numpy手动实现卷积代码

import numpy as np
import torch

def ZY_Conv(img, in_channels, out_channels, kernels, bias, stride=1, padding=0):
    # 得到参数
    N, C, H, W = img.shape
    kc,kci,kh, kw = kernels.shape
    p = padding

    #四周填充
    if p:
        img = np.pad(img, ((0, 0), (0, 0), (p, p), (p, p)), 'constant')  # padding along with all axis


        #得到输出的尺寸
    out_h = (H + 2 * padding - kh) // stride + 1
    out_w = (W + 2 * padding - kw) // stride + 1

    #卷积
    outputs = np.zeros([N, out_channels, out_h, out_w])
    for n in range(N):
        for out_c in range(out_channels):
            for in_c in range(in_channels):
                for h in range(out_h):
                    for w in range(out_w):
                        for x in range(kh):
                            for y in range(kw):
                                 outputs[n][out_c][h][w] += img[n][in_c][h * stride + x][w * stride + y] * kernels[out_c][in_c][x][y]
            #添加偏置
            if in_c == in_channels - 1:
               outputs[n][out_c][:][:] += bias[out_c]

    # 转化为tensor格式，内部数据为float32的形式
    outputs = torch.tensor(outputs, dtype=torch.float32)
    return outputs


if __name__ == '__main__':
    img = np.asarray(
        [[
            [
                [0, 1, 2, 3, 4],
                [1, 2, 3, 4, 5],
                [2, 3, 4, 5, 6],
                [3, 4, 5, 6, 7],
                [4, 5, 6, 7, 8],
            ],
            [
                [1, 2, 3, 4, 5],
                [2, 3, 4, 5, 6],
                [3, 4, 5, 6, 7],
                [4, 5, 6, 7, 8],
                [5, 6, 7, 8, 9]
            ],
            [
                [2, 3, 4, 5, 6],
                [3, 4, 5, 6, 7],
                [4, 5, 6, 7, 8],
                [5, 6, 7, 8, 9],
                [6, 7, 8, 9, 0]
            ],
        ]]
    )

    kernels = np.asarray(
        [
            [
                [
                    [0, 0, 0],
                    [0, 1, 0],
                    [0, 0, 0]
                ],
                [
                    [0, 0, 0],
                    [0, 1, 1],
                    [0, 0, 0]
                ],
                [
                    [0, 0, 0],
                    [0, 1, 0],
                    [0, 0, 1]
                ],
            ],

            [
                [
                    [1, 2, -1],
                    [0, 1, 0],
                    [0, 0, 0]
                ],
                [
                    [0, 0, 0],
                    [0, 1, 1],
                    [0, 0, 0]
                ],
                [
                    [0, 0, 0],
                    [0, 1, 0],
                    [0, 0, 1]
                ],
            ]
        ]
    )


    bias = np.asarray(
        [-0.0722, -0.1841]
    )


    in_channels = 3
    out_channels = 2

    outputs = ZY_Conv(img, in_channels, out_channels, kernels, bias, stride=2, padding=1)
    print(outputs)

卷积运行结果：

与pytorch自带卷积函数对比：

from __future__ import print_function, division
import torch
import os
import numpy as np
import torch.nn as nn
from   zy_conv2 import ZY_Conv

img = np.asarray(
    [[
        [
            [0.0, 1.0, 2.0, 3.0, 4.0],
            [1.0, 2.0, 3.0, 4.0, 5.0],
            [2.0, 3.0, 4.0, 5.0, 6.0],
            [3.0, 4.0, 5.0, 6.0, 7.0],
            [4.0, 5.0, 6.0, 7.0, 8.0],
        ],
        [
            [1.0, 2.0, 3.0, 4.0, 5.0],
            [2.0, 3.0, 4.0, 5.0, 6.0],
            [3.0, 4.0, 5.0, 6.0, 7.0],
            [4.0, 5.0, 6.0, 7.0, 8.0],
            [5.0, 6.0, 7.0, 8.0, 9.0]
        ],
        [
            [2.0, 3.0, 4.0, 5.0, 6.0],
            [3.0, 4.0, 5.0, 6.0, 7.0],
            [4.0, 5.0, 6.0, 7.0, 8.0],
            [5.0, 6.0, 7.0, 8.0, 9.0],
            [6.0, 7.0, 8.0, 9.0, 0.0]
        ],
    ]]
)

#将img转化为tensor类型
img= torch.tensor(img, dtype=torch.float32)
conv= nn.Conv2d(3, 2, 3, 1, 1)#设置输入通道为3，输出通道为2，卷积核大小为2，步长为2，零填充为1


#进行卷积操作，输出卷积核和偏置及卷积结果
out=conv(img)
print("卷积核：")
print(conv.weight)
print("偏执：")
print(conv.bias)
print("pytorch自带卷积函数运行结果：")
print(out)

#提取出卷积核与偏置矩阵
kernels=conv.weight
kernals=kernels.detach().numpy()
bias=conv.bias
bias=bias.detach().numpy()

in_channels = 3
out_channels = 2

#将提取出来的卷积核与偏置矩阵代入自编的卷积函数中，对比自编函数与pytorch自带函数的结果
outputs = ZY_Conv(img, in_channels, out_channels, kernels, bias, stride=1, padding=1)
print("自编卷积函数运行结果：")
print(outputs)

对比结果：