInfo aclAlpha = armcomputetensorutils::BuildArmComputeTensorInfo(alpha);
const arm_compute::TensorInfo aclOutput = armcomputetensorutils::BuildArmComputeTensorInfo(output);
return arm_compute::NEPReluLayer::validate(&aclInput,
&aclAlpha,
&aclOutput);
}
NeonPreluWorkload::NeonPreluWorkload(const PreluQueueDescriptor& descriptor,
const WorkloadInfo& info)
: BaseWorkload<PreluQueueDescriptor>(descriptor, info)
{
m_Data.ValidateInputsOutputs("NeonPreluWorkload", 1, 1);
arm_compute::ITensor& input = PolymorphicDowncast<IAclTensorHandle*>(m_Data.m_Inputs[0])->GetTensor();
arm_compute::ITensor& alpha = PolymorphicDowncast<IAclTensorHandle*>(m_Data.m_Inputs[1])->GetTensor();
arm_compute::ITensor& output = PolymorphicDowncast<IAclTensorHandle*>(m_Data.m_Outputs[0])->GetTensor();
auto layer = std::make_unique<arm_compute::NEPReluLayer>();
layer->configure(&input, &alpha, &output);
m_PreluLayer.reset(layer.release());
}
void NeonPreluWorkload::Execute() const
{
ARMNN_SCOPED_PROFILING_EVENT_NEON_GUID("NeonPreluWorkload_Execute", this->GetGuid());
m_PreluLayer->run();
}
} //namespace armnn
|
