目录
1 准备数据集
2 建立模型
3 构建损失函数和优化器
4 训练+测试
5 完整代码+运行结果
6 遇到问题
我们之前学习的案例中,输入x都是一个向量;在MNIST数据集中,我们需要输入的是一个图像,怎样,图像怎么才能输入到模型中进行训练呢?一种方法是我们可以把图像映射成一个矩向量,再输入到模型中进行训练。
怎样将一个图像映射成一个向量?
如图所示是MNIST数据集中一个方格的图像,它是由28x28=784个像素组成,其中越深的地方数值越接近0,越亮的地方数值越接近1。
因此可以将此图像按照对应的像素和数值映射成一个28x28的一个矩阵,如下图所示:
具体代码如下:
# 准备数据集
batch_size = 64
transform = transforms.Compose([
transforms.ToTensor(),
#均值、标准差
transforms.Normalize((0.1307, ), (0.3081, ))
])
train_dataset = datasets.MNIST(root='../dataset/mnist/',
train=True,
download=True,
transform=transform)
train_loader = DataLoader(train_dataset,
shuffle=True,
batch_size=batch_size)
test_dataset = datasets.MNIST(root='../dataset/mnist',
train=False,
download=True,
transform=transform)
test_loader = DataLoader(test_dataset,
shuffle=False,
batch_size=batch_size)
注:
全连接网络中,要求输入的是一个矩阵,因此需要将1x28x28的这个三阶的张量变成一个一阶的向量,因此将图像的每一行的向量横着拼起来变成一串,这样就变成了一个维度为1x784的向量,一共输入N个手写数图,因此,输入矩阵维度为(N,784)。这样就可以设计我们的模型,如下图所示:
具体代码:
# 设计模型
class Net(torch.nn.Module):
def __init__(self):
super(Net, self).__init__()
self.l1 = torch.nn.Linear(784, 512)
self.l2 = torch.nn.Linear(512, 256)
self.l3 = torch.nn.Linear(256, 128)
self.l4 = torch.nn.Linear(128, 64)
self.l5 = torch.nn.Linear(64, 10)
def forward(self, x):
x = x.view(-1, 784)
x = F.relu(self.l1(x))
x = F.relu(self.l2(x))
x = F.relu(self.l3(x))
x = F.relu(self.l4(x))
return self.l5(x)
model = Net()
具体代码:
# 构建损失函数和优化器
criterion = torch.nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)
具体代码:
# 定义训练函数
def train(epoch):
running_loss = 0.0
for batch_idx, data in enumerate(train_loader, 0):
inputs, target = data
optimizer.zero_grad()
# 前馈+反馈+更新
outputs = model(inputs)
loss = criterion(outputs, target)
loss.backward()
optimizer.step()
running_loss += loss.item()
# 每300次迭代输出一次
if batch_idx % 300 == 299:
print('[%d,%5d] loss:%.3f' % (epoch + 1, batch_idx + 1, running_loss / 300))
running_loss = 0.0
# 定义测试函数
def test():
correct = 0
total = 0
with torch.no_grad():
for data in test_loader:
images, labels = data
outputs = model(images)
# 沿着第一维度找最大值的下标
_, predicted = torch.max(outputs.data, dim=1)
total += labels.size(0)
correct += (predicted == labels).sum().item()
print('Accuracy on test set:%d %%' % (100 * correct / total))
# 实例化训练和测试
if __name__ == '__main__':
# 训练10轮
for epoch in range(10):
train(epoch)
test()
完整代码:
import torch
from torchvision import transforms
from torchvision import datasets
from torch.utils.data import DataLoader
import torch.nn.functional as F
import torch.optim as optim
# 准备数据集
batch_size = 64
transform = transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307, ), (0.3081, ))
])
train_dataset = datasets.MNIST(root='../dataset/mnist/',
train=True,
download=True,
transform=transform)
train_loader = DataLoader(train_dataset,
shuffle=True,
batch_size=batch_size)
test_dataset = datasets.MNIST(root='../dataset/mnist',
train=False,
download=True,
transform=transform)
test_loader = DataLoader(test_dataset,
shuffle=False,
batch_size=batch_size)
# 设计模型
class Net(torch.nn.Module):
def __init__(self):
super(Net, self).__init__()
self.l1 = torch.nn.Linear(784, 512)
self.l2 = torch.nn.Linear(512, 256)
self.l3 = torch.nn.Linear(256, 128)
self.l4 = torch.nn.Linear(128, 64)
self.l5 = torch.nn.Linear(64, 10)
def forward(self, x):
x = x.view(-1, 784)
x = F.relu(self.l1(x))
x = F.relu(self.l2(x))
x = F.relu(self.l3(x))
x = F.relu(self.l4(x))
return self.l5(x)
model = Net()
# 构建损失函数和优化器
criterion = torch.nn.CrossEntropyLoss()
optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.5)
# 定义训练函数
def train(epoch):
running_loss = 0.0
for batch_idx, data in enumerate(train_loader, 0):
inputs, target = data
optimizer.zero_grad()
# 前馈+反馈+更新
outputs = model(inputs)
loss = criterion(outputs, target)
loss.backward()
optimizer.step()
running_loss += loss.item()
if batch_idx % 300 == 299:
print('[%d,%5d] loss:%.3f' % (epoch + 1, batch_idx + 1, running_loss / 300))
running_loss = 0.0
# 定义测试函数
def test():
correct = 0
total = 0
with torch.no_grad():
for data in test_loader:
images, labels = data
outputs = model(images)
_, predicted = torch.max(outputs.data, dim=1)
total += labels.size(0)
correct += (predicted == labels).sum().item()
print('Accuracy on test set:%d %%' % (100 * correct / total))
# 实例化训练和测试
if __name__ == '__main__':
for epoch in range(10):
train(epoch)
test()
运行截图如下:
运行时遇到警告:The given NumPy array is not writeable,and PyTorch does not support non-writeable tensor,如图:
按照路径找到mnist.py文件:
点开修改:删除copy+False,就没有报错,程序可以继续运行了