[DL] Basic-Net

Nets

杂七杂八的介绍

  • LeNet 1994

是早期（1994年）的神经网络之一，用于手写数字识别，由卷积层，池化层，全连接层组成，网络结构如下图所示

Structure of LeNet
  • AlexNet 2012

是首个实用性很强的卷积神经网络由卷积操，池化层，全连接层，softmax层以及ReLU、Dropout构成。首次提出在2012年的ILSVRC大规模视觉识别竞赛上。其网络结构如下图所示：

Structure of AlexNet
  • VGGNet 2014

VGGNet出现在2014年ILSVRC上比赛上获得了分类项目的第二名和定位项目的第一名，VGGNet相对于AlexNet堆叠了更多基础模块导致网络深度达到近二十层，另外它将之前5x5，7x7的卷积核替换成3x3的小卷积核，用2x2池化代替3x3，去除了局部响应归一化

。在训练高级别的网络时，可以先训练低级别的网络，用前者获得的权重初始化高级别的网络，可以加速网络的收敛。网络参数如下表所示：

Structure of VGGNet

  • LeNet 1994

  • AlexNet 2012

  • VGGNet 2014

  • GoogleNet 2014

  • ResNet 2016

  • DenseNet 2017

  • SqueezeNet 2017

  • MobileNet 2017

  • SEnet 2018

LeNet

是一系列网络，包括LeNet 1-5

Yann LeCun 等人在 1990 年

7层神经网络，包括3个卷积层，2个池化层，2个全连接层。所有卷积核5x5， stride = 1. 池化为全局，激活函数为Sigmoid

用pytorch展现网络架构

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
import torch
from torch import nn
from d2l import torch as d2l
class Reshape(torch.nn.Module):
    def forward(self, x):
        return x.view(-1, 1, 28, 28)
    
net = torch.nn.Sequential(
    Reshape(), 
    nn.Conv2d(1, 6, kernel_size=5, padding=2), nn.Sigmoid(),  # padding表示在边缘加2，因为本来是32x32的
    nn.AvgPool2d(2, stride=2),
    nn.Conv2d(6, 16, kernel_size=5), nn.Sigmoid(),
    nn.AvgPool2d(2, stride=2), 
    nn.Flatten(),
    nn.Linear(16 * 5 * 5, 120), nn.Sigmoid(),
    nn.Linear(120, 84), nn.Sigmoid(),
    nn.Linear(84, 10)
)
1
2
3
4
X = torch.rand(size=(1, 1, 28, 28), dtype=torch.float32)
for layer in net:
    X = layer(X)
    print(layer.__class__.__name__, 'output shape: \t', X.shape)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
# 输出
Reshape output shape: 	 torch.Size([1, 1, 28, 28])
Conv2d output shape: 	 torch.Size([1, 6, 28, 28])
Sigmoid output shape: 	 torch.Size([1, 6, 28, 28])
AvgPool2d output shape: 	 torch.Size([1, 6, 14, 14])
Conv2d output shape: 	 torch.Size([1, 16, 10, 10])
Sigmoid output shape: 	 torch.Size([1, 16, 10, 10])
AvgPool2d output shape: 	 torch.Size([1, 16, 5, 5])
Flatten output shape: 	 torch.Size([1, 400])
Linear output shape: 	 torch.Size([1, 120])
Sigmoid output shape: 	 torch.Size([1, 120])
Linear output shape: 	 torch.Size([1, 84])
Sigmoid output shape: 	 torch.Size([1, 84])
Linear output shape: 	 torch.Size([1, 10])
1
2
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size=batch_size)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
def evaluate_accuracy_gpu(net, data_iter, device=None): #@save
    """使用GPU计算模型在数据集上的精度"""
    if isinstance(net, nn.Module):
        net.eval()  # 设置为评估模式
        if not device:
            device = next(iter(net.parameters())).device
    # 正确预测的数量，总预测的数量
    metric = d2l.Accumulator(2)
    with torch.no_grad():
        for X, y in data_iter:
            if isinstance(X, list):
                # BERT微调所需的（之后将介绍）
                X = [x.to(device) for x in X]
            else:
                X = X.to(device)
            y = y.to(device)
            metric.add(d2l.accuracy(net(X), y), y.numel())
    return metric[0] / metric[1]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
def train_ch6(net, train_iter, test_iter, num_epochs, lr, device):
    def init_weights(m):
        if type(m) == nn.Linear or type(m) == nn.Conv2d:
            nn.init.xavier_uniform_(m.weight)
    net.apply(init_weights)
    print("training on ", device)
    net.to(device)
    optimizer = torch.optim.SGD(net.parameters(), lr=lr)
    loss = nn.CrossEntropyLoss()
    animator = d2l.Animator(xlabel='epoch', xlim=[1, num_epochs],
                            legend=['train loss', 'train acc', 'test acc'])
    timer, num_batches = d2l.Timer(), len(train_iter)
    for epoch in range(num_epochs):
        metric = d2l.Accumulator(3)
        net.train()
        for i, (X, y) in enumerate(train_iter):
            timer.start()
            optimizer.zero_grad()
            X, y = X.to(device), y.to(device)
            y_hat = net(X)
            l = loss(y_hat, y)
            l.backward()
            optimizer.step()
            with torch.no_grad():
                metric.add(l * X.shape[0], d2l.accuracy(y_hat, y), X.shape[0])
            timer.stop()
            train_l = metric[0] / metric[2]
            train_acc = metric[1] / metric[2]
            if (i + 1) % (num_batches // 5) == 0 or i == num_batches - 1:
                animator.add(epoch + (i + 1) / num_batches,
                             (train_l, train_acc, None))
        test_acc = evaluate_accuracy_gpu(net, test_iter)
        animator.add(epoch + 1, (None, None, test_acc))
    print(f'loss {train_l:.3f}, train acc {train_acc:.3f}, '
          f'test acc {test_acc:.3f}')
    print(f'{metric[2] * num_epochs / timer.sum():.1f} examples/sec '
          f'on {str(device)}')
1
2
lr, num_epochs = 0.9, 20
train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
1
2
loss 0.356, train acc 0.868, test acc 0.849
56757.7 examples/sec on cuda:0
result

AlexNet

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
import torch
from torch import nn
from d2l import torch as d2l

net = nn.Sequential(
    # 这里使用一个11*11的更大窗口来捕捉对象。
    # 同时，步幅为4，以减少输出的高度和宽度。
    # 另外，输出通道的数目远大于LeNet
    nn.Conv2d(1, 96, kernel_size=11, stride=4, padding=1), nn.ReLU(),
    nn.MaxPool2d(kernel_size=3, stride=2),
    # 减小卷积窗口，使用填充为2来使得输入与输出的高和宽一致，且增大输出通道数
    nn.Conv2d(96, 256, kernel_size=5, padding=2), nn.ReLU(),
    nn.MaxPool2d(kernel_size=3, stride=2),
    # 使用三个连续的卷积层和较小的卷积窗口。
    # 除了最后的卷积层，输出通道的数量进一步增加。
    # 在前两个卷积层之后，汇聚层不用于减少输入的高度和宽度
    nn.Conv2d(256, 384, kernel_size=3, padding=1), nn.ReLU(),
    nn.Conv2d(384, 384, kernel_size=3, padding=1), nn.ReLU(),
    nn.Conv2d(384, 256, kernel_size=3, padding=1), nn.ReLU(),
    nn.MaxPool2d(kernel_size=3, stride=2),
    nn.Flatten(),
    # 这里，全连接层的输出数量是LeNet中的好几倍。使用dropout层来减轻过拟合
    nn.Linear(6400, 4096), nn.ReLU(),
    nn.Dropout(p=0.5),
    nn.Linear(4096, 4096), nn.ReLU(),
    nn.Dropout(p=0.5),
    # 最后是输出层。由于这里使用Fashion-MNIST，所以用类别数为10，而非论文中的1000
    nn.Linear(4096, 10))
1
2
3
4
X = torch.randn(1, 1, 224, 224)
for layer in net:
    X=layer(X)
    print(layer.__class__.__name__,'output shape:\t',X.shape)
Conv2d output shape:     torch.Size([1, 96, 54, 54])
ReLU output shape:   torch.Size([1, 96, 54, 54])
MaxPool2d output shape:  torch.Size([1, 96, 26, 26])
Conv2d output shape:     torch.Size([1, 256, 26, 26])
ReLU output shape:   torch.Size([1, 256, 26, 26])
MaxPool2d output shape:  torch.Size([1, 256, 12, 12])
Conv2d output shape:     torch.Size([1, 384, 12, 12])
ReLU output shape:   torch.Size([1, 384, 12, 12])
Conv2d output shape:     torch.Size([1, 384, 12, 12])
ReLU output shape:   torch.Size([1, 384, 12, 12])
Conv2d output shape:     torch.Size([1, 256, 12, 12])
ReLU output shape:   torch.Size([1, 256, 12, 12])
MaxPool2d output shape:  torch.Size([1, 256, 5, 5])
Flatten output shape:    torch.Size([1, 6400])
Linear output shape:     torch.Size([1, 4096])
ReLU output shape:   torch.Size([1, 4096])
Dropout output shape:    torch.Size([1, 4096])
Linear output shape:     torch.Size([1, 4096])
ReLU output shape:   torch.Size([1, 4096])
Dropout output shape:    torch.Size([1, 4096])
Linear output shape:     torch.Size([1, 10])
1
2
batch_size = 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=224)
1
2
lr, num_epochs = 0.01, 10
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
1
2
3
# 输出
loss 0.331, train acc 0.878, test acc 0.882
1635.7 examples/sec on cuda:0
result

VGGNet

使用块，更大更深的AlexNet

  • 带填充以保持分辨率的卷积层；

  • 非线性激活函数，如ReLU；

  • 汇聚层，如最大汇聚层。

1
2
3
4
5
6
7
8
9
10
11
12
13
import torch
from torch import nn
from d2l import torch as d2l

def vgg_block(num_convs, in_channels, out_channels):
    layers = []
    for _ in range(num_convs):
        layers.append(nn.Conv2d(in_channels, out_channels,
                                kernel_size=3, padding=1))
        layers.append(nn.ReLU())
        in_channels = out_channels
    layers.append(nn.MaxPool2d(kernel_size=2, stride=2))
    return nn.Sequential(*layers)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
conv_arch = ((1, 64), (1, 128), (2, 256), (2, 512), (2, 512))
def vgg(conv_arch):
    conv_blks = []
    in_channels = 1
    for (num_convs, out_channels) in conv_arch:
        conv_blks.append(vgg_block(num_convs, in_channels, out_channels))
        in_channels = out_channels
    return nn.Sequential(
        *conv_blks, 
        nn.Flatten(),
        nn.Linear(out_channels * 7 * 7, 4096), nn.ReLU(),
        nn.Dropout(0.5), nn.Linear(4096, 4096), nn.ReLU(),
        nn.Dropout(0.5), nn.Linear(4096, 10)
    )
net = vgg(conv_arch)
1
2
3
4
5
X = torch.randn(size=(1, 1, 224, 224))
for blk in net:
    X = blk(X)
    print(blk.__class__.__name__, 'output shape:\t', X.shape)
    
Sequential output shape:     torch.Size([1, 64, 112, 112])
Sequential output shape:     torch.Size([1, 128, 56, 56])
Sequential output shape:     torch.Size([1, 256, 28, 28])
Sequential output shape:     torch.Size([1, 512, 14, 14])
Sequential output shape:     torch.Size([1, 512, 7, 7])
Flatten output shape:    torch.Size([1, 25088])
Linear output shape:     torch.Size([1, 4096])
ReLU output shape:   torch.Size([1, 4096])
Dropout output shape:    torch.Size([1, 4096])
Linear output shape:     torch.Size([1, 4096])
ReLU output shape:   torch.Size([1, 4096])
Dropout output shape:    torch.Size([1, 4096])
Linear output shape:     torch.Size([1, 10])
1
2
3
ratio = 4
small_conv_arch = [(pair[0], pair[1] // ratio) for pair in conv_arch]
net = vgg(small_conv_arch)
1
2
3
4
X = torch.randn(size=(1, 1, 224, 224))
for blk in net:
    X = blk(X)
    print(blk.__class__.__name__, 'output shape:\t', X.shape)
Sequential output shape:     torch.Size([1, 16, 112, 112])
Sequential output shape:     torch.Size([1, 32, 56, 56])
Sequential output shape:     torch.Size([1, 64, 28, 28])
Sequential output shape:     torch.Size([1, 128, 14, 14])
Sequential output shape:     torch.Size([1, 128, 7, 7])
Flatten output shape:    torch.Size([1, 6272])
Linear output shape:     torch.Size([1, 4096])
ReLU output shape:   torch.Size([1, 4096])
Dropout output shape:    torch.Size([1, 4096])
Linear output shape:     torch.Size([1, 4096])
ReLU output shape:   torch.Size([1, 4096])
Dropout output shape:    torch.Size([1, 4096])
Linear output shape:     torch.Size([1, 10])
1
2
3
lr, num_epochs, batch_size = 0.05, 10, 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=224)
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
loss 0.177, train acc 0.935, test acc 0.894
1091.9 examples/sec on cuda:0
result

GoogleNet/Inception V3

Inception 块

输入被copy成四块

image-20230214141018466
image-20230214141928373

用了很多1x1卷积，降低通道数。

1
2
3
4
import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l


class Inception(nn.Module):
    # c1--c4是每条路径的输出通道数
    def __init__(self, in_channels, c1, c2, c3, c4, **kwargs):
        super(Inception, self).__init__(**kwargs)
        # 线路1，单1x1卷积层
        self.p1_1 = nn.Conv2d(in_channels, c1, kernel_size=1)
        # 线路2，1x1卷积层后接3x3卷积层
        self.p2_1 = nn.Conv2d(in_channels, c2[0], kernel_size=1)
        self.p2_2 = nn.Conv2d(c2[0], c2[1], kernel_size=3, padding=1)
        # 线路3，1x1卷积层后接5x5卷积层
        self.p3_1 = nn.Conv2d(in_channels, c3[0], kernel_size=1)
        self.p3_2 = nn.Conv2d(c3[0], c3[1], kernel_size=5, padding=2)
        # 线路4，3x3最大汇聚层后接1x1卷积层
        self.p4_1 = nn.MaxPool2d(kernel_size=3, stride=1, padding=1)
        self.p4_2 = nn.Conv2d(in_channels, c4, kernel_size=1)

    def forward(self, x):
        p1 = F.relu(self.p1_1(x))
        p2 = F.relu(self.p2_2(F.relu(self.p2_1(x))))
        p3 = F.relu(self.p3_2(F.relu(self.p3_1(x))))
        p4 = F.relu(self.p4_2(self.p4_1(x)))
        # 在通道维度上连结输出
        return torch.cat((p1, p2, p3, p4), dim=1)
1
2
3
b1 = nn.Sequential(nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3),
                   nn.ReLU(),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
1
2
3
4
5
b2 = nn.Sequential(nn.Conv2d(64, 64, kernel_size=1),
                   nn.ReLU(),
                   nn.Conv2d(64, 192, kernel_size=3, padding=1),
                   nn.ReLU(),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
1
2
3
b3 = nn.Sequential(Inception(192, 64, (96, 128), (16, 32), 32),
                   Inception(256, 128, (128, 192), (32, 96), 64),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
1
2
3
4
5
6
b4 = nn.Sequential(Inception(480, 192, (96, 208), (16, 48), 64),
                   Inception(512, 160, (112, 224), (24, 64), 64),
                   Inception(512, 128, (128, 256), (24, 64), 64),
                   Inception(512, 112, (144, 288), (32, 64), 64),
                   Inception(528, 256, (160, 320), (32, 128), 128),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1))
1
2
3
4
5
6
b5 = nn.Sequential(Inception(832, 256, (160, 320), (32, 128), 128),
                   Inception(832, 384, (192, 384), (48, 128), 128),
                   nn.AdaptiveAvgPool2d((1,1)),
                   nn.Flatten())

net = nn.Sequential(b1, b2, b3, b4, b5, nn.Linear(1024, 10))
1
2
3
4
X = torch.rand(size=(1, 1, 96, 96))
for layer in net:
    X = layer(X)
    print(layer.__class__.__name__,'output shape:\t', X.shape)
Sequential output shape:     torch.Size([1, 64, 24, 24])
Sequential output shape:     torch.Size([1, 192, 12, 12])
Sequential output shape:     torch.Size([1, 480, 6, 6])
Sequential output shape:     torch.Size([1, 832, 3, 3])
Sequential output shape:     torch.Size([1, 1024])
Linear output shape:     torch.Size([1, 10])
1
2
3
lr, num_epochs, batch_size = 0.1, 10, 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=96)
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
loss 0.246, train acc 0.907, test acc 0.893
1690.6 examples/sec on cuda:0
image-20230214200912069

ResNet

image-20230214201006292
image-20230214201016846 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l


class Residual(nn.Module):
    def __init__(self, input_channels, num_channels,
                 use_1x1conv=False, strides=1):
        super().__init__()
        self.conv1 = nn.Conv2d(input_channels, num_channels,
                               kernel_size=3, padding=1, stride=strides)
        self.conv2 = nn.Conv2d(num_channels, num_channels,
                               kernel_size=3, padding=1)
        if use_1x1conv:
            self.conv3 = nn.Conv2d(input_channels, num_channels,
                                   kernel_size=1, stride=strides)
        else:
            self.conv3 = None
        self.bn1 = nn.BatchNorm2d(num_channels)
        self.bn2 = nn.BatchNorm2d(num_channels)

    def forward(self, X):
        Y = F.relu(self.bn1(self.conv1(X)))
        Y = self.bn2(self.conv2(Y))
        if self.conv3:
            X = self.conv3(X)
        Y += X
        return F.relu(Y)
1
2
3
4
b1 = nn.Sequential(nn.Conv2d(1, 64, kernel_size=7, stride=2, padding=3),
                   nn.BatchNorm2d(64), nn.ReLU(),
                   nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
                  )
1
2
3
4
5
6
7
8
9
10
def resnet_block(input_channels, num_channels, num_residuals,
                 first_block=False):
    blk = []
    for i in range(num_residuals):
        if i == 0 and not first_block:
            blk.append(Residual(input_channels, num_channels,
                                use_1x1conv=True, strides=2))
        else:
            blk.append(Residual(num_channels, num_channels))
    return blk
1
2
3
4
b2 = nn.Sequential(*resnet_block(64, 64, 2, first_block=True))
b3 = nn.Sequential(*resnet_block(64, 128, 2))
b4 = nn.Sequential(*resnet_block(128, 256, 2))
b5 = nn.Sequential(*resnet_block(256, 512, 2))
1
2
3
net = nn.Sequential(b1, b2, b3, b4, b5,
                    nn.AdaptiveAvgPool2d((1,1)),
                    nn.Flatten(), nn.Linear(512, 10))
1
2
3
4
X = torch.rand(size=(1, 1, 224, 224))
for layer in net:
    X = layer(X)
    print(layer.__class__.__name__,'output shape:\t', X.shape)
Sequential output shape:     torch.Size([1, 64, 56, 56])
Sequential output shape:     torch.Size([1, 64, 56, 56])
Sequential output shape:     torch.Size([1, 128, 28, 28])
Sequential output shape:     torch.Size([1, 256, 14, 14])
Sequential output shape:     torch.Size([1, 512, 7, 7])
AdaptiveAvgPool2d output shape:  torch.Size([1, 512, 1, 1])
Flatten output shape:    torch.Size([1, 512])
Linear output shape:     torch.Size([1, 10])
1
2
3
lr, num_epochs, batch_size = 0.05, 10, 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=96)
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
loss 0.015, train acc 0.996, test acc 0.893
2613.9 examples/sec on cuda:0
image-20230214201027140
#notes
[DL] Basic-Net
http://jamil-yu.github.io/2023/02/19/[DL]Basic-Net/
Author
Jamil Yu
Posted on
February 19, 2023
Licensed under