說(shuō)明 0、前一部分叫做Feature Extraction，后一部分叫做classification

? ? ? ? 1、每一個(gè)卷積核它的通道數(shù)量要求和輸入通道是一樣的。這種卷積核的總數(shù)有多少個(gè)和你輸出通道的數(shù)量是一樣的。

? ? ? ? 2、卷積(convolution)后，C(Channels)變，W(width)和H(Height)可變可不變，取決于是否padding。subsampling(或pooling)后，C不變，W和H變。

? ? ? ? 3、卷積層：保留圖像的空間信息。

? ? ? ?4、卷積層要求輸入輸出是四維張量(B,C,W,H)，全連接層的輸入與輸出都是二維張量(B,Input_feature)。

? ? ? ? ? ? ?傳送門 PyTorch的nn.Linear（）詳解

? ? ? 5、卷積(線性變換)，激活函數(shù)(非線性變換)，池化；這個(gè)過程若干次后，view打平，進(jìn)入全連接層~

?

?

?

---

1. 卷積操作

import torch # 定義輸入、輸出通道 in_channels, out_channels = 5, 10 # 定義圖像尺寸 width, height = 100, 100 # 定義卷積核的大小，下式表示大小為3*3的正方形，同時(shí)，卷積核的通道數(shù)與輸入圖像的通道數(shù)一致，均為5 kernel_size = 3 # 定義一次輸入圖像的數(shù)量 batch_size = 1input = torch.randn(batch_size,in_channels,width,height)# out_channels 決定了卷積核的數(shù)量, 即一共有10個(gè)3*3*5的卷積核 conv_layer = torch.nn.Conv2d(in_channels,out_channels,kernel_size=kernel_size) output = conv_layer(input)print(input.shape) print(output.shape) print(conv_layer.weight.shape)

輸出：

torch.Size([1, 5, 100, 100]) torch.Size([1, 10, 98, 98]) torch.Size([10, 5, 3, 3])

有時(shí)，我們希望獲得與原圖像相同大小的卷積后的圖像，這時(shí)需要屬性padding，默認(rèn)為0

conv_layer_with_padding = torch.nn.Conv2d(in_channels,out_channels,padding=1,kernel_size = kernel_size) output_with_padding = conv_layer_with_padding(input) print(output_with_padding.shape)

輸出：

torch.Size([1, 10, 100, 100])

還有時(shí)，我們希望再次降低網(wǎng)絡(luò)的大小，以降低運(yùn)算量。此時(shí)引入卷積核移動(dòng)步長(zhǎng)stride的概念，默認(rèn)為1

conv_layer_with_stride = torch.nn.Conv2d(in_channels,out_channels,stride=2,kernel_size=kernel_size)output_with_stride = conv_layer_with_stride(input) print(output_with_stride.shape)

輸出：

torch.Size([1, 10, 49, 49])

2. 下采樣

下采樣與卷積無(wú)本質(zhì)區(qū)別，不同的在于目的。下采樣的目的是將數(shù)據(jù)維度再次減少。
最常用的下采樣手段是Max Pooling 最大池化。

input = [3,4,6,5,2,4,6,8,1,6,7,8,9,7,4,6, ] input = torch.Tensor(input).view(1,1,4,4) maxpooling_layer = torch.nn.MaxPool2d(kernel_size=2) # 注意，我們將kernel_size設(shè)為2，此時(shí)stride默認(rèn)也為2output = maxpooling_layer(input) print(output)

輸出：

tensor([[[[4., 8.],[9., 8.]]]])

3. 卷積神經(jīng)基礎(chǔ)代碼

?

代碼說(shuō)明：

1、torch.nn.Conv2d(1,10,kernel_size=3,stride=2,bias=False)

?1是指輸入的Channel，灰色圖像是1維的；10是指輸出的Channel，也可以說(shuō)第一個(gè)卷積層需要10個(gè)卷積核；kernel_size=3,卷積核大小是3x3；stride=2進(jìn)行卷積運(yùn)算時(shí)的步長(zhǎng)，默認(rèn)為1；bias=False卷積運(yùn)算是否需要偏置bias，默認(rèn)為False。padding = 0，卷積操作是否補(bǔ)0。

2、self.fc = torch.nn.Linear(320, 10)，這個(gè)320獲取的方式，可以通過x = x.view(batch_size, -1)

# print(x.shape)可得到(64,320)，64指的是batch，320就是指要進(jìn)行全連接操作時(shí)，輸入的特征維度。

import torch 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 import matplotlib.pyplot as plt# prepare datasetbatch_size = 64 transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.1307,), (0.3081,))])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 classclass Net(torch.nn.Module):def __init__(self):super(Net, self).__init__()self.conv1 = torch.nn.Conv2d(1, 10, kernel_size=5)self.conv2 = torch.nn.Conv2d(10, 20, kernel_size=5)self.pooling = torch.nn.MaxPool2d(2)self.fc = torch.nn.Linear(320, 10)def forward(self, x):# flatten data from (n,1,28,28) to (n, 784)batch_size = x.size(0)x = F.relu(self.pooling(self.conv1(x)))x = F.relu(self.pooling(self.conv2(x)))x = x.view(batch_size, -1) # -1 此處自動(dòng)算出的是320# print("x.shape",x.shape)x = self.fc(x)return xmodel = Net() device = torch.device("cuda" if torch.cuda.is_available() else "cpu") model.to(device)# construct loss and optimizer criterion = torch.nn.CrossEntropyLoss() optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)# training cycle forward, backward, update def train(epoch):running_loss = 0.0for batch_idx, data in enumerate(train_loader, 0):inputs, target = datainputs, target = inputs.to(device), target.to(device)optimizer.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 = dataimages, labels = images.to(device), labels.to(device)outputs = 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))return correct / totalif __name__ == '__main__':epoch_list = []acc_list = []for epoch in range(10):train(epoch)acc = test()epoch_list.append(epoch)acc_list.append(acc)plt.plot(epoch_list, acc_list)plt.ylabel('accuracy')plt.xlabel('epoch')plt.show()

總結(jié)

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

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