<address id="rphpx"><listing id="rphpx"></listing></address>

<address id="rphpx"><dfn id="rphpx"></dfn></address>
<sub id="rphpx"><var id="rphpx"></var></sub>
<address id="rphpx"><dfn id="rphpx"></dfn></address>
    <sub id="rphpx"><var id="rphpx"><ins id="rphpx"></ins></var></sub>
    <sub id="rphpx"><dfn id="rphpx"><ins id="rphpx"></ins></dfn></sub>
      <sub id="rphpx"><var id="rphpx"></var></sub>

      <sub id="rphpx"><var id="rphpx"></var></sub>

        <sub id="rphpx"><var id="rphpx"><ins id="rphpx"></ins></var></sub>

        <thead id="rphpx"><var id="rphpx"><output id="rphpx"></output></var></thead>

        <sub id="rphpx"><listing id="rphpx"></listing></sub>

          <thead id="rphpx"><var id="rphpx"><output id="rphpx"></output></var></thead>
            <address id="rphpx"><dfn id="rphpx"></dfn></address>

            pytoch-基本卷积网络结构, 参数提取,参数初始化

            基本卷积网络结构net.py 

            from torch import nn
            
            class SimpleCNN(nn.Module):
                def __init__(self):
                    super(SimpleCNN, self).__init__()
                    layer1 = nn.Sequential() # 将网络模型进行添加
                    layer1.add_module(conv1, nn.Conv2d(3, 32, 3, 1, padding=1)) # nn.Conv
                    layer1.add_module(relu1, nn.ReLU(True))
                    layer1.add_module(pool1, nn.MaxPool2d(2, 2))
                    self.layer1 = layer1
            
                    layer2 = nn.Sequential()
                    layer2.add_module(conv2, nn.Conv2d(32, 64, 3, 1, padding=1))
                    layer2.add_module(relu2, nn.ReLU(True))
                    layer2.add_module(pool2, nn.MaxPool2d(2, 2))
                    self.layer2 = layer2
            
                    layer3 = nn.Sequential()
                    layer3.add_module(conv3, nn.Conv2d(64, 128, 3, 1, padding=1))
                    layer3.add_module(relu3, nn.ReLU(True))
                    layer3.add_module(pool3, nn.MaxPool2d(2, 2))
                    self.layer3 = layer3
            
                    layer4 = nn.Sequential()
                    layer4.add_module(fc1, nn.Linear(2048, 512))
                    layer4.add_module(fc_relu1, nn.ReLU(True))
                    layer4.add_module(fc2, nn.Linear(512, 64))
                    layer4.add_module(fc_relu2, nn.ReLU(True))
                    layer4.add_module(fc3, nn.Linear(64, 10))
                    self.layer4 = layer4
            
                def forward(self, x):
                    conv1 = self.layer1(x)
                    conv2 = self.layer2(conv1)
                    conv3 = self.layer3(conv2)
                    fc_input = conv3.view(conv3.size(0), -1)
                    fc_out = self.layer4(fc_input)
            
                    return fc_out
            
            model = SimpleCNN()
            # print(model) # 打印输出网络结构

            提取前两层的网络结构

            new_model = nn.Sequential(*list(model.children())[:2])  # 提取前两层的网络结构, 构造nn.Sequential网络串接, * 表示将里面的内容一个个传进去

            提取所有层的网络结构

            conv_model = nn.Sequential()
            # 提取所有的卷积层操作, model.name_modules() 提取所有层的网络结构
            for name, layer in model.named_modules():
                if isinstance(layer, nn.Conv2d):
                    name = name.replace(., _)
                    conv_model.add_module(name, layer)
            print(conv_model)
            相关文章
            相关标签/搜索
            白小姐四不像必中一肖+正版+网址930