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
|
/*
* 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.
*/
/* ----------------------------------------------------------------------
* Project: CMSIS NN Library
* Title: arm_fully_connected_s16
* Description: Fully connected function compatible with TF Lite.
*
* $Date: 3. August 2021
* $Revision: V.1.0.0
*
* Target Processor: Cortex-M and Cortex-A cores
*
* -------------------------------------------------------------------- */
#include "arm_nnfunctions.h"
#include "arm_nnsupportfunctions.h"
/**
* @ingroup groupNN
*/
/**
* @addtogroup FC
* @{
*/
/*
* S16 basic fully-connected and matrix multiplication layer function for TensorFlow Lite
*
* Refer header file for details.
*
*/
arm_status arm_fully_connected_s16(const cmsis_nn_context *ctx,
const cmsis_nn_fc_params *fc_params,
const cmsis_nn_per_tensor_quant_params *quant_params,
const cmsis_nn_dims *input_dims,
const q15_t *input,
const cmsis_nn_dims *filter_dims,
const q7_t *kernel,
const cmsis_nn_dims *bias_dims,
const int64_t *bias,
const cmsis_nn_dims *output_dims,
q15_t *output)
{
(void)bias_dims;
(void)ctx;
(void)fc_params->filter_offset;
int32_t batch_cnt = input_dims->n;
const q31_t reduced_multiplier = REDUCE_MULTIPLIER(quant_params->multiplier);
while (batch_cnt)
{
arm_nn_vec_mat_mult_t_s16(input,
kernel,
bias,
output,
reduced_multiplier,
quant_params->shift,
filter_dims->n, /* col_dim or accum_depth */
output_dims->c, /* row_dim or output_depth */
fc_params->activation.min,
fc_params->activation.max);
input += filter_dims->n;
output += output_dims->c;
batch_cnt--;
}
return (ARM_MATH_SUCCESS);
}
int32_t arm_fully_connected_s16_get_buffer_size(const cmsis_nn_dims *filter_dims)
{
(void)filter_dims;
return 0;
}
/**
* @} end of FC group
*/
|
