resnet-50项目实例

### ResNet-50 实现概述

ResNet-50 是一种深度残差网络，在图像分类任务上表现出色。该模型通过引入跳跃连接解决了深层网络中的梯度消失问题[^3]。

#### 使用 PyTorch 的 ResNet-50 示例代码

以下是基于 PyTorch 构建并训练 ResNet-50 模型的一个简单例子：

import torch
from torchvision import models, transforms
from torch.utils.data import DataLoader
from torchvision.datasets import CIFAR10
import torch.nn as nn
import torch.optim as optim


def train_model(model, criterion, optimizer, dataloaders, device='cuda', num_epochs=25):
    model.to(device)

    for epoch in range(num_epochs):
        print(f'Epoch {epoch}/{num_epochs - 1}')
        print('-' * 10)

        for phase in ['train']:
            if phase == 'train':
                model.train()

            running_loss = 0.0
            running_corrects = 0

            for inputs, labels in dataloaders[phase]:
                inputs = inputs.to(device)
                labels = labels.to(device)

                outputs = model(inputs)
                loss = criterion(outputs, labels)

                if phase == 'train':
                    optimizer.zero_grad()
                    loss.backward()
                    optimizer.step()

                _, preds = torch.max(outputs, 1)
                running_loss += loss.item() * inputs.size(0)
                running_corrects += torch.sum(preds == labels.data)

            epoch_loss = running_loss / len(dataloaders[phase].dataset)
            epoch_acc = running_corrects.double() / len(dataloaders[phase].dataset)

            print(f'{phase} Loss: {epoch_loss:.4f} Acc: {epoch_acc:.4f}')

data_transforms = {
    'train': transforms.Compose([
        transforms.RandomResizedCrop(224),
        transforms.RandomHorizontalFlip(),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]),
}

image_datasets = {'train': CIFAR10(root='./data', train=True,
                                   download=True, transform=data_transforms['train'])}
dataloaders_dict = {'train': DataLoader(image_datasets[x], batch_size=4,
                                       shuffle=True, num_workers=4)}

device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')

model_ft = models.resnet50(pretrained=True)
num_ftrs = model_ft.fc.in_features
model_ft.fc = nn.Linear(num_ftrs, 10)  # 修改最后一层全连接层适应CIFAR-10数据集类别数量

criterion = nn.CrossEntropyLoss()
optimizer_ft = optim.SGD(model_ft.parameters(), lr=0.001, momentum=0.9)

model_ft = train_model(model_ft, criterion, optimizer_ft, dataloaders=dataloaders_dict)

此段代码展示了如何加载预训练的 ResNet-50 并针对特定的数据集调整最后几层以适配新的分类任务，同时设置了简单的训练循环来进行微调[^2]。

#### 使用 TensorFlow/Keras 的 ResNet-50 示例代码

对于那些更倾向于使用 TensorFlow 或 Keras API 来构建和训练模型的人来说，下面是一个类似的实现方式：

import tensorflow as tf
from tensorflow.keras.applications import ResNet50
from tensorflow.keras.layers import Dense, GlobalAveragePooling2D
from tensorflow.keras.models import Model
from tensorflow.keras.preprocessing.image import ImageDataGenerator

base_model = ResNet50(weights='imagenet', include_top=False)

x = base_model.output
x = GlobalAveragePooling2D()(x)
predictions = Dense(10, activation='softmax')(x)

model = Model(inputs=base_model.input, outputs=predictions)

for layer in base_model.layers:
    layer.trainable = False

datagen = ImageDataGenerator(
        rescale=1./255,
        shear_range=0.2,
        zoom_range=0.2,
        horizontal_flip=True)

train_generator = datagen.flow_from_directory(
        './data/train',
        target_size=(224, 224),
        batch_size=32,
        class_mode='categorical')

model.compile(optimizer='adam', loss='categorical_crossentropy')
model.fit(train_generator, epochs=10)

这段代码同样实现了迁移学习的过程，其中冻结了原始 ResNet 层，并只对新添加的部分进行了训练[^1]。