ncnn

1
// Tencent is pleased to support the open source community by making ncnn available.
2
//
3
// Copyright (C) 2021 THL A29 Limited, a Tencent company. All rights reserved.
4
//
5
// Licensed under the BSD 3-Clause License (the "License"); you may not use this file except
6
// in compliance with the License. You may obtain a copy of the License at
7
//
8
// https://opensource.org/licenses/BSD-3-Clause
9
//
10
// Unless required by applicable law or agreed to in writing, software distributed
11
// under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR
12
// CONDITIONS OF ANY KIND, either express or implied. See the License for the
13
// specific language governing permissions and limitations under the License.
14

15
#include "testutil.h"
16

17
static int test_convolution3d(int w, int h, int d, int c, int outch, int kernel, int dilation, int stride, int pad, int bias)
18
{
19
    ncnn::Mat a = RandomMat(w, h, d, c);
20

21
    ncnn::ParamDict pd;
22
    pd.set(0, outch);    // num_output
23
    pd.set(1, kernel);   // kernel_w
24
    pd.set(2, dilation); // dilation_w
25
    pd.set(3, stride);   // stride_w
26
    pd.set(4, pad);      // pad_w
27
    pd.set(5, bias);     // bias_term
28
    pd.set(6, outch * c * kernel * kernel * kernel);
29

30
    int activation_type = RAND() % 6; // 0 1 2 3 4 5
31
    ncnn::Mat activation_params(2);
32
    activation_params[0] = RandomFloat(-1, 0); // alpha
33
    activation_params[1] = RandomFloat(0, 1);  // beta
34
    pd.set(9, activation_type);
35
    pd.set(10, activation_params);
36

37
    std::vector<ncnn::Mat> weights(bias ? 2 : 1);
38
    weights[0] = RandomMat(outch * c * kernel * kernel * kernel);
39
    if (bias)
40
        weights[1] = RandomMat(outch);
41

42
    int ret = test_layer("Convolution3D", pd, weights, a);
43
    if (ret != 0)
44
    {
45
        fprintf(stderr, "test_convolution3d failed w=%d h=%d d=%d c=%d outch=%d kernel=%d dilation=%d stride=%d pad=%d bias=%d act=%d actparams=[%f,%f]\n", w, h, d, c, outch, kernel, dilation, stride, pad, bias, activation_type, activation_params[0], activation_params[1]);
46
    }
47

48
    return ret;
49
}
50

51
static int test_convolution3d_0()
52
{
53
    static const int kdsp[7][4] = {
54
        {1, 1, 1, 0},
55
        {1, 1, 2, 0},
56
        {2, 1, 1, 1},
57
        {2, 1, 2, -233},
58
        {3, 1, 1, 1},
59
        {3, 1, 2, 1},
60
        {3, 2, 1, -234},
61
    };
62

63
    for (int i = 0; i < 7; i++)
64
    {
65
        const int k = kdsp[i][0];
66
        const int d = kdsp[i][1];
67
        const int s = kdsp[i][2];
68
        const int p = kdsp[i][3];
69

70
        int ret = 0
71
                  || test_convolution3d(11, 10, 9, 1, 1, k, d, s, p, 1)
72
                  || test_convolution3d(11, 10, 9, 4, 13, k, d, s, p, 0)
73
                  || test_convolution3d(11, 10, 9, 13, 4, k, d, s, p, 1)
74
                  || test_convolution3d(11, 10, 9, 12, 12, k, d, s, p, 0)
75
                  || test_convolution3d(11, 10, 9, 8, 12, k, d, s, p, 1)
76
                  || test_convolution3d(11, 10, 9, 8, 13, k, d, s, p, 0)
77
                  || test_convolution3d(11, 10, 9, 13, 8, k, d, s, p, 1)
78
                  || test_convolution3d(11, 10, 9, 12, 16, k, d, s, p, 0)
79
                  || test_convolution3d(11, 10, 9, 15, 15, k, d, s, p, 0)
80
                  || test_convolution3d(11, 10, 9, 16, 16, k, d, s, p, 0);
81

82
        if (ret != 0)
83
            return -1;
84
    }
85

86
    return 0;
87
}
88

89
int main()
90
{
91
    SRAND(7767517);
92

93
    return test_convolution3d_0();
94
}
95