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
|
/*
* Copyright (c) 2019 Samsung Electronics Co., Ltd. All Rights Reserved
* Copyright (c) 2016-2018 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_NEELEMENTWISEOPERATIONFUNCS_H__
#define __ARM_COMPUTE_NEELEMENTWISEOPERATIONFUNCS_H__
#include <arm_neon.h>
namespace arm_compute
{
class ITensor;
class Window;
class QuantizationInfo;
} // namespace arm_compute
namespace arm_compute
{
float32x4x4_t load_quantized(const uint8_t *input1_ptr, const int32x4_t &offset,
const float32x4_t &scale);
void store_quantized(uint8_t *output_ptr, const float32x4x4_t &rf, const float32x4_t &offset,
const float32x4_t &invscale);
float32x4x4_t dup_quantized(uint8_t broadcast_value, int offset, float scale);
void elementwise_op_quantized(
const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window,
uint8_t (*scalar_func)(const float &, const float &, QuantizationInfo),
int (*broadcast_func)(int, int, int, const uint8_t *, float32x4x4_t, uint8_t *, int32x4_t,
float32x4_t, float32x4_t, float32x4_t, const bool),
int (*neon_func)(int, int, int, const uint8_t *, const uint8_t *, uint8_t *, int32x4_t,
int32x4_t, float32x4_t, float32x4_t, float32x4_t, float32x4_t));
void elementwise_op(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window,
float (*scalar_func)(const float &, const float &),
int (*broadcast_func)(int, int, int, const float *, const float &, float *,
const bool),
int (*neon_func)(int, int, int, const float *, const float *, float *));
void elementwise_op(const ITensor *in1, const ITensor *in2, ITensor *out, const Window &window,
uint8_t (*scalar_func)(const uint8_t &, const uint8_t &),
int (*broadcast_func)(int, int, int, const uint8_t *, const uint8_t &,
uint8_t *, const bool),
int (*neon_func)(int, int, int, const uint8_t *, const uint8_t *, uint8_t *));
} // namespace arm_compute
#endif // __ARM_COMPUTE_NEELEMENTWISEOPERATIONFUNCS_H__
|
