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
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
|
/* ----------------------------------------------------------------------
* Project: CMSIS DSP Library
* Title: arm_weighted_sum_f16.c
* Description: Weighted Sum
*
* $Date: 23 April 2021
* $Revision: V1.9.0
*
* Target Processor: Cortex-M and Cortex-A cores
* -------------------------------------------------------------------- */
/*
* Copyright (C) 2010-2021 ARM Limited or its affiliates. All rights reserved.
*
* SPDX-License-Identifier: Apache-2.0
*
* Licensed under the Apache License, Version 2.0 (the License); you may
* not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an AS IS BASIS, WITHOUT
* WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include <limits.h>
#include <math.h>
#include "dsp/support_functions_f16.h"
#if defined(ARM_FLOAT16_SUPPORTED)
/**
@ingroup groupSupport
*/
/**
@defgroup weightedsum Weighted Sum
Weighted sum of values
*/
/**
* @addtogroup weightedsum
* @{
*/
/**
* @brief Weighted sum
*
*
* @param[in] *in Array of input values.
* @param[in] *weigths Weights
* @param[in] blockSize Number of samples in the input array.
* @return Weighted sum
*
*/
#if defined(ARM_MATH_MVE_FLOAT16) && !defined(ARM_MATH_AUTOVECTORIZE)
#include "arm_helium_utils.h"
float16_t arm_weighted_sum_f16(const float16_t *in,const float16_t *weigths, uint32_t blockSize)
{
_Float16 accum1, accum2;
float16x8_t accum1V, accum2V;
float16x8_t inV, wV;
const float16_t *pIn, *pW;
uint32_t blkCnt;
pIn = in;
pW = weigths;
accum1V = vdupq_n_f16(0.0f16);
accum2V = vdupq_n_f16(0.0f16);
blkCnt = blockSize >> 3;
while (blkCnt > 0)
{
inV = vld1q(pIn);
wV = vld1q(pW);
pIn += 4;
pW += 4;
accum1V = vfmaq(accum1V, inV, wV);
accum2V = vaddq(accum2V, wV);
blkCnt--;
}
accum1 = vecAddAcrossF16Mve(accum1V);
accum2 = vecAddAcrossF16Mve(accum2V);
blkCnt = blockSize & 7;
while(blkCnt > 0)
{
accum1 += (_Float16)*pIn++ * (_Float16)*pW;
accum2 += (_Float16)*pW++;
blkCnt--;
}
return (accum1 / accum2);
}
#else
float16_t arm_weighted_sum_f16(const float16_t *in, const float16_t *weigths, uint32_t blockSize)
{
_Float16 accum1, accum2;
const float16_t *pIn, *pW;
uint32_t blkCnt;
pIn = in;
pW = weigths;
accum1=0.0f16;
accum2=0.0f16;
blkCnt = blockSize;
while(blkCnt > 0)
{
accum1 += (_Float16)*pIn++ * (_Float16)*pW;
accum2 += (_Float16)*pW++;
blkCnt--;
}
return(accum1 / accum2);
}
#endif /* defined(ARM_MATH_MVEF) && !defined(ARM_MATH_AUTOVECTORIZE) */
/**
* @} end of weightedsum group
*/
#endif /* #if defined(ARM_FLOAT16_SUPPORTED) */
|
