mnist数据识别-Dropout 02

最新推荐文章于 2022-01-04 22:24:02 发布

qq_41005604

最新推荐文章于 2022-01-04 22:24:02 发布

阅读量147

点赞数

分类专栏： Pytorch CNN LSTM 文章标签：深度学习

本文链接：https://blog.csdn.net/qq_41005604/article/details/115350992

版权

Pytorch CNN LSTM 专栏收录该内容

6 篇文章 0 订阅

订阅专栏

Dropout的使用主要是在定义网络结构

在01的基础上，把改变的网络结构放在这里了

# 定义网络结构

'''
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.layer1 = nn.Sequential【用顺序模型封装起来】(nn.Linear(784,500), nn.Dropout(p=0.5), nn.Tanh()【激活函数】)
        self.layer2 = nn.Sequential(nn.Linear(500,300), nn.Dropout(p=0.5), nn.Tanh())
        self.layer3 = nn.Sequential(nn.Linear(300,10), nn.Softmax(dim=1))【输出层一般不需要加dropout】
        
    def forward(self,x):
        # ([64, 1, 28, 28])->(64,784)
        x = x.view(x.size()[0], -1)
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        return x
'''
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.layer1 = nn.Sequential(nn.Linear(784,500), nn.Dropout(p=0.5), nn.Tanh())
        self.layer2 = nn.Sequential(nn.Linear(500,300), nn.Dropout(p=0.5), nn.Tanh())
        self.layer3 = nn.Sequential(nn.Linear(300,10), nn.Softmax(dim=1))
        
    def forward(self,x):
        # ([64, 1, 28, 28])->(64,784)
        x = x.view(x.size()[0], -1)
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        return x