【PyTorch笔记】pytorch入门教程15 模型训练套路

一般我们会把模型单独保存成一个py文件。然后在另一个文件操作。

模型文件：

import torch
from torch import nn


class Tudui(nn.Module):
    def __init__(self):
        super().__init__()
        self.model = nn.Sequential(
            nn.Conv2d(3,32,5,1,2),
            nn.MaxPool2d(2),
            nn.Conv2d(32,32,5,1,2),
            nn.MaxPool2d(2),
            nn.Conv2d(32,64,5,1,2),
            nn.MaxPool2d(2),
            nn.Flatten(),
            nn.Linear(64*4*4,64),
            nn.Linear(64,10)
        )

    def forward(self,x):
        x = self.model(x)
        return x

if __name__ == '__main__':
    # 上面这句代码的作用是，只有当该模块作为主程序运行时，才会执行下面的代码，而当该模块被作为一个模块被导入到其他程序中时，这段代码不会被执行。这是为了避免在导入该模块时就执行了一些不必要的代码。在此py中运行可以查看模型有没有错误，输出是不是自己想要的结果
    tudui = Tudui()
    input = torch.ones((64,3,32,32))
    output = tudui(input)
    print(output.shape)

训练套路如下：

import torchvision
from tensorboardX import SummaryWriter
from torch import nn
from torch.utils.data import DataLoader
from model import * #注意model和当前此文件必须在同一个文件夹下
train_data = torchvision.datasets.CIFAR10('./dataset',train = True,transform=torchvision.transforms.ToTensor())
test_data = torchvision.datasets.CIFAR10('./dataset',train = False,transform=torchvision.transforms.ToTensor())
train_data_size = len(train_data)
test_data_size = len(test_data)
print('训练集长度{}'.format(train_data_size))
print('测试集长度{}'.format(test_data_size))

train_dataloader = DataLoader(train_data,batch_size=64)
test_dataloader = DataLoader(test_data,batch_size=64)

tudui = Tudui()

loss_fn = nn.CrossEntropyLoss()

lr = 0.01
#也可以写成1e-2，都是为了后面方便修改
optimizer = torch.optim.SGD(tudui.parameters(),lr=lr)



total_train_step = 0
total_test_step = 0
epoch = 1

writer = SummaryWriter('logs_train')

for i in range(epoch):
    print('---------第{}轮训练开始--------'.format(i+1))

    # tudui.train() 网络里有BN或者dropout层的时候要有这句话
    for imgs,targets in train_dataloader:
        outputs = tudui(imgs)
        loss = loss_fn(outputs,targets)

        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        total_train_step += 1
        if total_train_step % 100 == 0:
            print('训练次数：{},Loss:{}'.format(total_train_step,loss.item()))
            writer.add_scalar('train_loss',loss.item(),total_train_step)
#test步骤：
    # tudui.eval() 网络里有BN或者dropout层的时候要有这句话
    total_test_loss = 0
    total_accuracy = 0

    with torch.no_grad():
        for imgs,targets in test_dataloader:
            outputs = tudui(imgs)
            loss = loss_fn(outputs,targets)
            total_test_loss += loss.item()
            accuracy = (outputs.argmax(1) == targets).sum()
            accuracy = accuracy.item()
            #argmax(1)是以轴1（行）为单位，选出每个单位的最大值，0的话就是从每一列选一个最大值。非假即1，求和是求出预测对的个数
            total_accuracy += accuracy
    print('整体测试集上的Loss:{}'.format(total_test_loss))
    print('整体测试集上的正确率:{}'.format(total_accuracy/test_data_size))
    writer.add_scalar('test_loss',total_test_loss,total_test_step)
    writer.add_scalar('test_accuracy',total_accuracy/test_data_size,total_test_step)
    total_test_step += 1

    torch.save(tudui,'tudui_{}.pth'.format(i))
    #这是保存每一轮训练完的模型
    # torch,save(tudui.state_dict(),'tudui_{}'.format(i))
writer.close()

这里在tensorboard上面看Loss图可能会有两条线，那是因为设置了smooth，把smoothing改成0就好了