Inception Moudel

1、卷積核超參數(shù)選擇困難，自動(dòng)找到卷積的最佳組合。

2、1x1卷積核，不同通道的信息融合。使用1x1卷積核雖然參數(shù)量增加了，但是能夠顯著的降低計(jì)算量(operations)

3、Inception Moudel由4個(gè)分支組成，要分清哪些是在Init里定義，哪些是在forward里調(diào)用。4個(gè)分支在dim=1(channels)上進(jìn)行concatenate。24+16+24+24 = 88

4、GoogleNet的Inception(Pytorch實(shí)現(xiàn))

代碼說(shuō)明：1、先使用類(lèi)對(duì)Inception Moudel進(jìn)行封裝

? ? ? ? ? ? ? ? ? 2、先是1個(gè)卷積層(conv,maxpooling,relu)，然后inceptionA模塊(輸出的channels是24+16+24+24=88)，接下來(lái)又是一個(gè)卷積層(conv,mp,relu),然后inceptionA模塊，最后一個(gè)全連接層(fc)。

? ? ? ? ? ? ? ? ?3、1408這個(gè)數(shù)據(jù)可以通過(guò)x = x.view(in_size, -1)后調(diào)用x.shape得到。

import torch import torch.nn as nn from torchvision import transforms from torchvision import datasets from torch.utils.data import DataLoader import torch.nn.functional as F import torch.optim as optim# prepare datasetbatch_size = 64 transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))]) # 歸一化,均值和方差train_dataset = datasets.MNIST(root='../dataset/mnist/', train=True, download=True, transform=transform) train_loader = DataLoader(train_dataset, shuffle=True, batch_size=batch_size) test_dataset = datasets.MNIST(root='../dataset/mnist/', train=False, download=True, transform=transform) test_loader = DataLoader(test_dataset, shuffle=False, batch_size=batch_size)# design model using class class InceptionA(nn.Module):def __init__(self, in_channels):super(InceptionA, self).__init__()self.branch1x1 = nn.Conv2d(in_channels, 16, kernel_size=1)self.branch5x5_1 = nn.Conv2d(in_channels, 16, kernel_size=1)self.branch5x5_2 = nn.Conv2d(16, 24, kernel_size=5, padding=2)self.branch3x3_1 = nn.Conv2d(in_channels, 16, kernel_size=1)self.branch3x3_2 = nn.Conv2d(16, 24, kernel_size=3, padding=1)self.branch3x3_3 = nn.Conv2d(24, 24, kernel_size=3, padding=1)self.branch_pool = nn.Conv2d(in_channels, 24, kernel_size=1)def forward(self, x):branch1x1 = self.branch1x1(x)branch5x5 = self.branch5x5_1(x)branch5x5 = self.branch5x5_2(branch5x5)branch3x3 = self.branch3x3_1(x)branch3x3 = self.branch3x3_2(branch3x3)branch3x3 = self.branch3x3_3(branch3x3)branch_pool = F.avg_pool2d(x, kernel_size=3, stride=1, padding=1)branch_pool = self.branch_pool(branch_pool)outputs = [branch1x1, branch5x5, branch3x3, branch_pool]return torch.cat(outputs, dim=1) # b,c,w,h c對(duì)應(yīng)的是dim=1class Net(nn.Module):def __init__(self):super(Net, self).__init__()self.conv1 = nn.Conv2d(1, 10, kernel_size=5)self.conv2 = nn.Conv2d(88, 20, kernel_size=5) # 88 = 24x3 + 16self.incep1 = InceptionA(in_channels=10) # 與conv1 中的10對(duì)應(yīng)self.incep2 = InceptionA(in_channels=20) # 與conv2 中的20對(duì)應(yīng)self.mp = nn.MaxPool2d(2)self.fc = nn.Linear(1408, 10) def forward(self, x):in_size = x.size(0)x = F.relu(self.mp(self.conv1(x)))x = self.incep1(x)x = F.relu(self.mp(self.conv2(x)))x = self.incep2(x)x = x.view(in_size, -1)x = self.fc(x)return xmodel = Net()# construct loss and optimizer criterion = torch.nn.CrossEntropyLoss() optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)# training cycle forward, backward, updatedef train(epoch):running_loss = 0.0for batch_idx, data in enumerate(train_loader, 0):inputs, target = dataoptimizer.zero_grad()outputs = model(inputs)loss = criterion(outputs, target)loss.backward()optimizer.step()running_loss += loss.item()if batch_idx % 300 == 299:print('[%d, %5d] loss: %.3f' % (epoch+1, batch_idx+1, running_loss/300))running_loss = 0.0def test():correct = 0total = 0with torch.no_grad():for data in test_loader:images, labels = dataoutputs = model(images)_, predicted = torch.max(outputs.data, dim=1)total += labels.size(0)correct += (predicted == labels).sum().item()print('accuracy on test set: %d %% ' % (100*correct/total))if __name__ == '__main__':for epoch in range(10):train(epoch)test()

視頻中截圖：

說(shuō)明：1、要解決的問(wèn)題：梯度消失

? ? ? ? ? ?2、跳連接，H(x) = F(x) + x,張量維度必須一樣，加完后再激活。不要做pooling，張量的維度會(huì)發(fā)生變化。

代碼說(shuō)明：

1、先是1個(gè)卷積層(conv,maxpooling,relu)，然后ResidualBlock模塊，接下來(lái)又是一個(gè)卷積層(conv,mp,relu),然后esidualBlock模塊模塊，最后一個(gè)全連接層(fc)。

import torch import torch.nn as nn from torchvision import transforms from torchvision import datasets from torch.utils.data import DataLoader import torch.nn.functional as F import torch.optim as optim# prepare datasetbatch_size = 64 transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))]) # 歸一化,均值和方差train_dataset = datasets.MNIST(root='../dataset/mnist/', train=True, download=True, transform=transform) train_loader = DataLoader(train_dataset, shuffle=True, batch_size=batch_size) test_dataset = datasets.MNIST(root='../dataset/mnist/', train=False, download=True, transform=transform) test_loader = DataLoader(test_dataset, shuffle=False, batch_size=batch_size)# design model using class class ResidualBlock(nn.Module):def __init__(self, channels):super(ResidualBlock, self).__init__()self.channels = channelsself.conv1 = nn.Conv2d(channels, channels, kernel_size=3, padding=1)self.conv2 = nn.Conv2d(channels, channels, kernel_size=3, padding=1)def forward(self, x):y = F.relu(self.conv1(x))y = self.conv2(y)return F.relu(x + y)class Net(nn.Module):def __init__(self):super(Net, self).__init__()self.conv1 = nn.Conv2d(1, 16, kernel_size=5)self.conv2 = nn.Conv2d(16, 32, kernel_size=5) # 88 = 24x3 + 16self.rblock1 = ResidualBlock(16)self.rblock2 = ResidualBlock(32)self.mp = nn.MaxPool2d(2)self.fc = nn.Linear(512, 10) # 暫時(shí)不知道1408咋能自動(dòng)出來(lái)的def forward(self, x):in_size = x.size(0)x = self.mp(F.relu(self.conv1(x)))x = self.rblock1(x)x = self.mp(F.relu(self.conv2(x)))x = self.rblock2(x)x = x.view(in_size, -1)x = self.fc(x)return xmodel = Net()# construct loss and optimizer criterion = torch.nn.CrossEntropyLoss() optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)# training cycle forward, backward, updatedef train(epoch):running_loss = 0.0for batch_idx, data in enumerate(train_loader, 0):inputs, target = dataoptimizer.zero_grad()outputs = model(inputs)loss = criterion(outputs, target)loss.backward()optimizer.step()running_loss += loss.item()if batch_idx % 300 == 299:print('[%d, %5d] loss: %.3f' % (epoch+1, batch_idx+1, running_loss/300))running_loss = 0.0def test():correct = 0total = 0with torch.no_grad():for data in test_loader:images, labels = dataoutputs = model(images)_, predicted = torch.max(outputs.data, dim=1)total += labels.size(0)correct += (predicted == labels).sum().item()print('accuracy on test set: %d %% ' % (100*correct/total))if __name__ == '__main__':for epoch in range(10):train(epoch)test()

1

總結(jié)

以上是生活随笔為你收集整理的卷积神经网络(高级篇) Inception Moudel的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問(wèn)題。

如果覺(jué)得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。