365天深度学习训练营-第P6周：好莱坞明星识别

目录

 一、前言

 二、我的环境

三、代码实现

1、设置GPU

2、导入数据

3、划分数据集

 4、调用VGG16模型

 5、训练代码

6、测试函数

7、设置动态学习率

8、开始训练

 8、数据可视化

 9、模型评估

四、拔高部分

1、手动搭建VGG-16模型

 2、用简单的CNN模型进行训练

 3、用MobileNetV2进行训练

---

 一、前言

>- ** 本文为[365天深度学习训练营](https://mp.weixin.qq.com/s/xLjALoOD8HPZcH563En8bQ) 中的学习记录博客**
>- ** 参考文章：365天深度学习训练营-第P6周：好莱坞明星识别（训练营内部成员可读）**
>- ** 原作者：[K同学啊|接辅导、项目定制](https://mtyjkh.blog.csdn.net/)**

● 难度：夯实基础⭐⭐
● 语言：Python3、Pytorch3
● 时间：12月23日-12月28日
要求：

自己搭建VGG-16网络框架
调用官方的VGG-16网络框架
如何查看模型的参数量以及相关指标

拔高（可选）：

测试集准确率达到60%（难度有点大，但是这个过程可以学到不少）
手动搭建VGG-16网络框架

 二、我的环境

语言环境：Python3.7

编译器：jupyter notebook

深度学习环境：TensorFlow2

三、代码实现

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

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

device

2、导入数据

#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]
classeNames


# 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("./6-data/",transform=train_transforms)
total_data


total_data.class_to_idx

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


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=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

 

 4、调用VGG16模型

from torchvision.models import vgg16

device = "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

 5、训练代码

# 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

6、测试函数

# 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

7、设置动态学习率

# 设置动态学习率
def adjust_learning_rate(optimizer, epoch, start_lr):
    # 每 2 个epoch衰减到原来的 0.98
    lr = start_lr * (0.92 ** (epoch // 2))
    for param_group in optimizer.param_groups:
        param_group['lr'] = lr


learn_rate = 1e-4  # 初始学习率
optimizer = torch.optim.SGD(model.parameters(), lr=learn_rate)

8、开始训练

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

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

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)

    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')

 

 8、数据可视化

# 数据可视化
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

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.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()

 9、模型评估

#模型评估
best_model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, best_model, loss_fn)

四、拔高部分

1、手动搭建VGG-16模型

class Model(nn.Module):
    def __init__(self):
        super(Model, self).__init__()
        self.sequ1=nn.Sequential(
            nn.Conv2d(3, 64, kernel_size=3, stride=1, padding=1),     # 64*224*224
            nn.ReLU(inplace=True),
            nn.Conv2d(64, 64, kernel_size=3, stride=1, padding=1),    # 64*224*224
            nn.ReLU(inplace=True),
            nn.MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False), # 64*112*112
            nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1),   # 128*112*112
            nn.ReLU(inplace=True),
            nn.Conv2d(128, 128, kernel_size=3, stride=1, padding=1),  # 128*112*112
            nn.ReLU(inplace=True),
            nn.MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False), # 128*56*56
            nn.Conv2d(128, 256, kernel_size=3, stride=1, padding=1),  # 256*56*56
            nn.ReLU(inplace=True),
            nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1),  # 256*56*56
            nn.ReLU(inplace=True),
            nn.Conv2d(256, 256, kernel_size=3, stride=1, padding=1),  # 256*56*56
            nn.ReLU(inplace=True),
            nn.MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False), # 256*28*28
            nn.Conv2d(256, 512, kernel_size=3, stride=1, padding=1),  # 512*28*28
            nn.ReLU(inplace=True),
            nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),  # 512*28*28
            nn.ReLU(inplace=True),
            nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),  # 512*28*28
            nn.ReLU(inplace=True),
            nn.MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False), # 512*14*14
            nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),  # 512*14*14
            nn.ReLU(inplace=True),
            nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),  # 512*14*14
            nn.ReLU(inplace=True),
            nn.Conv2d(512, 512, kernel_size=3, stride=1, padding=1),  # 512*14*14
            nn.ReLU(inplace=True),
            nn.MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)  # 512*7*7
        )
        self.pool2=nn.AdaptiveAvgPool2d(output_size=(7, 7))    # 512*7*7
        self.sequ3=nn.Sequential(
            nn.Linear(in_features=25088, out_features=4096, bias=True),
            nn.ReLU(inplace=True),
            nn.Dropout(p=0.5, inplace=False),
            nn.Linear(in_features=4096, out_features=4096, bias=True),
            nn.ReLU(inplace=True),
            nn.Dropout(p=0.5, inplace=False),
            nn.Linear(in_features=4096, out_features=17, bias=True)
        )
    
    def forward(self, x):
       x = self.sequ1(x)
       x = self.pool2(x)
       x = self.sequ3(x)
       
       return x

 2、用简单的CNN模型进行训练

 

 

 3、用MobileNetV2进行训练

import torch
import torch.nn as nn
import torchvision

# 分类个数
num_class = 17

# DW卷积
def Conv3x3BNReLU(in_channels,out_channels,stride,groups):
    return nn.Sequential(
            # stride=2 wh减半，stride=1 wh不变
            nn.Conv2d(in_channels=in_channels, out_channels=out_channels, kernel_size=3, stride=stride, padding=1, groups=groups),
            nn.BatchNorm2d(out_channels),
            nn.ReLU6(inplace=True)
        )

# PW卷积
def Conv1x1BNReLU(in_channels,out_channels):
    return nn.Sequential(
            nn.Conv2d(in_channels=in_channels, out_channels=out_channels, kernel_size=1, stride=1),
            nn.BatchNorm2d(out_channels),
            nn.ReLU6(inplace=True)
        )

# # PW卷积(Linear) 没有使用激活函数
def Conv1x1BN(in_channels,out_channels):
    return nn.Sequential(
            nn.Conv2d(in_channels=in_channels, out_channels=out_channels, kernel_size=1, stride=1),
            nn.BatchNorm2d(out_channels)
        )

class InvertedResidual(nn.Module):
    # t = expansion_factor,也就是扩展因子，文章中取6
    def __init__(self, in_channels, out_channels, expansion_factor, stride):
        super(InvertedResidual, self).__init__()
        self.stride = stride
        self.in_channels = in_channels
        self.out_channels = out_channels
        mid_channels = (in_channels * expansion_factor)
        # print("expansion_factor:", expansion_factor)
        # print("mid_channels:",mid_channels)

        # 先1x1卷积升维，再1x1卷积降维
        self.bottleneck = nn.Sequential(
            # 升维操作: 扩充维度是 in_channels * expansion_factor (6倍)
            Conv1x1BNReLU(in_channels, mid_channels),
            # DW卷积,降低参数量
            Conv3x3BNReLU(mid_channels, mid_channels, stride, groups=mid_channels),
            # 降维操作: 降维度 in_channels * expansion_factor(6倍) 降维到指定 out_channels 维度
            Conv1x1BN(mid_channels, out_channels)
        )

        # 第一种: stride=1 才有shortcut 此方法让原本不相同的channels相同
        if self.stride == 1:
            self.shortcut = Conv1x1BN(in_channels, out_channels)

        # 第二种: stride=1 切 in_channels=out_channels 才有 shortcut
        # if self.stride == 1 and in_channels == out_channels:
        #     self.shortcut = ()

    def forward(self, x):
        out = self.bottleneck(x)
        # 第一种:
        out = (out+self.shortcut(x)) if self.stride==1 else out
        # 第二种:
        # out = (out + x) if self.stride == 1 and self.in_channels == self.out_channels else out
        return out

class MobileNetV2(nn.Module):
    # num_class为分类个数, t为扩充因子
    def __init__(self, num_classes=num_class, t=6):
        super(MobileNetV2,self).__init__()

        # 3 -> 32 groups=1 不是组卷积 单纯的卷积操作
        self.first_conv = Conv3x3BNReLU(3,32,2,groups=1)

        # 32 -> 16 stride=1 wh不变
        self.layer1 = self.make_layer(in_channels=32, out_channels=16, stride=1, factor=1, block_num=1)
        # 16 -> 24 stride=2 wh减半
        self.layer2 = self.make_layer(in_channels=16, out_channels=24, stride=2, factor=t, block_num=2)
        # 24 -> 32 stride=2 wh减半
        self.layer3 = self.make_layer(in_channels=24, out_channels=32, stride=2, factor=t, block_num=3)
        # 32 -> 64 stride=2 wh减半
        self.layer4 = self.make_layer(in_channels=32, out_channels=64, stride=2, factor=t, block_num=4)
        # 64 -> 96 stride=1 wh不变
        self.layer5 = self.make_layer(in_channels=64, out_channels=96, stride=1, factor=t, block_num=3)
        # 96 -> 160 stride=2 wh减半
        self.layer6 = self.make_layer(in_channels=96, out_channels=160, stride=2, factor=t, block_num=3)
        # 160 -> 320 stride=1 wh不变
        self.layer7 = self.make_layer(in_channels=160, out_channels=320, stride=1, factor=t, block_num=1)
        # 320 -> 1280 单纯的升维操作
        self.last_conv = Conv1x1BNReLU(320,1280)

        self.avgpool = nn.AvgPool2d(kernel_size=7,stride=1)
        self.dropout = nn.Dropout(p=0.2)
        self.linear = nn.Linear(in_features=1280,out_features=num_classes)
        self.init_params()

    def make_layer(self, in_channels, out_channels, stride, factor, block_num):
        layers = []
        # 与ResNet类似，每层Bottleneck单独处理，指定stride。此层外的stride均为1
        layers.append(InvertedResidual(in_channels, out_channels, factor, stride))
        # 这些叠加层stride均为1，in_channels = out_channels, 其中 block_num-1 为重复次数
        for i in range(1, block_num):
            layers.append(InvertedResidual(out_channels, out_channels, factor, 1))
        return nn.Sequential(*layers)

    # 初始化权重操作
    def init_params(self):
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                nn.init.kaiming_normal_(m.weight)
                nn.init.constant_(m.bias, 0)
            elif isinstance(m, nn.Linear) or isinstance(m, nn.BatchNorm2d):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)

    def forward(self, x):
        x = self.first_conv(x)  # torch.Size([1, 32, 112, 112])
        x = self.layer1(x)      # torch.Size([1, 16, 112, 112])
        x = self.layer2(x)      # torch.Size([1, 24, 56, 56])
        x = self.layer3(x)      # torch.Size([1, 32, 28, 28])
        x = self.layer4(x)      # torch.Size([1, 64, 14, 14])
        x = self.layer5(x)      # torch.Size([1, 96, 14, 14])
        x = self.layer6(x)      # torch.Size([1, 160, 7, 7])
        x = self.layer7(x)      # torch.Size([1, 320, 7, 7])
        x = self.last_conv(x)   # torch.Size([1, 1280, 7, 7])
        x = self.avgpool(x)     # torch.Size([1, 1280, 1, 1])
        x = x.view(x.size(0),-1)    # torch.Size([1, 1280])
        x = self.dropout(x)
        x = self.linear(x)      # torch.Size([1, 5])
        return x

device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))

model = MobileNetV2().to(device)
model

虽然算法是新算法 但是准确率并没有提高 仅供参考

MobileNet v2网络是由google团队在2018年提出的，相比MobileNet V1网络，准确率更高，模型更小。

网络中的亮点 ：

• Inverted Residuals （倒残差结构 ）
• Linear Bottlenecks（结构的最后一层采用线性层）