pytorch实现加载保存查看checkpoint文件

# 方式一：保存加载整个state_dict（推荐）
# 保存
torch.save(model.state_dict(), PATH)
# 加载
model.load_state_dict(torch.load(PATH))
# 测试时不启用 BatchNormalization 和 Dropout
model.eval()

# 方式二：保存加载整个模型
# 保存
torch.save(model, PATH)
# 加载
model = torch.load(PATH)
model.eval()

# 方式三：保存用于继续训练的checkpoint或者多个模型
# 保存
torch.save({
            'epoch': epoch,
            'model_state_dict': model.state_dict(),
            ...
            }, PATH)
# 加载
checkpoint = torch.load(PATH)
start_epoch=checkpoint['epoch']
model.load_state_dict(checkpoint['model_state_dict'])
# 测试时
model.eval()
# 或者训练时
model.train()

# GPU上保存，CPU上加载
# 保存
torch.save(model.state_dict(), PATH)
# 加载
device = torch.device('cpu')
model.load_state_dict(torch.load(PATH, map_location=device))
# 如果是多gpu保存，需要去除关键字中的module，见第4部分

# GPU上保存，GPU上加载
# 保存
torch.save(model.state_dict(), PATH)
# 加载
device = torch.device("cuda")
model.load_state_dict(torch.load(PATH))
model.to(device)

# CPU上保存，GPU上加载
# 保存
torch.save(model.state_dict(), PATH)
# 加载
device = torch.device("cuda")
# 选择希望使用的GPU
model.load_state_dict(torch.load(PATH, map_location="cuda:0"))  
model.to(device)

# 打印模型的 state_dict
print("Model's state_dict:")
for param_tensor in model.state_dict():
    print(param_tensor, "\t", model.state_dict()[param_tensor].size())

#解决1：加载时将module去掉

# 创建一个不包含`module.`的新OrderedDict
from collections import OrderedDict
new_state_dict = OrderedDict()
for k, v in state_dict.items():
    name = k[7:] # 去掉 `module.`
    new_state_dict[name] = v
# 加载参数
model.load_state_dict(new_state_dict)

# 解决2：保存checkpoint时不保存module
torch.save(model.module.state_dict(), PATH)

'''本文件用于举例说明pytorch保存和加载文件的方法''' 
import torch as torch
import torchvision as tv
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
import torchvision.transforms as transforms
import os
  
# 参数声明
batch_size = 32
epochs = 10
WORKERS = 0  # dataloder线程数
test_flag = False  # 测试标志，True时加载保存好的模型进行测试
ROOT = '/home/pxt/pytorch/cifar'  # MNIST数据集保存路径
log_dir = '/home/pxt/pytorch/logs/cifar_model.pth'  # 模型保存路径
# 加载MNIST数据集
transform = tv.transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize([0.5, 0.5, 0.5], [0.5, 0.5, 0.5])])
 
train_data = tv.datasets.CIFAR10(root=ROOT, train=True, download=True, transform=transform)
test_data = tv.datasets.CIFAR10(root=ROOT, train=False, download=False, transform=transform)
 
train_load = torch.utils.data.DataLoader(train_data, batch_size=batch_size, shuffle=True, num_workers=WORKERS)
test_load = torch.utils.data.DataLoader(test_data, batch_size=batch_size, shuffle=False, num_workers=WORKERS)
 
 
# 构造模型
class Net(nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = nn.Conv2d(3, 64, 3, padding=1)
        self.conv2 = nn.Conv2d(64, 128, 3, padding=1)
        self.conv3 = nn.Conv2d(128, 256, 3, padding=1)
        self.conv4 = nn.Conv2d(256, 256, 3, padding=1)
        self.pool = nn.MaxPool2d(2, 2)
        self.fc1 = nn.Linear(256 * 8 * 8, 1024)
        self.fc2 = nn.Linear(1024, 256)
        self.fc3 = nn.Linear(256, 10)
 
    def forward(self, x):
        x = F.relu(self.conv1(x))
        x = self.pool(F.relu(self.conv2(x)))
        x = F.relu(self.conv3(x))
        x = self.pool(F.relu(self.conv4(x)))
        x = x.view(-1, x.size()[1] * x.size()[2] * x.size()[3])
        x = F.relu(self.fc1(x))
        x = F.relu(self.fc2(x))
        x = self.fc3(x)
        return x
  
model = Net().cpu()
 
criterion = nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.01)
 
 
# 模型训练
def train(model, train_loader, epoch):
    model.train()
    train_loss = 0
    for i, data in enumerate(train_loader, 0):
        x, y = data
        x = x.cpu()
        y = y.cpu()
 
        optimizer.zero_grad()
        y_hat = model(x)
        loss = criterion(y_hat, y)
        loss.backward()
        optimizer.step()
        train_loss += loss
        print('正在进行第{}个epoch中的第{}次循环'.format(epoch,i))
 
    loss_mean = train_loss / (i + 1)
    print('Train Epoch: {}\t Loss: {:.6f}'.format(epoch, loss_mean.item()))
 
 
# 模型测试
def test(model, test_loader):
    model.eval()
    test_loss = 0
    correct = 0
    with torch.no_grad():
        for i, data in enumerate(test_loader, 0):
            x, y = data
            x = x.cpu()
            y = y.cpu()
 
            optimizer.zero_grad()
            y_hat = model(x)
            test_loss += criterion(y_hat, y).item()
            pred = y_hat.max(1, keepdim=True)[1]
            correct += pred.eq(y.view_as(pred)).sum().item()
        test_loss /= (i + 1)
        print('Test set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)\n'.format(
            test_loss, correct, len(test_data), 100. * correct / len(test_data)))
  
def main():
    # 如果test_flag=True,则加载已保存的模型并进行测试，测试以后不进行此模块以后的步骤
    if test_flag:
        # 加载保存的模型直接进行测试机验证
        checkpoint = torch.load(log_dir)
        model.load_state_dict(checkpoint['model'])
        optimizer.load_state_dict(checkpoint['optimizer'])
        start_epoch = checkpoint['epoch']
        test(model, test_load)
        return
 
    # 如果有保存的模型，则加载模型，并在其基础上继续训练
    if os.path.exists(log_dir):
        checkpoint = torch.load(log_dir)
        model.load_state_dict(checkpoint['model'])
        optimizer.load_state_dict(checkpoint['optimizer'])
        start_epoch = checkpoint['epoch']
        print('加载 epoch {} 成功！'.format(start_epoch))
    else:
        start_epoch = 0
        print('无保存了的模型，将从头开始训练！')
 
    for epoch in range(start_epoch+1, epochs):
        train(model, train_load, epoch)
        test(model, test_load)
        # 保存模型
        state = {'model':model.state_dict(), 'optimizer':optimizer.state_dict(), 'epoch':epoch}
        torch.save(state, log_dir)
 
if __name__ == '__main__':
    main()