深度学习实战基础案例——卷积神经网络（CNN）基于Xception的猫狗识别|第2例

今天使用轻量级的一个网络Xception做一个简单的猫狗识别案例，我的环境具体如下：

• pytorch：2.0
• python：3.8
• jupyter notebook

一、环境准备

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

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')

二、数据预处理

读取数据：

import os,PIL,random,pathlib

data_dir = 'dataset/'
data_dir = pathlib.Path(data_dir)

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

# 输出
['cat', 'dog']

数据处理

# 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。
])

test_transform = transforms.Compose([
    transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸
    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_dir,transform=train_transforms)
total_data

将数据集进行分类

total_data.class_to_idx

# 输出
{
    
    'cat': 0, 'dog': 1}

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

数据集加载

batch_size = 4

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

查看数据集形状

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

三、构建模型

Xception的具体网络结构如下所示：

import math
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.model_zoo as model_zoo
from torch.nn import init
import torch


class SeparableConv2d(nn.Module):
    def __init__(self, in_channels, out_channels, kernel_size=1, stride=1, padding=0, dilation=1, bias=False):
        super(SeparableConv2d, self).__init__()

        self.conv1 = nn.Conv2d(in_channels, in_channels, kernel_size, stride, padding, dilation, groups=in_channels,
                               bias=bias)
        self.pointwise = nn.Conv2d(in_channels, out_channels, 1, 1, 0, 1, 1, bias=bias)

    def forward(self, x):
        x = self.conv1(x)
        x = self.pointwise(x)
        return x

class Block(nn.Module):
    def __init__(self, in_filters, out_filters, reps, strides=1, start_with_relu=True, grow_first=True):
        # :parm reps:块重复次数
        super(Block, self).__init__()

        # Middle flow无需做这一步，而其余块需要，以做跳连
        # 1)Middle flow输入输出特征图个数始终一致，且Stride恒为1
        # 1）其余快stride=2，这样可以将特征图尺寸减半，获得与最大池化减半特征图尺寸同样的效果
        if out_filters != in_filters or strides != 1:
            self.skip = nn.Conv2d(in_filters, out_filters, 1, stride=strides, bias=False)
            self.skipbn = nn.BatchNorm2d(out_filters)
        else:
            self.skip = None

        self.relu = nn.ReLU(inplace=True)
        rep = []

        filters = in_filters
        if grow_first:
            rep.append(self.relu)
            # 这里的卷积不改变特征图尺寸
            rep.append(SeparableConv2d(in_filters, out_filters, 3, stride=1, padding=1, bias=False))
            rep.append(nn.BatchNorm2d(out_filters))
            filters = out_filters

        for i in range(reps - 1):
            rep.append(self.relu)
            rep.append(SeparableConv2d(filters, filters, 3, stride=1, padding=1, bias=False))
            rep.append(nn.BatchNorm2d(filters))

        if not grow_first:
            rep.append(self.relu)
            rep.append(SeparableConv2d(in_filters, out_filters, 3, stride=1, padding=1, bias=False))
            rep.append(nn.BatchNorm2d(out_filters))

        if not start_with_relu:
            rep = rep[1:]
        else:
            rep[0] = nn.ReLU(inplace=False)

        # Middle flow 的stride恒为1，因此无需做池化，而其余块需要
        # 其余块的stride=2，因此这里的最大池化可以将特征图尺寸减半
        if strides != 1:
            rep.append(nn.MaxPool2d(3, strides, 1))
        self.rep = nn.Sequential(*rep)

    def forward(self, inp):
        x = self.rep(inp)

        if self.skip is not None:
            skip = self.skip(inp)
            skip = self.skipbn(skip)
        else:
            skip = inp

        x += skip
        return x

class Xception(nn.Module):
    def __init__(self, num_classes):
        super(Xception, self).__init__()
        self.num_classes = num_classes # 总分类数

        ###############################定义 Entry flow#################################
        self.conv1 = nn.Conv2d(in_channels=3, out_channels=32, kernel_size=3, stride=2, padding=0, bias=False)
        self.bn1 = nn.BatchNorm2d(32)
        self.relu = nn.ReLU(inplace=True)

        self.conv2 = nn.Conv2d(in_channels=32, out_channels=64, kernel_size=3, stride=1, padding=0, bias=False)
        self.bn2 = nn.BatchNorm2d(64)
        # do relu here

        # Block中的参数顺序：in_filters,out_filters,reps,stride,start_with_relu,grow_first
        self.block1 = Block(64, 128, 2, 2, start_with_relu=False, grow_first=True)
        self.block2 = Block(128, 256, 2, 2, start_with_relu=True, grow_first=True)
        self.block3 = Block(256, 728, 2, 2, start_with_relu=True, grow_first=True)

        ##############################定义 Middle flow################################
        self.block4 = Block(728, 728, 3, 1, start_with_relu=True, grow_first=True)
        self.block5 = Block(728, 728, 3, 1, start_with_relu=True, grow_first=True)
        self.block6 = Block(728, 728, 3, 1, start_with_relu=True, grow_first=True)
        self.block7 = Block(728, 728, 3, 1, start_with_relu=True, grow_first=True)

        self.block8 = Block(728, 728, 3, 1, start_with_relu=True, grow_first=True)
        self.block9 = Block(728, 728, 3, 1, start_with_relu=True, grow_first=True)
        self.block10 = Block(728, 728, 3, 1, start_with_relu=True, grow_first=True)
        self.block11 = Block(728, 728, 3, 1, start_with_relu=True, grow_first=True)


        #############################定义 Exit flow###################################
        self.block12 = Block(728, 1024, 2, 2, start_with_relu=True, grow_first=False)

        self.conv3 = SeparableConv2d(1024, 1536, 3, 1, 1)
        self.bn3 = nn.BatchNorm2d(1536)


        # do relu here
        self.conv4 = SeparableConv2d(1536, 2048, 3, 1, 1)
        self.bn4 = nn.BatchNorm2d(2048)

        self.fc = nn.Linear(2048, num_classes)

        ###############################################################################




        #--------------------init weights---------------------#
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
                m.weight.data.normal_(0, math.sqrt(2. / n))
            elif isinstance(m, nn.BatchNorm2d):
                m.weight.data.fill_(1)
                m.bias.data.zero_()
        #----------------------------------------------------------------




    def forward(self, x):
        ###########################定义 Entry flow ######################################
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)

        x = self.conv2(x)
        x = self.bn2(x)
        x = self.relu(x)

        x = self.block1(x)
        x = self.block2(x)
        x = self.block3(x)

        ######################## 定义 Middle flow#######################################
        x = self.block4(x)
        x = self.block5(x)
        x = self.block6(x)
        x = self.block7(x)
        x = self.block8(x)
        x = self.block9(x)
        x = self.block10(x)
        x = self.block11(x)

        ######################### 定义 Exit flow #######################################
        x = self.block12(x)

        x = self.conv3(x)
        x = self.bn3(x)
        x = self.relu(x)

        x = self.conv4(x)
        x = self.bn4(x)
        x = self.relu(x)

        x = F.adaptive_avg_pool2d(x, (1,1))
        x = x.view(x.size(0), -1)
        x = self.fc(x)

        return x

四、实例化模型

device = "cuda:0" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))

xception = Xception(num_classes = 2)
model = xception.to(device)
model

五、训练模型

5.1 构建训练函数

# 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

5.2 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

5.3 开始正式训练

import copy

optimizer  = torch.optim.Adam(model.parameters(), lr= 1e-3)
loss_fn    = nn.CrossEntropyLoss() # 创建损失函数

epochs     = 5

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):

    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)

    # 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(best_model.state_dict(), PATH)

print('Done')

六、可视化精度和损失

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()

七、个体预测

随便去网上找一张猫狗图片，进行预测。

# 预测

import matplotlib.pyplot as plt
from PIL import Image
from torchvision.transforms import transforms
import torch
import matplotlib.pyplot as plt

plt.rcParams['font.sans-serif']=['SimHei']  #解决中文显示乱码问题
plt.rcParams['axes.unicode_minus']=False  #解决坐标轴负数的负号显示问题



data_transform = transforms.Compose([
    transforms.Resize([224, 224]),  # 将输入图片resize成统一尺寸
    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。
])


img = Image.open("cat.jpg")
plt.imshow(img)
img = data_transform(img)
img = torch.unsqueeze(img, dim=0)
name=['狗','猫']
model_weight_path = "best_model.pth"
model = Xception(num_classes = 2)
model.load_state_dict(torch.load(model_weight_path))
model.eval()
with torch.no_grad():
    output = torch.squeeze(model(img))

    predict = torch.softmax(output, dim=0)
    # 获得最大可能性索引
    predict_cla = torch.argmax(predict).numpy()
    print('索引为', predict_cla)
print('预测结果为：{},置信度为: {}'.format(name[predict_cla], predict[predict_cla].item()))
plt.show()

总结

• Xception（又称为 Extreme Inception）是一种卷积神经网络架构，在 2016 年由 Google 提出，它的名字是由 ‘Extreme’ 和 ‘Inception’ 两个词汇组成的。Xception 基于 Inception V3 模型基础进行改进，使用深度可分离卷积来代替传统的卷积，从而更加有效地减少了模型的参数数量和计算复杂度。
• 在传统的 Inception 模型中，每个计算单元采用了两个卷积层，一个 1x1 的卷积层用于降低特征图的通道数，紧接着是一个 3x3 的卷积层用于进行特征提取。而 Xception 则将 1x1 和 3x3 卷积逐次分开，使用了深度可分离卷积作为基本的计算单元。深度可分离卷积将标准卷积分解为两部分，首先使用深度卷积来处理每个输入通道，然后再使用 1x1 的逐点卷积来融合通道，从而获得与标准卷积近似的特征提取效果。而深度可分离卷积相较于标准卷积而言，可以明显降低参数量和训练计算量。
• 而使用深度可分离卷积单元取代了传统的卷积操作之后，Xception模型在计算效率，模型大小上都相比 Inception V3 有大幅的提升。
• 总之，Xception模型的优势是在极大的减少了网络参数量和计算复杂度的同时，可以保持卓越的性能表现。因此，Xception模型已经被广泛地应用与图像分类、目标检测等任务中。