实战PyTorch VGG-16人脸识别：从17%调优至60%+准确率

• 🍨 本文为🔗365天深度学习训练营 中的学习记录博客
• 🍖 原作者：K同学啊
  

---

一、前期准备

设置GPU

import torch
import torch.nn as nn
import torchvision.transforms as transforms
import torchvision
from torchvision import transforms, datasets
import os,PIL,pathlib,warnings
import torchinfo
from torchinfo import summary

warnings.filterwarnings("ignore")             #忽略警告信息

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device   # 输出：device(type='cuda')

导入数据

import os,PIL,random,pathlib

data_dir = './6-data/'
data_dir = pathlib.Path(data_dir)

data_paths  = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[1] for path in data_paths]

train_transforms = transforms.Compose([
    transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸
    transforms.RandomHorizontalFlip(), # 随机水平翻转
    transforms.ToTensor(),          # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间
    transforms.Normalize(           # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛
        mean=[0.485, 0.456, 0.406], 
        std=[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
])
total_data = datasets.ImageFolder("./data/",transform=train_transforms)
train_size = int(0.8 * len(total_data))
test_size  = len(total_data) - train_size
train_dataset, test_dataset = torch.utils.data.random_split(total_data, [train_size, test_size])

构建数据加载器

batch_size = 32

train_dl = torch.utils.data.DataLoader(train_dataset,
                                           batch_size=batch_size,
                                           shuffle=True,
                                           num_workers=1)
test_dl = torch.utils.data.DataLoader(test_dataset,
                                          batch_size=batch_size,
                                          shuffle=False,
                                          num_workers=1)

二、调用官方VGG-16模型并进行变体

1.调用模型并变体

加载vgg模型后，先冻结features部分的所有参数使其无法进行训练，再解冻最后10层；然后修改vgg的classifier模块，修改vgg16模型中最后一层全连接层，输出目标类别个数，然后在classifier模块加入dropout层以此减少过拟合。

from torchvision.models import vgg16
# 加载预训练模型，并且对模型进行微调
model = vgg16(pretrained = True).to(device) # 加载预训练的vgg16模型

for param in model.parameters():
    param.requires_grad = False # 冻结模型的参数，这样子在训练的时候只训练最后一层的参数
# 只解冻最后两个卷积块（conv5_x）和分类器的最后层
# VGG16的features部分最后10层是conv5的卷积层
for param in model.features[-5:].parameters():  # 解冻后5层（conv5）
    param.requires_grad = True

model.classifier = nn.Sequential(
    nn.Linear(512*7*7, 4096),
    nn.BatchNorm1d(4096),       # 新增BN层
    nn.ReLU(),
    nn.Dropout(0.6),            # 提升Dropout率（原0.5）
    nn.Linear(4096, len(classeNames))
)
model.to(device)

2.查看变体后vgg模型结构

torchinfo的summary一直是查看模型结构的好帮手。

summary(model, (3, 224, 224), batch_dim = 0, col_names = ("input_size", "output_size", "num_params"), verbose = 0)

输出结果：

三、构建训练函数和测试函数

1、构建训练函数

def train(dataloader, model, loss_fn, optimizer):
    size = len(dataloader.dataset)  # 训练集的大小
    num_batches = len(dataloader)   # 批次数目, (size/batch_size，向上取整)

    train_loss, train_acc = 0, 0  # 初始化训练损失和正确率
    
    for X, y in dataloader:  # 获取图片及其标签
        X, y = X.to(device), y.to(device)
        
        # 计算预测误差
        pred = model(X)          # 网络输出
        loss = loss_fn(pred, y)  # 计算网络输出和真实值之间的差距，targets为真实值，计算二者差值即为损失
        
        # 反向传播
        optimizer.zero_grad()  # grad属性归零
        loss.backward()        # 反向传播
        optimizer.step()       # 每一步自动更新
        
        # 记录acc与loss
        train_acc  += (pred.argmax(1) == y).type(torch.float).sum().item()
        train_loss += loss.item()
            
    train_acc  /= size
    train_loss /= num_batches

    return train_acc, train_loss

2、构建测试函数

def test (dataloader, model, loss_fn):
    size        = len(dataloader.dataset)  # 测试集的大小
    num_batches = len(dataloader)          # 批次数目, (size/batch_size，向上取整)
    test_loss, test_acc = 0, 0
    
    # 当不进行训练时，停止梯度更新，节省计算内存消耗
    with torch.no_grad():
        for imgs, target in dataloader:
            imgs, target = imgs.to(device), target.to(device)
            
            # 计算loss
            target_pred = model(imgs)
            loss        = loss_fn(target_pred, target)
            
            test_loss += loss.item()
            test_acc  += (target_pred.argmax(1) == target).type(torch.float).sum().item()

    test_acc  /= size
    test_loss /= num_batches

    return test_acc, test_loss

四、正式训练

1、构建优化器及学习率调度器

optimizer = torch.optim.AdamW([
    {'params': model.features[-15:].parameters(), 'lr': 3e-4},
    {'params': model.classifier.parameters(), 'lr': 1e-3}
], weight_decay=5e-4)
# 学习率衰减
scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(
    optimizer, 
    mode='min',       # 监控验证损失的最小值
    factor=0.1,       # 学习率衰减因子（new_lr = lr * factor）
    patience=3,       # 容忍3个epoch没有改善
    verbose=True,     # 打印调整信息
    min_lr=1e-6       # 学习率下限
)

2、正式训练

import copy

loss_fn    = nn.CrossEntropyLoss() # 创建损失函数
epochs     = 40

train_loss = []
train_acc  = []
test_loss  = []
test_acc   = []

best_acc = 0    # 设置一个最佳准确率，作为最佳模型的判别指标

for epoch in range(epochs):
    # 更新学习率（使用自定义学习率时使用）
    # adjust_learning_rate(optimizer, epoch, learn_rate)
    
    model.train()
    epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, optimizer)
    
    model.eval()
    epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
    scheduler.step(epoch_test_loss) # 更新学习率（调用官方动态学习率接口时使用）
    
    # 保存最佳模型到 best_model
    if epoch_test_acc > best_acc:
        best_acc   = epoch_test_acc
        best_model = copy.deepcopy(model)
    
    train_acc.append(epoch_train_acc)
    train_loss.append(epoch_train_loss)
    test_acc.append(epoch_test_acc)
    test_loss.append(epoch_test_loss)
    
    # 获取当前的学习率
    lr = optimizer.state_dict()['param_groups'][0]['lr']
    
    template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}, Lr:{:.2E}')
    print(template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, 
                          epoch_test_acc*100, epoch_test_loss, lr))
    
# 保存最佳模型到文件中
PATH = './best_model.pth'  # 保存的参数文件名
torch.save(model.state_dict(), PATH)

print('Done')

训练结果如下：

五、模型结果可视化

import matplotlib.pyplot as plt
#隐藏警告
import warnings
warnings.filterwarnings("ignore")               #忽略警告信息
plt.rcParams['font.sans-serif']    = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False      # 用来正常显示负号
plt.rcParams['figure.dpi']         = 100        #分辨率

from datetime import datetime
current_time = datetime.now() # 获取当前时间

epochs_range = range(epochs)

plt.figure(figsize=(12, 3))
plt.subplot(1, 2, 1)

plt.plot(epochs_range, train_acc, label='Training Accuracy')
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.xlabel(current_time) # 打卡请带上时间戳，否则代码截图无效

plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, label='Training Loss')
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

六、用模型识别单张图片

from PIL import Image 

classes = list(total_data.class_to_idx)

def predict_one_image(image_path, model, transform, classes):
    
    test_img = Image.open(image_path).convert('RGB')
    plt.imshow(test_img)  # 展示预测的图片

    test_img = transform(test_img)
    img = test_img.to(device).unsqueeze(0)
    
    model.eval()
    output = model(img)

    _,pred = torch.max(output,1)
    pred_class = classes[pred]
    print(f'预测结果是：{pred_class}')

# 预测训练集中的某张照片
predict_one_image(image_path='./data/Angelina Jolie/001_fe3347c0.jpg', 
                  model=model, 
                  transform=train_transforms, 
                  classes=classes)

学习成果速览

1. 模型调优

• 迁移学习：冻结底层+解冻顶层卷积
• 过拟合控制：Dropout(0.7)、权重衰减(5e-4)、标签平滑

2. 训练策略

• 分层学习率（特征层1e-5，分类层5e-5）
• 动态调整：ReduceLROnPlateau