1 |
|
/** |
2 |
|
* HMLP (High-Performance Machine Learning Primitives) |
3 |
|
* |
4 |
|
* Copyright (C) 2014-2017, The University of Texas at Austin |
5 |
|
* |
6 |
|
* This program is free software: you can redistribute it and/or modify |
7 |
|
* it under the terms of the GNU General Public License as published by |
8 |
|
* the Free Software Foundation, either version 3 of the License, or |
9 |
|
* (at your option) any later version. |
10 |
|
* |
11 |
|
* This program is distributed in the hope that it will be useful, |
12 |
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
13 |
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
14 |
|
* GNU General Public License for more details. |
15 |
|
* |
16 |
|
* You should have received a copy of the GNU General Public License |
17 |
|
* along with this program. If not, see the LICENSE file. |
18 |
|
* |
19 |
|
**/ |
20 |
|
|
21 |
|
|
22 |
|
|
23 |
|
/** CONV2D templates */ |
24 |
|
#include <primitives/conv2d.hpp> |
25 |
|
|
26 |
|
/** Haswell kernels */ |
27 |
|
#include <rank_k_d8x6.hpp> |
28 |
|
|
29 |
|
using namespace hmlp::cnn; |
30 |
|
|
31 |
|
|
32 |
|
|
33 |
|
void conv2d |
34 |
|
( |
35 |
|
int w0, int h0, int d0, int s, int p, int batchSize, |
36 |
|
float *B, int w1, int h1, int d1, |
37 |
|
float *A, |
38 |
|
float *C |
39 |
|
) |
40 |
|
{ |
41 |
|
printf( "not implemented yet\n" ); |
42 |
|
exit( 1 ); |
43 |
|
}; |
44 |
|
|
45 |
|
|
46 |
|
void conv2d |
47 |
|
( |
48 |
|
int w0, int h0, int d0, int s, int p, int batchSize, |
49 |
|
double *B, int w1, int h1, int d1, |
50 |
|
double *A, |
51 |
|
double *C |
52 |
|
) |
53 |
|
{ |
54 |
|
rank_k_asm_d8x6 semiringkernel; |
55 |
|
rank_k_asm_d8x6 microkernel; |
56 |
|
|
57 |
|
conv2d< |
58 |
|
72, 960, 256, 8, 6, |
59 |
|
72, 960, 8, 6, 32, |
60 |
|
false, |
61 |
|
rank_k_asm_d8x6, |
62 |
|
rank_k_asm_d8x6, |
63 |
|
double, double, double, double> |
64 |
|
( |
65 |
|
w0, h0, d0, s, p, batchSize, |
66 |
|
B, |
67 |
|
w1, h1, d1, |
68 |
|
A, |
69 |
|
C, |
70 |
|
semiringkernel, |
71 |
|
microkernel |
72 |
|
); |
73 |
|
}; |
74 |
|
|
75 |
|
|
76 |
|
void dconv2d |
77 |
|
( |
78 |
|
int w0, int h0, int d0, int s, int p, int batchSize, |
79 |
|
float *B, int w1, int h1, int d1, |
80 |
|
float *A, |
81 |
|
float *C |
82 |
|
) |
83 |
|
{ |
84 |
|
conv2d( w0, h0, d0, s, p, batchSize, B, w1, h1, d1, A, C ); |
85 |
|
}; |
86 |
|
|
87 |
|
|
88 |
|
void dconv2d |
89 |
|
( |
90 |
|
int w0, int h0, int d0, int s, int p, int batchSize, |
91 |
|
double *B, int w1, int h1, int d1, |
92 |
|
double *A, |
93 |
|
double *C |
94 |
|
) |
95 |
|
{ |
96 |
|
conv2d( w0, h0, d0, s, p, batchSize, B, w1, h1, d1, A, C ); |
97 |
|
}; |
98 |
|
|
99 |
|
|
100 |
|
|
101 |
|
|
102 |
|
|
103 |
|
void dconv2d_ref |
104 |
|
( |
105 |
|
int w0, int h0, int d0, int s, int p, int batchSize, |
106 |
|
double *B, int w1, int h1, int d1, |
107 |
|
double *A, |
108 |
|
double *C |
109 |
|
) |
110 |
|
{ |
111 |
|
conv2d_ref |
112 |
|
( |
113 |
|
w0, h0, d0, s, p, batchSize, |
114 |
|
B, |
115 |
|
w1, h1, d1, |
116 |
|
A, |
117 |
|
C |
118 |
|
); |
119 |
|
}; |
120 |
|
|
121 |
|
|
122 |
|
|