import torch
import torchvision
from torch.utils.tensorboard import SummaryWriter
from modle import *
#modle文件要和当前文件在一个文件夹下
from torch import nn
from torch.utils.data import DataLoader
#准备数据集
train_data = torchvision.datasets.CIFAR10("./dataset1", train=True, download=True,
transform=torchvision.transforms.ToTensor())
test_data = torchvision.datasets.CIFAR10("./dataset1", train=False, download=True,
transform=torchvision.transforms.ToTensor())
#length 长度
train_data_size = len(train_data)
test_data_size = len(test_data)
#如果train_data_size=10,训练数据集的长度为:10
print("训练数据集的长度:{}".format(train_data_size))#在后面按住ctrl+D,可以在下面直接进行复制。
print("测试数据集的长度:{}".format(test_data_size))
#利用Dataloader来加载数据集
train_dataloader = DataLoader(train_data, batch_size=64)
test_dataloader = DataLoader(test_data, batch_size=64)
#创建网络模型
feng = Feng()
#损失函数
loss_fn = nn.CrossEntropyLoss()
#优化器
#learning_rate = 0.01
#1e-2=1*(10)*(-2)=1/100=0.01
learning_rate = 1e-2
optimizer = torch.optim.SGD(feng.parameters(), lr=learning_rate)
#设置训练网络的一些参数
#记录训练的次数
total_train_step = 0
#记录测试的次数
total_test_step = 0
#训练的轮数
epoch = 10
#添加tensorboard
writer = SummaryWriter("../logs_train")
for i in range(epoch):
print("--------第{}轮训练开始------".format(i+1))
#训练步骤开始
for data in train_dataloader:
imgs, targets = data
outputs = feng(imgs)
loss = loss_fn(outputs, targets)
#优化器梯度清零
optimizer.zero_grad()
#反向传播,得到每个参数节点的梯度
loss.backward()
#调用优化器,对梯度参数进行优化
optimizer.step()
total_train_step = total_train_step + 1
#整体的步骤满100的时候,才打印
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)
#测试步骤开始
total_test_loss = 0
#整体正确的个数为0
total_accuracy = 0
with torch.no_grad():#网络模型的梯度都去除
for data in test_dataloader:
imgs, targets = data
outputs = feng(imgs)
loss = loss_fn(outputs, targets)#计算loss误差
total_test_loss = total_test_loss + loss.item()
#每小部分正确的个数
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_test_step = total_test_step + 1
#保存每一轮的训练结果
torch.save(feng, "feng_{}.path".format(i))
print("模型已保存")
writer.close()