Pytorch教程[10]完整模型训练套路

一般的模型构建都是按照下图这样的流程

下面分享一个自己手动搭建的网络

from model import *
import torchvision
import torch
from torch.utils.tensorboard import SummaryWriter
from torchvision import transforms
from torch import nn
from torch.utils.data import DataLoader

#数据增强
data_transforms = transforms.Compose([
        transforms.RandomRotation(45),
        transforms.ToTensor(),
    ])

#准备数据集
#train_data = torchvision.datasets.CIFAR10(root="D:\pythonProject_pytorchstudy", train=True, transform=torchvision.transforms.ToTensor(), download=False)
#test_data = torchvision.datasets.CIFAR10(root="D:\pythonProject_pytorchstudy", train=False, transform=torchvision.transforms.ToTensor(), download=False)
train_data = torchvision.datasets.CIFAR10(root="D:\pythonProject_pytorchstudy", train=True, transform=data_transforms, download=False)
test_data = torchvision.datasets.CIFAR10(root="D:\pythonProject_pytorchstudy", train=False, transform=torchvision.transforms.ToTensor(), download=False)



#数据集长度
train_data_size = len(train_data)
test_data_size = len(test_data)
print("训练集的长度为:{}".format(train_data_size))
print("测试集的长度为:{}".format(test_data_size))



#利用Dataloader加载数据集
train_dataloader =DataLoader(train_data,batch_size=64)
test_dataloader =DataLoader(test_data,batch_size=64)

#搭建神经网络
#model.py

#创建网络模型
Yolo = My_Model()

################################
if torch.cuda.is_available():  #
    Yolo = My_Model().cuda()   #
################################

#损失函数
loss_fn = nn.CrossEntropyLoss()

################################
if torch.cuda.is_available():  #
    loss_fn = loss_fn.cuda()   #
################################

#优化器
learning_rate = 0.01 #1e-2 = 1 x (10)^(-2) =1/100 =0.01
optimizer  = torch.optim.SGD(Yolo.parameters(), lr = learning_rate, )


#设置训练网络的参数
total_train_step = 0
#记录测试次数
total_test_step = 0
#训练轮数
epoch = 10

#添加tensorboard
writer = SummaryWriter("D:\pythonProject_pytorchstudy\cifar-10-batches-py\logs_train")


for i in range(epoch):
    print("第{}轮训练开始".format(i+1))
    #训练步骤开始
    Yolo.train()
    for data in train_dataloader:
        imgs,targets = data
        ################################
        if torch.cuda.is_available():  #
            imgs = imgs.cuda()         #
            targets = targets.cuda()   #
        ################################
        outputs = Yolo(imgs)
        loss  = loss_fn(outputs,targets)

        optimizer.zero_grad()
        loss.backward()
        optimizer.step()
        total_train_step += 1

        if total_train_step % 30 ==0:
            print("Iteration：{},loss:{}".format(total_train_step,loss.item()))
            writer.add_scalar("train_loss", loss.item(),total_train_step)
    #测试步骤开始
    Yolo.eval()
    total_test_loss = 0
    total_accuracy = 0
    with torch.no_grad(): #让网络中的梯度没有
        for data in test_dataloader:
            imgs, targets = data
            ################################
            if torch.cuda.is_available():  #
                imgs = imgs.cuda()         #
                targets = targets.cuda()   #
            ################################
            outputs = Yolo(imgs)
            loss = loss_fn(outputs,targets)
            total_test_loss = total_test_loss + loss.item()
            accuracy = (outputs.argmax(1) == targets).sum()
            total_accuracy = 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_train_step += 1

    torch.save(Yolo,"YOLO_{}".format(i+1))
    #torch.save(Yolo.state_dict(),"Yolo_{}.pth".format(i+1))
    print("模型已保存")

writer.close()

import torch
from torch import nn


class My_Model(nn.Module):
    def __init__(self):
        super(My_Model, self).__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

    # Yolo = My_Model()
    # input = torch.ones(64,3,32,32)
    # output = Yolo(input)
    # print(output.shape)