machine-learning/07_convolution/07_modern_cnn.ipynb

现代卷积神经网络

目录

• 深度卷积神经网络（AlexNet）
• 使用块的网络（VGG）
• 网络中的网络（NiN）
• 含并行连结的网络（GoogLeNet）
• 批量规范化
• 残差网络（ResNet）

深度卷积神经网络（AlexNet）

In [1]:

import torch
from torch import nn
from d2l import torch as d2l

net = nn.Sequential(
    # 这里，我们使用一个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))

我们构造一个 单通道数据，来观察每一层输出的形状

In [2]:

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])

Fashion-MNIST图像的分辨率 低于ImageNet图像。 我们将它们增加到224 \times 224

In [3]:

batch_size = 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=224)

训练AlexNet。注意：这里是欠拟合的标准曲线

In [4]:

lr, num_epochs = 0.01, 10
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())

loss 0.333, train acc 0.878, test acc 0.878
1762.7 examples/sec on cuda:0

使用块的网络（VGG）

VGG块

In [1]:

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)

VGG网络

原始VGG网络有5个卷积块，其中前两个块各有一个卷积层，后三个块各包含两个卷积层。 第一个模块有64个输出通道，每个后续模块将输出通道数量翻倍，直到该数字达到512。由于该网络使用8个卷积层和3个全连接层，因此它通常被称为VGG-11。

In [2]:

conv_arch = ((1, 64), (1, 128), (2, 256), (2, 512), (2, 512))

In [3]:

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)

观察每个层输出的形状

In [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, 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])

由于VGG-11比AlexNet计算量更大，因此我们构建了一个通道数较少的网络

In [5]:

ratio = 4
small_conv_arch = [(pair[0], pair[1] // ratio) for pair in conv_arch]
net = vgg(small_conv_arch)

模型训练。仔细观察：这里出现了过拟合

In [6]:

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 0.172, train acc 0.936, test acc 0.914
1119.5 examples/sec on cuda:0

网络中的网络（NiN）

NiN块

In [2]:

import torch
from torch import nn
from d2l import torch as d2l


def nin_block(in_channels, out_channels, kernel_size, strides, padding):
    return nn.Sequential(
        nn.Conv2d(in_channels, out_channels, kernel_size, strides, padding),
        nn.ReLU(),
        nn.Conv2d(out_channels, out_channels, kernel_size=1), nn.ReLU(),
        nn.Conv2d(out_channels, out_channels, kernel_size=1), nn.ReLU())

NiN模型

In [3]:

net = nn.Sequential(
    nin_block(1, 96, kernel_size=11, strides=4, padding=0),
    nn.MaxPool2d(3, stride=2),
    nin_block(96, 256, kernel_size=5, strides=1, padding=2),
    nn.MaxPool2d(3, stride=2),
    nin_block(256, 384, kernel_size=3, strides=1, padding=1),
    nn.MaxPool2d(3, stride=2),
    nn.Dropout(0.5),
    nin_block(384, 10, kernel_size=3, strides=1, padding=1),  # 标签类别数是10
    nn.AdaptiveAvgPool2d((1, 1)),
    nn.Flatten())  # 将四维的输出转成二维的输出，其形状为(批量大小,10)

查看每个块的输出形状

In [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, 96, 54, 54])
MaxPool2d output shape:	 torch.Size([1, 96, 26, 26])
Sequential output shape:	 torch.Size([1, 256, 26, 26])
MaxPool2d output shape:	 torch.Size([1, 256, 12, 12])
Sequential output shape:	 torch.Size([1, 384, 12, 12])
MaxPool2d output shape:	 torch.Size([1, 384, 5, 5])
Dropout output shape:	 torch.Size([1, 384, 5, 5])
Sequential output shape:	 torch.Size([1, 10, 5, 5])
AdaptiveAvgPool2d output shape:	 torch.Size([1, 10, 1, 1])
Flatten output shape:	 torch.Size([1, 10])

训练模型

In [5]:

lr, num_epochs, batch_size = 0.1, 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 2.303, train acc 0.100, test acc 0.100
1445.5 examples/sec on cuda:0

注意：这是典型的训练失败案例。

• 思考：如何（调参以实现）提高准确性？

含并行连结的网络（GoogLeNet）

In [1]:

import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l

Inception块

In [2]:

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)

GoogLeNet模型

In [6]:

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))

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))

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))

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))

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))

为了使Fashion-MNIST上的训练短小精悍，我们将输入的高和宽从224降到96

In [7]:

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])

训练模型

In [8]:

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 0.244, train acc 0.908, test acc 0.896
3490.2 examples/sec on cuda:0

批量规范化

从零实现

In [1]:

import torch
from torch import nn
from d2l import torch as d2l

In [2]:

def batch_norm(X, gamma, beta, moving_mean, moving_var, eps, momentum):
    if not torch.is_grad_enabled(): # 判断当前模式是训练模式还是预测模式
        # 如果是在预测模式下，直接使用传入的移动平均所得的均值和方差
        X_hat = (X - moving_mean) / torch.sqrt(moving_var + eps)
    else:
        assert len(X.shape) in (2, 4)
        if len(X.shape) == 2: # 使用全连接层的情况，计算特征维上的均值和方差
            mean = X.mean(dim=0)
            var = ((X - mean) ** 2).mean(dim=0)
        else: # 使用二维卷积层的情况，计算通道维上（axis=1）的均值和方差
            # 这里我们需要保持X的形状以便后面可以做广播运算
            mean = X.mean(dim=(0, 2, 3), keepdim=True)
            var = ((X - mean) ** 2).mean(dim=(0, 2, 3), keepdim=True)
        # 训练模式下，用当前的均值和方差做标准化
        X_hat = (X - mean) / torch.sqrt(var + eps)
        # 更新移动平均的均值和方差
        moving_mean = momentum * moving_mean + (1.0 - momentum) * mean
        moving_var = momentum * moving_var + (1.0 - momentum) * var
    Y = gamma * X_hat + beta # 缩放和移位
    return Y, moving_mean.data, moving_var.data

创建一个正确的BatchNorm层

In [3]:

class BatchNorm(nn.Module):
    def __init__(self, num_features, num_dims):
        # num_features：全连接层的输出数量或卷积层的输出通道数。
        # num_dims：2表示全连接层，4表示卷积层
        super().__init__()
        if num_dims == 2:
            shape = (1, num_features)
        else:
            shape = (1, num_features, 1, 1)
        # 参与求梯度和迭代的拉伸和偏移参数，分别初始化成1和0
        self.gamma = nn.Parameter(torch.ones(shape))
        self.beta = nn.Parameter(torch.zeros(shape))
        # 非模型参数：均值、方差
        self.moving_mean = torch.zeros(shape)
        self.moving_var = torch.ones(shape)

    def forward(self, X):
        if self.moving_mean.device != X.device:
            # 如果X不在内存上，将moving_mean和moving_var复制到X所在显存上
            self.moving_mean = self.moving_mean.to(X.device)
            self.moving_var = self.moving_var.to(X.device)
        # 保存更新过的moving_mean和moving_var
        Y, self.moving_mean, self.moving_var = batch_norm(
            X, self.gamma, self.beta, self.moving_mean,
            self.moving_var, eps=1e-5, momentum=0.9)
        return Y

应用BatchNorm 于LeNet模型

In [4]:

net = nn.Sequential(
    nn.Conv2d(1, 6, kernel_size=5), BatchNorm(6, num_dims=4), nn.Sigmoid(),
    nn.AvgPool2d(kernel_size=2, stride=2),
    nn.Conv2d(6, 16, kernel_size=5), BatchNorm(16, num_dims=4), nn.Sigmoid(),
    nn.AvgPool2d(kernel_size=2, stride=2), nn.Flatten(),
    nn.Linear(16*4*4, 120), BatchNorm(120, num_dims=2), nn.Sigmoid(),
    nn.Linear(120, 84), BatchNorm(84, num_dims=2), nn.Sigmoid(),
    nn.Linear(84, 10))

在Fashion-MNIST数据集上训练网络。注意：学习率大很多

In [5]:

lr, num_epochs, batch_size = 1.0, 10, 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())

loss 0.263, train acc 0.902, test acc 0.857
31378.5 examples/sec on cuda:0

拉伸参数gamma和偏移参数beta

In [6]:

net[1].gamma.reshape((-1,)), net[1].beta.reshape((-1,))

Out [6]:

(tensor([2.9182, 3.9908, 2.5709, 2.1960, 3.0324, 2.0170], device='cuda:0',
        grad_fn=<ReshapeAliasBackward0>),
 tensor([-0.4980,  2.6303, -2.3259, -1.4185, -3.2187, -2.1353], device='cuda:0',
        grad_fn=<ReshapeAliasBackward0>))

简明实现

In [7]:

net = nn.Sequential(
    nn.Conv2d(1, 6, kernel_size=5), nn.BatchNorm2d(6), nn.Sigmoid(),
    nn.AvgPool2d(kernel_size=2, stride=2),
    nn.Conv2d(6, 16, kernel_size=5), nn.BatchNorm2d(16), nn.Sigmoid(),
    nn.AvgPool2d(kernel_size=2, stride=2), nn.Flatten(),
    nn.Linear(256, 120), nn.BatchNorm1d(120), nn.Sigmoid(),
    nn.Linear(120, 84), nn.BatchNorm1d(84), nn.Sigmoid(),
    nn.Linear(84, 10))

使用相同超参数来训练模型。注意：通常高级API运行速度快得多，因为它的代码已编译为C++或CUDA

In [8]:

d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())

loss 0.281, train acc 0.896, test acc 0.843
55970.4 examples/sec on cuda:0

残差网络（ResNet）

残差块

In [1]:

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)

输入和输出形状一致

In [2]:

blk = Residual(3,3)
X = torch.rand(4, 3, 6, 6)
Y = blk(X)
Y.shape

Out [2]:

torch.Size([4, 3, 6, 6])

增加输出通道数的同时，减半输出的高和宽

In [3]:

blk = Residual(3,6, use_1x1conv=True, strides=2)
blk(X).shape

Out [3]:

torch.Size([4, 6, 3, 3])

ResNet模型

In [7]:

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))

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

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))

net = nn.Sequential(b1, b2, b3, b4, b5,
                    nn.AdaptiveAvgPool2d((1,1)),
                    nn.Flatten(), nn.Linear(512, 10))

观察一下ResNet中不同模块的输入形状是如何变化的

In [8]:

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])

训练模型

In [9]:

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 0.009, train acc 0.998, test acc 0.922
4702.7 examples/sec on cuda:0