代码示例转自: pytorch tips:使pytorch代码在不改动情况在有GPU自动在GPU运行
1. 定义训练方式:
方法一:
device=torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model=model.to(device)
x=x.to(device)
y=y.to(device)
方法二:
model=model.cuda()
x=x.cuda()
y=y.cuda()
推荐使用方法一:
#gpu or not
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
2. 训练代码相应修改:
import torch
import torch.nn as nn
import torch.optim as optim
import torchvision
import torchvision.transforms as transforms
import argparse
import torch.utils.data
from resnet import ResNet18
#gpu or not
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
parser = argparse.ArgumentParser(description='Python CIFAR10 Training')
parser.add_argument('--outf', default='./model', help='folder to output images and model checkpoints')
parser.add_argument('--net', default=',.model/Resnet18.pth', help='path to net(to continue training)')
args = parser.parse_args()
#超参数设置
EPOCH = 135
pre_epoch = 0
BATCH_SIZE = 128
LR = 0.1
#准备数据集并预处理
transform_train = transforms.Compose([
transforms.RandomCrop(size=32, padding=4), #先padding,再随机截取32*32
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010)),#R,G,B每层的归一化用到的均值和方差
])
transform_test = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.4914, 0.4822, 0.4465), (0.2023, 0.1994, 0.2010)),
])
trianset = torchvision.datasets.CIFAR10(root='./data', train=True, download=True, transform=transform_train) #训练数据集
trainloader = torch.utils.data.DataLoader(dataset=trianset, batch_size=BATCH_SIZE, shuffle=True, num_workers=2) #生成一个个batch进行批训练,顺序打乱
testset = torchvision.datasets.CIFAR10(root='./data', train=False, download=True,transform=transform_test) #测试数据集
testloader = torch.utils.data.DataLoader(dataset=testset, batch_size=100,shuffle = False, num_workers=2)
classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck')
#模型定义ResNet
net = ResNet18().to(device)
#定义损失函数和优化方法
criterion = nn.CrossEntropyLoss() #损失函数为交叉熵,多用于多分类问题
optimizer = optim.SGD(net.parameters(), lr=LR, momentum=0.9, weight_decay=5e-4)
#训练
if __name__ == "__main__":
best_acc = 85 #2 初始化best test accuracy
print("Start Training Resnet-18!")
with open("acc.txt", "w") as f:
with open("log.txt", "w") as f2:
for epoch in range(pre_epoch, EPOCH):
print('\nEpoch: %d' % (epoch+1))
net.train()
sum_loss=0
correct = 0
total = 0
for i,data in enumerate(trainloader, 0):
#准备数据
length = len(trainloader)
inputs, labels = data
################################
数据这里进行了修改
################################
inputs, labels = inputs.to(device), labels.to(device)
optimizer.zero_grad()
#forward + backward
outputs = net(inputs)
loss = criterion(outputs, labels)
loss.backward()
optimizer.step()
#每训练一个batch打印一次loss和准确率
sum_loss += loss.item()
_,predicted = torch.max(outputs.data,1)
total += labels.size(0)
correct += predicted.eq(labels.data).cpu().sum()
print('[epoch: %d, iter: %d] Loss: %.03f | Acc: %.3f%% '
% (epoch+1, (i+1+epoch*length), sum_loss/(i+1), 100. * correct/ total))
f2.write('%03d %05d |Loss: %.03f | Acc: %.3f%% '
% (epoch+1, (i+1+epoch*length), sum_loss/(i+1), 100. * correct/ total))
f2.write('\n')
f2.flush()
#每次训练完一个epoch测试以下准确率
print("Waiting Test!")
with torch.no_grad():
correct = 0
total = 0
for data in testloader:
net.eval()
images, labels = data
################################
数据这里进行了修改
################################
images, labels = images.to(device), labels.to(device)
outputs = net(images)
#取得最高分的那个类(outputs.data的索引号
_, predicted = torch.max(outputs.data, 1)
total += labels.size(0)
correct += (predicted == labels).sum()
print('测试分类准确率为: %.3f%%' % (100*correct/total))
acc = 100. * correct / total
#将每次测试结果实时写入acc.txt文件中
print('Saving model......')
torch.sava(net.state_dict(), '%s/net_%03d.pth' % (args.outf, epoch+1))
f.write("EPOCH=%03d, Accuracy=.3f%%" % (epoch+1, acc))
f.write('\n')
f.flush()
# 记录最佳测试分类准确率并写入best_acc.txt文件中
if acc > best_acc:
f3 = open("best_acc.txt", "w")
f3.write("EPOCH=%d,best_acc=%.3f%%" % (epoch+1, acc))
f3.close()
best_acc = acc
print("Training Finished, TotalEPOCH=%d" % EPOCH)