在模型训练中，你如何知道模型是否训练得很好，或者它是否满足了你想要的需求？ 因此，当我们完成一轮训练时，通过测试数据集中的损失或准确性来评估我们的模型
不再调整测试数据集中，只使用现有模型进行测试
# 准备数据集 import torch.optim import torchvision from torch import nn from torch.utils.data import DataLoader from model import * train_data = torchvision.datasets.CIFAR10("./dataset", train=True, transform=torchvision.transforms.ToTensor(), download=True) test_data = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor(), download=True) # length train_data_size = len(train_data) test_data_siz = len(test_data) print(f"训练数据集的长度如下：{
train_data_size}") print(f"测试数据集的长度如下：{
test_data_siz}") # 利用DataLoader加载 train_data_loader = DataLoader(train_data, batch_size=64) test_data_loader = DataLoader(test_data, batch_size=64) # 创建网络模型 lixinyu = Lixinyu() # 损失函数 loss_fn = nn.CrossEntropyLoss) # 优化器 # learning_rate = 0.01 learning_rate = 1e-2 optimizer = torch.optim.SGD(lixinyu.parameters(), learning_rate) # 设置训练网络参数 # 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 = 5 for i in range(epoch): print(f"--------------------第{
i+1}轮训练开始了-----------------") # 训练步骤开始 for data in train_data_loader: imgs, targets = data output = lixinyu(imgs) loss = loss_fn(output, targets) optimizer.zero_grad() loss.backward() optimizer.step() total_train_step += 1 print(f"训练次数{
total_train_step} 损失值{
loss.item()}") # loss.item()标准写法，这里直接写loss也行 # 测试步骤开始 total_test_loss = 0 with torch.no_grad(): for data in test_data_loader: imgs, targets = data outputs = lixinyu(imgs) loss = loss_fn(outputs, targets) total_test_loss = total_test_loss + loss.item() print(f'整体测试loss为：{
total_test_loss}') 训练次数781 损失值2.112248659133911
训练次数782 损失值2.173518180847168
整体测试loss为：318.5319519042969
--------------------第2轮训练开始了-----------------
训练次数783 损失值2.0448567867279053
发现存在训练loss打印过多，不易发现测试loss
if total_train_step % 100 == 0:##########
print(f"训练次数{
total_train_step} 损失值{
loss.item()}") # loss.item()标准写法，这里直接写loss也行
优化，使用tensorboard
# 准备数据集
import torch.optim
import torchvision
from torch import nn
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
from model import *
train_data = torchvision.datasets.CIFAR10("./dataset", train=True, transform=torchvision.transforms.ToTensor(),
download=True)
test_data = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor(),
download=True)
# length
train_data_size = len(train_data)
test_data_siz = len(test_data)
print(f"训练数据集的长度为：{
train_data_size}")
print(f"测试数据集的长度为：{
test_data_siz}")
# 利用DataLoader加载
train_data_loader = DataLoader(train_data, batch_size=256)
test_data_loader = DataLoader(test_data, batch_size=256)
# 创建网络模型
lixinyu = Lixinyu()
# 损失函数
loss_fn = nn.CrossEntropyLoss()
# 优化器
# learning_rate = 0.01
learning_rate = 1e-2
optimizer = torch.optim.SGD(lixinyu.parameters(), learning_rate)
# 设置训练网络参数
# 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 = 5
# 添加tensorboard
writer = SummaryWriter("p28")
for i in range(epoch):
print(f"--------------------第{
i+1}轮训练开始了-----------------")
# 训练步骤开始
for data in train_data_loader:
imgs, targets = data
output = lixinyu(imgs)
loss = loss_fn(output, targets)
optimizer.zero_grad()
loss.backward()
optimizer.step()
total_train_step += 1
if total_train_step % 100 == 0:
print(f"训练次数{
total_train_step} 损失值{
loss.item()}") # loss.item()标准写法，这里直接写loss也行
writer.add_scalar("train_loss", loss.item(), total_train_step)
# 测试步骤开始
total_test_loss = 0
with torch.no_grad():
for data in test_data_loader:
imgs, targets = data
outputs = lixinyu(imgs)
loss = loss_fn(outputs, targets)
total_test_loss = total_test_loss + loss.item()
print(f'整体测试loss为：{
total_test_loss}')
writer.add_scalar("total_test_loss", total_test_loss, total_test_step)
total_test_step = total_test_step + 1
writer.close()
每一轮保存模型
# 准备数据集
import torch.optim
import torchvision
from torch import nn
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
from model import *
train_data = torchvision.datasets.CIFAR10("./dataset", train=True, transform=torchvision.transforms.ToTensor(),
download=True)
test_data = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor(),
download=True)
# length
train_data_size = len(train_data)
test_data_siz = len(test_data)
print(f"训练数据集的长度为：{
train_data_size}")
print(f"测试数据集的长度为：{
test_data_siz}")
# 利用DataLoader加载
train_data_loader = DataLoader(train_data, batch_size=256)
test_data_loader = DataLoader(test_data, batch_size=256)
# 创建网络模型
lixinyu = Lixinyu()
# 损失函数
loss_fn = nn.CrossEntropyLoss()
# 优化器
# learning_rate = 0.01
learning_rate = 1e-2
optimizer = torch.optim.SGD(lixinyu.parameters(), learning_rate)
# 设置训练网络参数
# 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 = 5
# 添加tensorboard
writer = SummaryWriter("p28")
for i in range(epoch):
print(f"--------------------第{
i+1}轮训练开始了-----------------")
# 训练步骤开始
for data in train_data_loader:
imgs, targets = data
output = lixinyu(imgs)
loss = loss_fn(output, targets)
optimizer.zero_grad()
loss.backward()
optimizer.step()
total_train_step += 1
if total_train_step % 100 == 0:
print(f"训练次数{
total_train_step} 损失值{
loss.item()}") # loss.item()标准写法，这里直接写loss也行
writer.add_scalar("train_loss", loss.item(), total_train_step)
# 测试步骤开始
total_test_loss = 0
with torch.no_grad():
for data in test_data_loader:
imgs, targets = data
outputs = lixinyu(imgs)
loss = loss_fn(outputs, targets)
total_test_loss = total_test_loss + loss.item()
print(f'整体测试loss为：{
total_test_loss}')
writer.add_scalar("total_test_loss", total_test_loss, total_test_step)
total_test_step = total_test_step + 1
torch.save(lixinyu, f"lixinyu_epoch{
i}.pth")##############
print("模型已保存")#############
writer.close()
分类问题loss来反映并不如直接用正确率，在目标检测，实例分割中存在不同
1跨列，横向看，0跨行，列向看
测试正确率
D:\anaconda\python.exe C:/Users/ASUS/AppData/Roaming/JetBrains/PyCharm2021.3/extensions/test.py
tensor([1, 1])
tensor([False, True])
Process finished with exit code 0
# 准备数据集
import torch.optim
import torchvision
from torch import nn
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
from model import *
train_data = torchvision.datasets.CIFAR10("./dataset", train=True, transform=torchvision.transforms.ToTensor(),
download=True)
test_data = torchvision.datasets.CIFAR10("./dataset", train=False, transform=torchvision.transforms.ToTensor(),
download=True)
# length
train_data_size = len(train_data)
test_data_siz = len(test_data)
print(f"训练数据集的长度为：{
train_data_size}")
print(f"测试数据集的长度为：{
test_data_siz}")
# 利用DataLoader加载
train_data_loader = DataLoader(train_data, batch_size=256)
test_data_loader = DataLoader(test_data, batch_size=256)
# 创建网络模型
lixinyu = Lixinyu()
# 损失函数
loss_fn = nn.CrossEntropyLoss()
# 优化器
# learning_rate = 0.01
learning_rate = 1e-2
optimizer = torch.optim.SGD(lixinyu.parameters(), learning_rate)
# 设置训练网络参数
# 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 = 5
# 添加tensorboard
writer = SummaryWriter("p28")
for i in range(epoch):
print(f"--------------------第{
i+1}轮训练开始了-----------------")
# 训练步骤开始
for data in train_data_loader:
imgs, targets = data
output = lixinyu(imgs)
loss = loss_fn(output, targets)
optimizer.zero_grad()
loss.backward()
optimizer.step()
total_train_step += 1
if total_train_step % 100 == 0:
print(f"训练次数{
total_train_step} 损失值{
loss.item()}") # loss.item()标准写法，这里直接写loss也行
writer.add_scalar("train_loss", loss.item(), total_train_step)
# 测试步骤开始
total_test_loss = 0
total_accuracy = 0
with torch.no_grad():
for data in test_data_loader:
imgs, targets = data
outputs = lixinyu(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(f'整体测试loss为：{
total_test_loss}')
print(f'整体测试数据集上的正确率：{
total_accuracy/test_data_siz}')
writer.add_scalar("total_test_loss", total_test_loss, total_test_step)
writer.add_scalar("total_accuracy", total_accuracy/test_data_siz, total_test_step)
total_test_step = total_test_step + 1
torch.save(lixinyu, f"lixinyu_epoch{
i}.pth")
print("模型已保存")
writer.close()