import torch
import torch.nn as nn
import matplotlib.pyplot as plt
import torchvision
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device
使用dataset下载CIFAR10数据集,并划分好训练集与测试集。
使用dataloader加载数据,并设置好基本的batch_size
train_ds = torchvision.datasets.CIFAR10('data',
train=True,
transform=torchvision.transforms.ToTensor(), # 将数据类型转化为Tensor
download=True)
test_ds = torchvision.datasets.CIFAR10('data',
train=False,
transform=torchvision.transforms.ToTensor(), # 将数据类型转化为Tensor
download=True)
batch_size = 32
train_dl = torch.utils.data.DataLoader(train_ds,
batch_size=batch_size,
shuffle=True)
test_dl = torch.utils.data.DataLoader(test_ds,
batch_size=batch_size)
# 取一个批次查看数据格式
# 数据的shape为:[batch_size, channel, height, weight]
# 其中batch_size为自己设定,channel,height和weight分别是图片的通道数,高度和宽度。
imgs, labels = next(iter(train_dl))
imgs.shape
torch.Size([32, 3, 32, 32])
import numpy as np
# 指定图片大小,图像大小为20宽、5高的绘图(单位为英寸inch)
plt.figure(figsize=(20, 5))
for i, imgs in enumerate(imgs[:20]):
# 维度缩减
npimg = imgs.numpy().transpose((1,2,0))
# 将整个figure分成2行10列,绘制第i+1个子图。
plt.subplot(2, 10, i+1)
plt.imshow(npimg, cmap=plt.cm.binary)
plt.axis('off')
对于一般的CNN网络来说,都是由特征提取网络和分类网络构成,其中特征提取网络用于提取图片的特征,分类网络用于将图片进行分类。
import torch.nn.functional as F
num_classes = 10 # 图片的类别数
class Model(nn.Module):
def __init__(self):
super().__init__()
# 特征提取网络
self.conv1 = nn.Conv2d(3, 64, kernel_size=3)
self.pool1 = nn.MaxPool2d(2)
self.conv2 = nn.Conv2d(64, 64, kernel_size=3)
self.pool2 = nn.MaxPool2d(2)
self.conv3 = nn.Conv2d(64, 128, kernel_size=3)
self.pool3 = nn.MaxPool2d(2)
# 分类网络
self.fc1 = nn.Linear(512, 256)
self.fc2 = nn.Linear(256, num_classes)
# 前向传播
def forward(self, x):
x = self.pool1(F.relu(self.conv1(x)))
x = self.pool2(F.relu(self.conv2(x)))
x = self.pool3(F.relu(self.conv3(x)))
x = torch.flatten(x, start_dim=1)
x = F.relu(self.fc1(x))
x = self.fc2(x)
return x
打印并加载模型
from torchinfo import summary
# 将模型转移到GPU中(我们模型运行均在GPU中进行)
model = Model().to(device)
summary(model)
================================================================= Layer (type:depth-idx) Param # ================================================================= Model -- ├─Conv2d: 1-1 1,792 ├─MaxPool2d: 1-2 -- ├─Conv2d: 1-3 36,928 ├─MaxPool2d: 1-4 -- ├─Conv2d: 1-5 73,856 ├─MaxPool2d: 1-6 -- ├─Linear: 1-7 131,328 ├─Linear: 1-8 2,570 ================================================================= Total params: 246,474 Trainable params: 246,474 Non-trainable params: 0 =================================================================
loss_fn = nn.CrossEntropyLoss() # 创建损失函数
learn_rate = 1e-2 # 学习率
opt = torch.optim.SGD(model.parameters(),lr=learn_rate)
1. optimizer.zero_grad()
函数会遍历模型的所有参数,通过内置方法截断反向传播的梯度流,再将每个参数的梯度值设为0,即上一次的梯度记录被清空。
2. loss.backward()
PyTorch的反向传播(即tensor.backward())是通过autograd包来实现的,autograd包会根据tensor进行过的数学运算来自动计算其对应的梯度。
具体来说,torch.tensor是autograd包的基础类,如果你设置tensor的requires_grads为True,就会开始跟踪这个tensor上面的所有运算,如果你做完运算后使用tensor.backward(),所有的梯度就会自动运算,tensor的梯度将会累加到它的.grad属性里面去。
更具体地说,损失函数loss是由模型的所有权重w经过一系列运算得到的,若某个w的requires_grads为True,则w的所有上层参数(后面层的权重w)的.grad_fn属性中就保存了对应的运算,然后在使用loss.backward()后,会一层层的反向传播计算每个w的梯度值,并保存到该w的.grad属性中。
如果没有进行tensor.backward()的话,梯度值将会是None,因此loss.backward()要写在optimizer.step()之前。
3. optimizer.step()
step()函数的作用是执行一次优化步骤,通过梯度下降法来更新参数的值。因为梯度下降是基于梯度的,所以在执行optimizer.step()函数前应先执行loss.backward()函数来计算梯度。
注意:optimizer只负责通过梯度下降进行优化,而不负责产生梯度,梯度是tensor.backward()方法产生的。
# 训练循环
def train(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset) # 训练集的大小,一共60000张图片
num_batches = len(dataloader) # 批次数目,1875(60000/32)
train_loss, train_acc = 0, 0 # 初始化训练损失和正确率
for X, y in dataloader: # 获取图片及其标签
X, y = X.to(device), y.to(device)
# 计算预测误差
pred = model(X) # 网络输出
loss = loss_fn(pred, y) # 计算网络输出和真实值之间的差距,targets为真实值,计算二者差值即为损失
# 反向传播
optimizer.zero_grad() # grad属性归零
loss.backward() # 反向传播
optimizer.step() # 每一步自动更新
# 记录acc与loss
train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
train_loss += loss.item()
train_acc /= size
train_loss /= num_batches
return train_acc, train_loss
测试函数和训练函数大致相同,但是由于不进行梯度下降对网络权重进行更新,所以不需要传入优化器
def test (dataloader, model, loss_fn):
size = len(dataloader.dataset) # 测试集的大小,一共10000张图片
num_batches = len(dataloader) # 批次数目,313(10000/32=312.5,向上取整)
test_loss, test_acc = 0, 0
# 当不进行训练时,停止梯度更新,节省计算内存消耗
with torch.no_grad():
for imgs, target in dataloader:
imgs, target = imgs.to(device), target.to(device)
# 计算loss
target_pred = model(imgs)
loss = loss_fn(target_pred, target)
test_loss += loss.item()
test_acc += (target_pred.argmax(1) == target).type(torch.float).sum().item()
test_acc /= size
test_loss /= num_batches
return test_acc, test_loss
1. model.train()
model.train()的作用是启用 Batch Normalization 和 Dropout。
如果模型中有BN层(Batch Normalization)和Dropout,需要在训练时添加model.train()。model.train()是保证BN层能够用到每一批数据的均值和方差。对于Dropout,model.train()是随机取一部分网络连接来训练更新参数。
2. model.eval()
model.eval()的作用是不启用 Batch Normalization 和 Dropout。
如果模型中有BN层(Batch Normalization)和Dropout,在测试时添加model.eval()。model.eval()是保证BN层能够用全部训练数据的均值和方差,即测试过程中要保证BN层的均值和方差不变。对于Dropout,model.eval()是利用到了所有网络连接,即不进行随机舍弃神经元。
训练完train样本后,生成的模型model要用来测试样本。在model(test)之前,需要加上model.eval(),否则的话,有输入数据,即使不训练,它也会改变权值。这是model中含有BN层和Dropout所带来的的性质。
epochs = 10
train_loss = []
train_acc = []
test_loss = []
test_acc = []
for epoch in range(epochs):
model.train()
epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, opt)
model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
train_acc.append(epoch_train_acc)
train_loss.append(epoch_train_loss)
test_acc.append(epoch_test_acc)
test_loss.append(epoch_test_loss)
template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%,Test_loss:{:.3f}')
print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss))
print('Done')
Epoch: 1, Train_acc:12.3%, Train_loss:2.292, Test_acc:16.8%,Test_loss:2.244 Epoch: 2, Train_acc:24.0%, Train_loss:2.056, Test_acc:28.7%,Test_loss:1.933 Epoch: 3, Train_acc:32.1%, Train_loss:1.852, Test_acc:38.2%,Test_loss:1.709 Epoch: 4, Train_acc:39.9%, Train_loss:1.654, Test_acc:43.3%,Test_loss:1.567 Epoch: 5, Train_acc:44.2%, Train_loss:1.541, Test_acc:45.3%,Test_loss:1.494 Epoch: 6, Train_acc:47.5%, Train_loss:1.450, Test_acc:48.5%,Test_loss:1.420 Epoch: 7, Train_acc:50.9%, Train_loss:1.368, Test_acc:52.5%,Test_loss:1.343 Epoch: 8, Train_acc:53.6%, Train_loss:1.298, Test_acc:54.0%,Test_loss:1.293 Epoch: 9, Train_acc:56.3%, Train_loss:1.234, Test_acc:56.3%,Test_loss:1.241 Epoch:10, Train_acc:58.3%, Train_loss:1.182, Test_acc:57.5%,Test_loss:1.200 Done
import matplotlib.pyplot as plt
#隐藏警告
import warnings
warnings.filterwarnings("ignore") #忽略警告信息
plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号
plt.rcParams['figure.dpi'] = 100 #分辨率
epochs_range = range(epochs)
plt.figure(figsize=(12, 3))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, train_acc, label='Training Accuracy')
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, label='Training Loss')
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()