P6周：VGG-16算法-Pytorch实现人脸识别

•  🍨 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

我的环境

语言环境：Python 3.8.12

编译器：jupyter notebook

深度学习环境：torch 1.12.0+cu113

一、前期准备

1.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,warningswarnings.filterwarnings("ignore")             #忽略警告信息device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device

device(type='cuda')

2.导入数据

import os,PIL,random,pathlibdata_dir = 'F:/jupyter lab/DL-100-days/datasets/Hollywood_stars_photos/'
data_dir = pathlib.Path(data_dir)data_paths  = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[5] for path in data_paths]
classeNames

['Angelina Jolie','Brad Pitt','Denzel Washington','Hugh Jackman','Jennifer Lawrence','Johnny Depp','Kate Winslet','Leonardo DiCaprio','Megan Fox','Natalie Portman','Nicole Kidman','Robert Downey Jr','Sandra Bullock','Scarlett Johansson','Tom Cruise','Tom Hanks','Will Smith']

# 关于transforms.Compose的更多介绍可以参考：https://blog.csdn.net/qq_38251616/article/details/124878863
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("F:/jupyter lab/DL-100-days/datasets/Hollywood_stars_photos/",transform=train_transforms)
total_data

Dataset ImageFolderNumber of datapoints: 1800Root location: F:/jupyter lab/DL-100-days/datasets/Hollywood_stars_photos/StandardTransform
Transform: Compose(Resize(size=[224, 224], interpolation=bilinear, max_size=None, antialias=None)ToTensor()Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]))

total_data.class_to_idx

{'Angelina Jolie': 0,'Brad Pitt': 1,'Denzel Washington': 2,'Hugh Jackman': 3,'Jennifer Lawrence': 4,'Johnny Depp': 5,'Kate Winslet': 6,'Leonardo DiCaprio': 7,'Megan Fox': 8,'Natalie Portman': 9,'Nicole Kidman': 10,'Robert Downey Jr': 11,'Sandra Bullock': 12,'Scarlett Johansson': 13,'Tom Cruise': 14,'Tom Hanks': 15,'Will Smith': 16}

3.划分数据集 

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])
train_dataset, test_dataset

(<torch.utils.data.dataset.Subset at 0x187391e5a60>,<torch.utils.data.dataset.Subset at 0x187391e5b20>)

batch_size = 32train_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=True,num_workers=1)

for X, y in test_dl:print("Shape of X [N, C, H, W]: ", X.shape)print("Shape of y: ", y.shape, y.dtype)break

Shape of X [N, C, H, W]:  torch.Size([32, 3, 224, 224])
Shape of y:  torch.Size([32]) torch.int64

二、调用官方的VGG16模型

from torchvision.models import vgg16device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))# 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# 修改classifier模块的第6层（即：(6): Linear(in_features=4096, out_features=2, bias=True)）
# 注意查看我们下方打印出来的模型
model.classifier._modules['6'] = nn.Linear(4096,len(classeNames)) # 修改vgg16模型中最后一层全连接层，输出目标类别个数
model.to(device)  
model

Using cuda device
VGG((features): Sequential((0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(1): ReLU(inplace=True)(2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(3): ReLU(inplace=True)(4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(6): ReLU(inplace=True)(7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(8): ReLU(inplace=True)(9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(11): ReLU(inplace=True)(12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(13): ReLU(inplace=True)(14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(15): ReLU(inplace=True)(16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(18): ReLU(inplace=True)(19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(20): ReLU(inplace=True)(21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(22): ReLU(inplace=True)(23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)(24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(25): ReLU(inplace=True)(26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(27): ReLU(inplace=True)(28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))(29): ReLU(inplace=True)(30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False))(avgpool): AdaptiveAvgPool2d(output_size=(7, 7))(classifier): Sequential((0): Linear(in_features=25088, out_features=4096, bias=True)(1): ReLU(inplace=True)(2): Dropout(p=0.5, inplace=False)(3): Linear(in_features=4096, out_features=4096, bias=True)(4): ReLU(inplace=True)(5): Dropout(p=0.5, inplace=False)(6): Linear(in_features=4096, out_features=17, bias=True))
)

三、训练循环 

1.编写训练函数

# 训练循环
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losstrain_acc  += (pred.argmax(1) == y).type(torch.float).sum().item()train_loss += loss.item()train_acc  /= sizetrain_loss /= num_batchesreturn train_acc, train_loss

2.编写测试函数 

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losstarget_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  /= sizetest_loss /= num_batchesreturn test_acc, test_loss

3.设置动态学习率 

# 调用官方动态学习率接口时使用
learn_rate = 1e-3 # 初始学习率
lambda1 = lambda epoch: 0.92 ** (epoch // 4)
optimizer = torch.optim.SGD(model.parameters(), lr=learn_rate)
scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lambda1) #选定调整方法

 4.正式训练

import copyloss_fn    = nn.CrossEntropyLoss() # 创建损失函数
epochs     = 40train_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)scheduler.step() # 更新学习率（调用官方动态学习率接口时使用）model.eval()epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)# 保存最佳模型到 best_modelif epoch_test_acc > best_acc:best_acc   = epoch_test_accbest_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_model123.pth'  # 保存的参数文件名
torch.save(model.state_dict(), PATH)print('Done')

Epoch: 1, Train_acc:11.9%, Train_loss:2.810, Test_acc:14.4%, Test_loss:2.639, Lr:1.00E-03
Epoch: 2, Train_acc:17.4%, Train_loss:2.590, Test_acc:15.3%, Test_loss:2.512, Lr:1.00E-03
Epoch: 3, Train_acc:17.8%, Train_loss:2.483, Test_acc:17.5%, Test_loss:2.400, Lr:1.00E-03
Epoch: 4, Train_acc:21.0%, Train_loss:2.409, Test_acc:20.8%, Test_loss:2.344, Lr:9.20E-04
Epoch: 5, Train_acc:22.1%, Train_loss:2.331, Test_acc:23.9%, Test_loss:2.289, Lr:9.20E-04
...........
Epoch:36, Train_acc:42.2%, Train_loss:1.763, Test_acc:38.9%, Test_loss:1.858, Lr:4.72E-04
Epoch:37, Train_acc:43.8%, Train_loss:1.744, Test_acc:39.2%, Test_loss:1.872, Lr:4.72E-04
Epoch:38, Train_acc:44.0%, Train_loss:1.747, Test_acc:39.4%, Test_loss:1.872, Lr:4.72E-04
Epoch:39, Train_acc:43.9%, Train_loss:1.744, Test_acc:39.7%, Test_loss:1.859, Lr:4.72E-04
Epoch:40, Train_acc:44.4%, Train_loss:1.731, Test_acc:39.7%, Test_loss:1.885, Lr:4.34E-04
Done

四、结果可视化

1.Loss与Accuracy图

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()

 

2.对指定图片进行预测 

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='F:/jupyter lab/DL-100-days/datasets/Hollywood_stars_photos/Angelina Jolie/001_fe3347c0.jpg', model=model, transform=train_transforms, classes=classes)

预测结果是：Angelina Jolie

 3.模型评估

best_model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, best_model, loss_fn)

epoch_test_acc, epoch_test_loss

(0.3972222222222222, 1.8640953699747722)

# 查看是否与我们记录的最高准确率一致
epoch_test_acc

0.3972222222222222

五、学习心得

1.本次使用pytorch深度学习环境对官方的VGG-16模型进行调用，并且保存最佳模型权重。VGG-16模型的最大特点是深度。除此之外，其卷积层都采用3x3的卷积核和步长为1的卷积操作，同时卷积层后都有ReLU激活函数，从而降低过拟合风险。

2.训练过程中发现训练和测试的acc都过低（约为20%），通过调整动态学习率予以调整，初始学习率增大一个数量级之后，此问题得到一定的解决。

3.下一步将自行搭建VGG-16模型。