卷积层详解

# 1.卷积原理
# Conv1d代表一维卷积，Conv2d代表二维卷积，Conv3d代表三维卷积。
# kernel_size在训练过程中不断调整，定义为3就是3 * 3的卷积核，实际我们在训练神经网络过程中其实就是对kernel_size不断调整。
# 可以根据输入的参数获得输出的情况
# Input(N_in,C_in,H_in,W_in)
# Output(N_out,C_out,H_out,W_out)
# H_out={H_in+2xpadding[0]-dilation[0]x(kernel_size[0]-1)-1}/{stride[0]}+1 整体向下取整数
# W_out={W_in+2xpadding[0]-dilation[1]x(kernel_size[1]-1)-1}/{stride[1]}+1 整体向下取整数
import torch
import torchvision
from torch import nn
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter

# 2.搭建卷积层
dataset = torchvision.datasets.CIFAR10(root='D:\PyCharm\CIFAR10', train=False,
                                       transform=torchvision.transforms.ToTensor(), download=False)
dataloader = DataLoader(dataset=dataset, batch_size=64)


class MyModel(nn.Module):
    def __init__(self):
        super(MyModel, self).__init__()
        # 彩色图像输入为3层，我们想让它的输出为6层，选3 * 3 的卷积
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=6, kernel_size=3, stride=1, padding=0)

    def forward(self, x):
        x = self.conv1(x)
        return x


model = MyModel()
print(model)

# 3.卷积层处理图片
dataset = torchvision.datasets.CIFAR10(root='D:\PyCharm\CIFAR10', train=False,
                                       transform=torchvision.transforms.ToTensor())
dataloader = DataLoader(dataset=dataset, batch_size=64)


class MyModel(nn.Module):
    def __init__(self):
        super(MyModel, self).__init__()
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=6, kernel_size=3, stride=1, padding=0)

    def forward(self, x):
        x = self.conv1(x)
        return x


model = MyModel()
for data in dataloader:
    imgs, targets = data
    output = model(imgs)
    print(imgs.shape)  # 输入为3通道32×32的64张图片
    print(output.shape)  # 输出为6通道30×30的64张图片

# tensorboard显示
dataset = torchvision.datasets.CIFAR10(root='D:\PyCharm\CIFAR10', train=False,
                                       transform=torchvision.transforms.ToTensor())
dataloader = DataLoader(dataset=dataset, batch_size=64)


class MyModel(nn.Module):
    def __init__(self):
        super(MyModel, self).__init__()
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=6, kernel_size=3, stride=1, padding=0)

    def forward(self, x):
        x = self.conv1(x)
        return x


model = MyModel()
writer = SummaryWriter("logs9")
step = 0
for data in dataloader:
    imgs, targets = data
    output = model(imgs)
    print(imgs.shape)
    print(output.shape)
    for i, img in enumerate(imgs):
        img_input_hwc = img.permute(1, 2, 0)
        writer.add_image('imput', img_input_hwc, step, dataformats="HWC")
    output = torch.reshape(output, (-1, 3, 30, 30))
    # 把原来6个通道拉为3个通道，为了保证所有维度总数不变，其余的分量分到第一个维度中
    for i, img in enumerate(imgs):
        img_output_hwc = img.permute(1, 2, 0)
        writer.add_image('output', img_output_hwc, step, dataformats="HWC")
    step = step + 1
writer.close()