VGG-19架构 pytorch实现
import torch
import torch.nn as nn
from torchinfo import summary
class VGG19(nn.Module):
def __init__(self):
super().__init__()
self.conv1=nn.Conv2d(3,64,kernel_size=3,padding=1)
self.conv2=nn.Conv2d(64,64,kernel_size=3,padding=1)
self.pool1 = nn.MaxPool2d(kernel_size=2,stride=2)
self.conv3=nn.Conv2d(64,128,kernel_size=3,padding=1)
self.conv4=nn.Conv2d(128,128,kernel_size=3,padding=1)
self.pool2=nn.MaxPool2d(kernel_size=2,stride=2)
self.conv5 = nn.Conv2d(128, 256, kernel_size=3,padding=1)
self.conv6 = nn.Conv2d(256, 256, kernel_size=3,padding=1)
self.conv7 = nn.Conv2d(256, 256, kernel_size=3,padding=1)
self.conv8 = nn.Conv2d(256, 256, kernel_size=3,padding=1)
self.pool3=nn.MaxPool2d(kernel_size=2,stride=2)
self.conv9 = nn.Conv2d(256, 512, kernel_size=3,padding=1)
self.conv10 = nn.Conv2d(512, 512, kernel_size=3,padding=1)
self.conv11= nn.Conv2d(512, 512, kernel_size=3,padding=1)
self.conv12= nn.Conv2d(512, 512, kernel_size=3,padding=1)
self.pool4=nn.MaxPool2d(kernel_size=2,stride=2,padding=1)
self.conv13 = nn.Conv2d(512, 512, kernel_size=3,padding=1)
self.conv14 = nn.Conv2d(512, 512, kernel_size=3,padding=1)
self.conv15 = nn.Conv2d(512, 512, kernel_size=3,padding=1)
self.conv16 = nn.Conv2d(512, 512, kernel_size=3,padding=1)
self.pool5=nn.MaxPool2d(kernel_size=2,stride=2)
self.fc1=nn.Linear(7*7*512,4096)
self.fc2=nn.Linear(4096,4096)
self.fc3=nn.Linear(4096,10)
self.relu=nn.ReLU()
self.softmax=nn.Softmax(dim=1)
print("success")
def forward(self,x):
x = self.relu(self.conv1(x))
x = self.relu(self.conv2(x))
x = self.pool1(x)
x = self.relu(self.conv3(x))
x = self.relu(self.conv4(x))
x = self.pool2(x)
x = self.relu(self.conv5(x))
x = self.relu(self.conv6(x))
x = self.relu(self.conv7(x))
x = self.relu(self.conv8(x))
x = self.pool3(x)
x = self.relu(self.conv9(x))
x = self.relu(self.conv10(x))
x = self.relu(self.conv11(x))
x = self.relu(self.conv12(x))
x = self.pool4(x)
x = self.relu(self.conv13(x))
x = self.relu(self.conv14(x))
x = self.relu(self.conv15(x))
x = self.relu(self.conv16(x))
x = self.pool5(x)
# x = x.view(x.size()[0], -1)
x = x.view(-1,7*7*512)
x = self.relu(self.fc1(x))
x = self.relu(self.fc2(x))
output = self.softmax(self.fc3(x))
return output
net= VGG19()
# print(net)
data=torch.ones(size=(10,3,224,224))
net(data)
print(net(data).shape)
print(net(data))
# summary(net)
结果：
=================================================================
Layer (type:depth-idx) Param #
=================================================================
VGG19 --
├─Conv2d: 1-1 1,792
├─Conv2d: 1-2 36,928
├─MaxPool2d: 1-3 --
├─Conv2d: 1-4 73,856
├─Conv2d: 1-5 147,584
├─MaxPool2d: 1-6 --
├─Conv2d: 1-7 295,168
├─Conv2d: 1-8 590,080
├─Conv2d: 1-9 590,080
├─Conv2d: 1-10 590,080
├─MaxPool2d: 1-11 --
├─Conv2d: 1-12 1,180,160
├─Conv2d: 1-13 2,359,808
├─Conv2d: 1-14 2,359,808
├─Conv2d: 1-15 2,359,808
├─MaxPool2d: 1-16 --
├─Conv2d: 1-17 2,359,808
├─Conv2d: 1-18 2,359,808
├─Conv2d: 1-19 2,359,808
├─Conv2d: 1-20 2,359,808
├─MaxPool2d: 1-21 --
├─Linear: 1-22 102,764,544
├─Linear: 1-23 16,781,312
├─Linear: 1-24 40,970
├─ReLU: 1-25 --
├─Softmax: 1-26 --