densenet121实现图片大小为48*48四分类代码

以下是使用PyTorch实现DenseNet121进行48x48四分类的样例代码：

import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import Dataset, DataLoader
import torchvision.transforms as transforms
import numpy as np
from PIL import Image

# 定义数据集类
class MyDataset(Dataset):
    def __init__(self, data_path, transform=None):
        self.transform = transform
        self.data = np.load(data_path)
        self.imgs = self.data[:, :-1].reshape(-1, 48, 48).astype(np.uint8)[:, :, :, None]
        self.labels = self.data[:, -1].astype(np.long)

    def __getitem__(self, index):
        img = Image.fromarray(self.imgs[index])
        if self.transform is not None:
            img = self.transform(img)
        label = self.labels[index]
        return img, label

    def __len__(self):
        return len(self.data)

# 定义模型类
class DenseNet(nn.Module):
    def __init__(self):
        super(DenseNet, self).__init__()
        self.densenet = torch.hub.load('pytorch/vision:v0.6.0', 'densenet121', pretrained=True)
        num_ftrs = self.densenet.classifier.in_features
        self.densenet.classifier = nn.Linear(num_ftrs, 4)

    def forward(self, x):
        x = self.densenet(x)
        return x

# 定义训练函数
def train(model, device, train_loader, optimizer, criterion):
    model.train()
    for batch_idx, (data, target) in enumerate(train_loader):
        data, target = data.to(device), target.to(device)
        optimizer.zero_grad()
        output = model(data)
        loss = criterion(output, target)
        loss.backward()
        optimizer.step()

# 定义测试函数
def test(model, device, test_loader, criterion):
    model.eval()
    test_loss = 0
    correct = 0
    with torch.no_grad():
        for data, target in test_loader:
            data, target = data.to(device), target.to(device)
            output = model(data)
            test_loss += criterion(output, target).item()
            pred = output.argmax(dim=1, keepdim=True)
            correct += pred.eq(target.view_as(pred)).sum().item()

    test_loss /= len(test_loader.dataset)
    accuracy = correct / len(test_loader.dataset) * 100.
    print('Test set: Average loss: {:.4f}, Accuracy: {}/{} ({:.2f}%)'.format(
        test_loss, correct, len(test_loader.dataset), accuracy))

if __name__ == '__main__':
    # 加载数据和数据增强
    transform_train = transforms.Compose([
        transforms.RandomCrop(48, padding=4),
        transforms.RandomHorizontalFlip(),
        transforms.ToTensor(),
        transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),
    ])
    transform_test = transforms.Compose([
        transforms.ToTensor(),
        transforms.Normalize((0.485, 0.456, 0.406), (0.229, 0.224, 0.225)),
    ])
    train_dataset = MyDataset('train_data.npy', transform_train)
    test_dataset = MyDataset('test_data.npy', transform_test)
    train_loader = DataLoader(train_dataset, batch_size=64, shuffle=True)
    test_loader = DataLoader(test_dataset, batch_size=64, shuffle=False)

    # 定义模型和优化器
    device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
    model = DenseNet().to(device)
    optimizer = optim.Adam(model.parameters(), lr=1e-3)
    criterion = nn.CrossEntropyLoss()

    # 训练和测试
    for epoch in range(1, 21):
        train(model, device, train_loader, optimizer, criterion)
        test(model, device, test_loader, criterion)

其中，`MyDataset`类用于定义数据集，`DenseNet`类用于定义模型，`train`函数用于训练模型，`test`函数用于测试模型。在`if __name__ == '__main__':`中，加载数据、定义模型、定义优化器、定义损失函数、训练和测试模型。需要注意的是，在`MyDataset`中，需要将图片数据reshape成(48, 48, 1)的形状，且标签数据需要转换为`long`类型。