构建深度网络模型实现甲骨文识别

珞山春风

已于 2024-08-05 21:20:07 修改

阅读量473

点赞数 15

文章标签：人工智能深度学习机器学习 python

于 2024-07-19 17:31:47 首次发布

本文链接：https://blog.csdn.net/weixin_53721065/article/details/140555343

版权

本项目实现了构建一个深度网络来进行对甲骨文的识别，主要项目在甲骨文项目下，数据集爬取代码文件时爬取数据集的爬虫代码。

在jiaguwen项目中，datasets下面是数据集，分为训练集和测试集，每个文件夹的子文件夹里面的所有图片代表同一个字的甲骨文。
model.py是构建的模型
train.py是训练代码
predict.py是测试单张图片并显示结果的代码，可修改相应的图片路径进行测试
main.py是测试多张甲骨文图片并统计结果的代码
我预训练好的98%以上准确率的模型在results文件夹下

首先选用残差网络（ResNet）来构建自己的模型架构

部分代码如下：

# 定义ResNet类
class ResNet(nn.Module):
    def __init__(self, block, blocks_num, num_classes=1000, include_top=True, groups=1, width_per_group=64):
        super(ResNet, self).__init__()
        self.include_top = include_top
        # maxpool的输出通道数为64，残差结构的输入通道为64
        self.in_channel = 64
        self.groups = groups
        self.width_per_group = width_per_group
        self.conv1 = nn.Conv2d(3, self.in_channel, kernel_size=7, stride=2, padding=3, bias=False)
        self.bn1 = nn.BatchNorm2d(self.in_channel)
        self.relu = nn.ReLU(inplace=True)
        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
        # 浅层步长为1，深层的步长为2
        # block：定义了两种残差模块
        # block_num:定义了残差快的个数
        self.layer1 = self._make_layer(block, 64, blocks_num[0])
        self.layer2 = self._make_layer(block, 128, blocks_num[1], stride=2)
        self.layer3 = self._make_layer(block, 256, blocks_num[2], stride=2)
        self.layer4 = self._make_layer(block, 512, blocks_num[3], stride=2)
        if self.include_top:
            self.avgpool = nn.AdaptiveAvgPool2d((1, 1))  # 自适应平均池化，指定输出（H，w），通道数不变
            self.fc = nn.Linear(512 * block.expansion, num_classes)
        # 遍历网络中的每一层
        # 继承nn.Moudle类中的一个方法：self.Moudle(),它会返回该网络中所有的moudles
        for m in self.modules():
            if isinstance(m, nn.Conv2d):
                # kaiming正态分布初始化，使得卷积层反向传播的输出的方差都为1
                # fan_in权重是通过线性层隐形确定
                # fan_out：通过创建随机矩阵显式创建权重
                nn.init.kaiming_normal_(m.weight, mode='fan_out', nonlinearity='relu')

    def _make_layer(self, block, channel, block_num, stride=1):
        downsample = None
        if stride != 1 or self.in_channel != channel * block.expansion:
            downsample = nn.Sequential(
                nn.Conv2d(self.in_channel, channel * block.expansion, kernel_size=1, stride=stride, bias=False),
                nn.BatchNorm2d(channel * block.expansion)
            )
        layers = []
        layers.append(block(self.in_channel,
                            channel,
                            downsample=downsample,
                            stride=stride,
                            groups=self.groups,
                            width_per_group=self.width_per_group))
        self.in_channel = channel * block.expansion

        for _ in range(1, block_num):
            layers.append(block(self.in_channel,
                                channel,
                                groups=self.groups,
                                width_per_group=self.width_per_group))
        # Sequential:自定义顺序连接成模型，生成网络结构
        return nn.Sequential(*layers)

    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)
        x = self.maxpool(x)
        # 上述都是静态层，下边是动态层
        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)

        if self.include_top:
            x = self.avgpool(x)
            x = torch.flatten(x, 1)
            x = self.fc(x)
        return x

根据不同需要可选取不同的层数：

# 34层的resnet
def resnet34(num_classes=376, include_top=True):
    # https://download.pytorch.org/models/resnet34-333f7ec4.pth
    return ResNet(BasicBlock, [3, 4, 6, 3], num_classes=num_classes, include_top=include_top)


# 50层的resnet
def resnet50(num_classes=376, include_top=True):
    # https://download.pytorch.org/models/resnet50-19c8e357.pth
    return ResNet(Bottleneck, [3, 4, 6, 3], num_classes=num_classes, include_top=include_top)


# 101层的resnet
def resnet101(num_classes=376, include_top=True):
    # https://download.pytorch.org/models/resnet101-5d3b4d8f.pth
    return ResNet(Bottleneck, [3, 4, 23, 3], num_classes=num_classes, include_top=include_top)

然后编写训练模型的代码，这里选用了34层的模型，部分代码如下：

   train_loader = torch.utils.data.DataLoader(train_dataset,
                                               batch_size=batch_size, shuffle=True,
                                               num_workers=nw)
    # 加载测试数据集
    validate_dataset = datasets.ImageFolder(root=os.path.join(image_path, "val"),
                                            transform=data_transform["val"])
    # 测试集长度
    val_num = len(validate_dataset)
    validate_loader = torch.utils.data.DataLoader(validate_dataset,
                                                  batch_size=batch_size, shuffle=False,
                                                  num_workers=nw)

    print("using {} images for training, {} images for validation.".format(train_num,
                                                                           val_num))

    # 模型实例化
    net = resnet34()
    net.to(device)

然后就可以在项目下得到最优的模型，我这里训练了准确度98%的模型在results文件夹下：