pytorch实战——卷积层

最新推荐文章于 2024-01-25 08:45:42 发布

cfqp

最新推荐文章于 2024-01-25 08:45:42 发布

阅读量221

点赞数

分类专栏： pytorch python 神经网络与深度学习文章标签： pytorch 深度学习神经网络 Powered by 金山文档

本文链接：https://blog.csdn.net/cfqp123456/article/details/128739006

版权

pytorch 同时被 3 个专栏收录

4 篇文章 0 订阅

订阅专栏

python

3 篇文章 0 订阅

订阅专栏

神经网络与深度学习

1 篇文章 0 订阅

订阅专栏

import torchvision
from torch import nn
from torch.utils.data import DataLoader

# 测试集
from torch.utils.tensorboard import SummaryWriter

test_set = torchvision.datasets.CIFAR10("./dataset", train=False, transform=
torchvision.transforms.ToTensor(), download=True)

# 加载测试集
dataloader = DataLoader(test_set, batch_size=64)


# 定义神经网络
class Tingyu(nn.Module):
    def __init__(self):
        super(Tingyu, self).__init__()
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=3, kernel_size=3, stride=1,
                               padding=0)

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


# 创建实例,查看定义的神经网络
tingyu = Tingyu()
print(tingyu)
# Tingyu(
#   (conv1): Conv2d(3, 3, kernel_size=(3, 3), stride=(1, 1))
# )


# 图片送进神经网络，使用tensorboard展示
writer = SummaryWriter("conv_logs")
step = 0
for data in dataloader:
    images, targets = data
    # print(images.shape)   # (64,3,32,32)
    writer.add_images("input", images, step)
    output = tingyu(images)
    # print(output.shape)    #(64,3,30,30)
    # output = torch.reshape(output, (-1, 3, 30, 30))
    writer.add_images("output", output, step)
    step = step + 1

writer.close()