先 使用卷积层来学习图片空间信息,然后使用全连接层来转换到类别空间。
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 = nn.Sequential(
Reshape(),
nn.Conv2d(1, 6, kernel_size=5, padding=2), nn.Sigmoid(),
nn.AvgPool2d(kernel_size=2, stride=2),
nn.Conv2d(6, 16, kernel_size=5), nn.Sigmoid(),
nn.AvgPool2d(kernel_size=2, stride=2),
nn.Flatten(),
nn.Linear(16 * 5 * 5, 120), nn.Sigmoid(),
nn.Linear(120, 84), nn.Sigmoid(),
nn.Linear(84, 10))
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)
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]
#@save
def train_ch6(net, train_iter, test_iter, num_epochs, lr, device):
"""用GPU训练模型"""
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):
# 训练损失之和,训练准确率之和,样本数
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)}')
lr, num_epochs = 0.9, 10
train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
特点:
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。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。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))
X = torch.randn(1, 1, 224, 224)
for layer in net:
X=layer(X)
print(layer.__class__.__name__,'output shape:\t',X.shape)
batch_size = 128
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=224)
lr, num_epochs = 0.01, 10
d2l.train_ch6(net, train_iter, test_iter, num_epochs, lr, d2l.try_gpu())
经典卷积神经网络组成:1.带填充以保持分辨率的卷积层 2.非线性激活函数,如relu 3.池化层
VGG块与之类似
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)
def vgg(conv_arch):
conv_blks = []
in_channels = 1
# 卷积层部分
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(),
# 全连接层部分
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)
X = torch.randn(size=(1, 1, 224, 224))
for blk in net:
X = blk(X)
print(blk.__class__.__name__,'output shape:\t',X.shape)
ratio = 4
small_conv_arch = [(pair[0], pair[1] // ratio) for pair in conv_arch]
net = vgg(small_conv_arch)
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())
LeNet、AlexNet和VGG的共同特点是都使用了全连接层,但全连接层需要过多的参数量,会导致占用过多内存、计算带宽、容易过拟合。
卷积层参数:(较少)
卷积层后第一个全连接层参数:LeNet 16*5*5*120=48k AlexNet 256*5*5*4096=26M
VGG 512*7*7*4096=102M
NiN简单的解决方案:在每个像素的通道上分别使用多层感知机
一个卷积层后跟两个全连接层,步幅1,无填充,输出形状跟卷积层输出一样,起到全连接层的作用。作用就是对通道数进行混合
无全连接层,交替使用NiN块和步幅为2的最大池化层,逐步减小高宽和增大通道数;最后使用全局平均池化层得到输出,其输出就是类别数。
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())
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),
# 标签类别数是10
nin_block(384, 10, kernel_size=3, strides=1, padding=1),
nn.AdaptiveAvgPool2d((1, 1)),
# 将四维的输出转成二维的输出,其形状为(批量大小,10)
nn.Flatten())
X = torch.rand(size=(1, 1, 224, 224))
for layer in net:
X = layer(X)
print(layer.__class__.__name__,'output shape:\t', X.shape)
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())
吸收了NiN中串联网络的思想,并在此基础上做了改进。 重点是解决了什么样大小的卷积核最合适的问题。 毕竟,以前流行的网络使用小到1×1,大到11×11的卷积核。 本文的一个观点是,有时使用不同大小的卷积核组合是有利的。
由四条并行路径组成。 前三条路径使用窗口大小为1×1、3×3和5×5的卷积层,从不同空间大小中提取信息。 中间的两条路径在输入上执行1×1卷积,以减少通道数,从而降低模型的复杂性。 第四条路径使用3×3最大汇聚层,然后使用1×1卷积层来改变通道数。 这四条路径都使用合适的填充来使输入与输出的高和宽一致,最后我们将每条线路的输出在通道维度上连结,并构成Inception块的输出。在Inception块中,通常调整的超参数是每层输出通道数。(白色用来改变通道数,蓝色用来抽取信息,不抽取通道信息,只抽取空间信息)
一共使用9个Inception块和全局平均汇聚层的堆叠来生成其估计值。Inception块之间的最大汇聚层可降低维度。 第一个模块类似于AlexNet和LeNet,Inception块的组合从VGG继承,全局平均汇聚层避免了在最后使用全连接层。
损失出现在最后,后面的层训练较快;数据在最底部,底部层训练较慢,底部层一变化,所有都得跟着变,最后的那些层需要重新学习多次,导致收敛变慢。批量归一化解决在学习底部层的时候避免变化顶部。
方法:固定小批量里的均值和方差
然后再做额外的调整(可学习的参数)
可学习的参数为和,作用在全连接层和卷积层输出上,激活函数前或全连接层和卷积层输入上。对全连接层,作用在特征维(如2维时,每一列是一个特征);对于卷积层,作用在通道维。
总结:批量归一化固定小批量中的均值和方差,然后学习出适合的偏移和缩放
可以加速收敛速度(允许用更大的学习率),但一般不改变模型精度
import torch
from torch import nn
from d2l import torch as d2l
def batch_norm(X, gamma, beta, moving_mean, moving_var, eps, momentum):
# 通过is_grad_enabled来判断当前模式是训练模式还是预测模式
if not torch.is_grad_enabled():
# 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
class BatchNorm(nn.Module):
# num_features:完全连接层的输出数量或卷积层的输出通道数。
# num_dims:2表示完全连接层,4表示卷积层
def __init__(self, num_features, num_dims):
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))
# 非模型参数的变量初始化为0和1
self.moving_mean = torch.zeros(shape)
self.moving_var = torch.ones(shape)
def forward(self, X):
# 如果X不在内存上,将moving_mean和moving_var
# 复制到X所在显存上
if self.moving_mean.device != X.device:
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
残差网络的核心思想是:每个附加层都应该更容易地包含原始函数作为其元素之一。
残差块里首先有2个有相同输出通道数的3×3卷积层。 每个卷积层后接一个批量规范化层和ReLU激活函数。 然后我们通过跨层数据通路,跳过这2个卷积运算,将输入直接加在最后的ReLU激活函数前。 这样的设计要求2个卷积层的输出与输入形状一样,从而使它们可以相加。 如果想改变通道数,就需要引入一个额外的1×1卷积层来将输入变换成需要的形状后再做相加运算。
import torch
from torch import nn
from torch.nn import functional as F
from d2l import torch as d2l
class Residual(nn.Module): #@save
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)
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
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))
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))
X = torch.rand(size=(1, 1, 224, 224))
for layer in net:
X = layer(X)
print(layer.__class__.__name__,'output shape:\t', X.shape)
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())