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
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
|
/*
* Copyright (c) 2017 ARM Limited.
*
* SPDX-License-Identifier: MIT
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to
* deal in the Software without restriction, including without limitation the
* rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
* sell copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in all
* copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#ifndef __ARM_COMPUTE_CLGEMMLOWPOUTPUTSTAGE_H__
#define __ARM_COMPUTE_CLGEMMLOWPOUTPUTSTAGE_H__
#include "arm_compute/runtime/CL/ICLSimpleFunction.h"
/** This file contains all available output stages for GEMMLowp on OpenCL.
*
* In gemmlowp, the "output stage" is the process that takes a final int32 accumulator value (the output of @ref CLGEMMLowpMatrixMultiplyCore),
* and processes it to obtain the final ASYMM8 value.
*
* More information about the GEMMLowp output stage can be found at https://github.com/google/gemmlowp/blob/master/doc/output.md
*/
namespace arm_compute
{
class ITensor;
/** Basic function to execute CLGEMMLowpQuantizeDownInt32ToUint8Scale on OpenCL.
*
* CLGEMMLowpQuantizeDownInt32ToUint8Scale depends on 3 parameters: result_offset, result_mult_int, result_shift
* The final result is:
*
* ((input[i][k] + result_offset) * result_mult_int) >> result_shift
*
* In case the bias tensor is provided, the final result is:
*
* ((input[i][k] + result_offset) * result_mult_int + bias[k]) >> result_shift
*
* This function calls the following OpenCL kernels:
*
* -# @ref CLGEMMLowpQuantizeDownInt32ToUint8ScaleKernel
*
* @note The function accepts also 2 optional input arguments (min and max) which can be used to implement "rectified linear unit" activation functions
* before the result is shifted right by result_shift
*/
class CLGEMMLowpQuantizeDownInt32ToUint8Scale : public ICLSimpleFunction
{
public:
/** Initialise the kernel's inputs, output
*
* @param[in] input Input tensor. It is the output of @ref CLGEMMLowpMatrixMultiplyCore function. Data type supported: S32
* @param[in] bias Biases tensor. Only shared biases supported and it can be a nullptr if the addition of biases is not required.
* Biases are 1D tensor with dimensions [OFM]. Data type supported: Same as @p input.
* @param[out] output Output tensor. Data type supported: Data type supported: QASYMM8
* @param[in] result_offset Offset to be added to each element of the input matrix
* @param[in] result_mult_int Value to be multiplied to each element of the input matrix when once the result_offset has been add
* @param[in] result_shift Number of bits to shift right the result before converting back to QASYMM8
* @param[in] min (Optional) Min value used to saturate down the output result before converting back to QASYMM8
* @param[in] max (Optional) Max value used to saturate up the output result before converting back to QASYMM8,
* Along with @p min, this value can be used to implement "rectified linear unit" activation functions
*/
void configure(const ICLTensor *input, const ICLTensor *bias, ICLTensor *output, int result_offset, int result_mult_int, int result_shift, int min = 0, int max = 0);
};
}
#endif /*__ARM_COMPUTE_CLGEMMLOWPOUTPUTSTAGE_H__ */
|
