Premiers pas avec pytorch pour apprendre les réseaux de neurones convolutifs CNN

Premiers pas avec pytorch pour apprendre les réseaux de neurones convolutifs CNN

résultat de l'opération

 [1,  300] loss: 1.008
 [1,  600] loss: 0.289
 [1,  900] loss: 0.203
Accuracy on test set: 95 % 
 [2,  300] loss: 0.163
 [2,  600] loss: 0.135
 [2,  900] loss: 0.123
Accuracy on test set: 96 % 
 [3,  300] loss: 0.109
 [3,  600] loss: 0.107
 [3,  900] loss: 0.099
Accuracy on test set: 97 % 
 [4,  300] loss: 0.092
 [4,  600] loss: 0.085
 [4,  900] loss: 0.084
Accuracy on test set: 97 % 
 [5,  300] loss: 0.078
 [5,  600] loss: 0.077
 [5,  900] loss: 0.074
Accuracy on test set: 98 % 
 [6,  300] loss: 0.069
 [6,  600] loss: 0.070
 [6,  900] loss: 0.068
Accuracy on test set: 98 % 
 [7,  300] loss: 0.065
 [7,  600] loss: 0.061
 [7,  900] loss: 0.061
Accuracy on test set: 98 % 
 [8,  300] loss: 0.060
 [8,  600] loss: 0.056
 [8,  900] loss: 0.059
Accuracy on test set: 98 % 
 [9,  300] loss: 0.051
 [9,  600] loss: 0.052
 [9,  900] loss: 0.057
Accuracy on test set: 98 % 
 [10,  300] loss: 0.052
 [10,  600] loss: 0.049
 [10,  900] loss: 0.052
Accuracy on test set: 98 % 

Process finished with exit code 0

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


#step1 准备数据集

batch_size = 64
transform = transforms.Compose([
    transforms.ToTensor(),
    transforms.Normalize((0.137,),(0.3081,))
])

train_dataset = datasets.MNIST(root='../dataset/mnist',
                               train=True,
                               download=True,
                               transform=transform)

train_loder = DataLoader(train_dataset,
                         shuffle=True,
                         batch_size=batch_size)

test_dataset = datasets.MNIST(root='../dataset/mnist',
                               train=False,
                               download=True,
                               transform=transform)

test_loder = DataLoader(test_dataset,
                        shuffle=False,
                        batch_size=batch_size)

#step2 搭建网络
class Net(torch.nn.Module):
    def __init__(self):
        super(Net, self).__init__()
        self.conv1 = torch.nn.Conv2d(1,10,kernel_size=5)   #卷积层
        self.conv2 = torch.nn.Conv2d(10,20,kernel_size=5)
        self.pooling = torch.nn.MaxPool2d(2)   #池化
        self.fc = torch.nn.Linear(320,10)      #full connecting 全连接层

    def forward(self,x):
        batch_size = x.size(0)
        x = F.relu(self.pooling(self.conv1(x)))   # 卷积层 -> 池化 -> relu 层
        x = F.relu(self.pooling(self.conv2(x)))
        x = x.view(batch_size,-1)
        x = self.fc(x)
        return x

model = Net()

#使用GPU进行加速
device = torch.device("cuda:0" if torch.cuda.is_available() else"cpu")
model.to(device)

criterion = torch.nn.CrossEntropyLoss()
optimizer = torch.optim.SGD(model.parameters(),lr=0.01)


#step3 训练
def train(epoch):
    running_loss = 0.0
    for batch_idx,data in enumerate(train_loder,0):
        inputs,target = data
        inputs,target = inputs.to(device),target.to(device)
        optimizer.zero_grad()    #梯度清零

        #forward + backward + update
        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

def test():
    correct = 0
    total = 0
    with torch.no_grad():      #不计算梯度
        for data in test_loder:
            inputs,target = data
            inputs, target = inputs.to(device), target.to(device)
            outputs = model(inputs)
            _,predicted = torch.max(outputs.data,dim=1)
            total += target.size(0)
            correct += (predicted == target).sum().item()
    print('Accuracy on test set: %d %% '%(100 * correct / total))

if __name__ == '__main__':
    for epoch in range(10):
        train(epoch)
        test()