昇思25天学习打卡营第23天|计算机视觉-ResNet50图像分类

昇思25天学习打卡营第23天|计算机视觉-ResNet50图像分类

ResNet网络

**背景：**传统的卷积神经网络都是将一系列的卷积层和池化层堆叠得到的，但当网络堆叠到一定深度时，就会出现退化问题，即一味地加深网络并没能减少训练误差和测试误差。

**意义：**残差网络结构（Residual Network）：使用ResNet网络可以实现搭建较深的网络结构（突破1000层），并且ResNet网络层数越深，其训练误差和测试误差越小。有了这个结论之后，神经网络开始进入算力才是硬道理的时代。除此之外，残差连接使得梯度可以更快地传递，提升了网络的收敛速度。

**应用：**ResNet在图像分类、目标检测、语音识别等任务中表现出色。

残差网络结构

数据处理

from download import download

url = "https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/datasets/cifar-10-binary.tar.gz"

download(url, "./datasets-cifar10-bin", kind="tar.gz", replace=True)

下载后的数据集目录结构如下：

datasets-cifar10-bin/cifar-10-batches-bin
├── batches.meta.text #类别
├── data_batch_1.bin
├── data_batch_2.bin
├── data_batch_3.bin
├── data_batch_4.bin
├── data_batch_5.bin
├── readme.html
└── test_batch.bin

然后，使用mindspore.dataset.Cifar10Dataset接口来加载数据集，并进行相关图像增强操作。

import mindspore as ms
import mindspore.dataset as ds
import mindspore.dataset.vision as vision
import mindspore.dataset.transforms as transforms
from mindspore import dtype as mstype

data_dir = "./datasets-cifar10-bin/cifar-10-batches-bin"  # 数据集根目录
batch_size = 256  # 批量大小
image_size = 32  # 训练图像空间大小
workers = 4  # 并行线程个数
num_classes = 10  # 分类数量


def create_dataset_cifar10(dataset_dir, usage, resize, batch_size, workers):

    data_set = ds.Cifar10Dataset(dataset_dir=dataset_dir,
                                 usage=usage,
                                 num_parallel_workers=workers,
                                 shuffle=True)

    trans = []
    if usage == "train":
        trans += [
            vision.RandomCrop((32, 32), (4, 4, 4, 4)),
            vision.RandomHorizontalFlip(prob=0.5)
        ]

    trans += [
        vision.Resize(resize),
        vision.Rescale(1.0 / 255.0, 0.0),
        vision.Normalize([0.4914, 0.4822, 0.4465], [0.2023, 0.1994, 0.2010]),
        vision.HWC2CHW()
    ]

    target_trans = transforms.TypeCast(mstype.int32)

    # 数据映射操作
    data_set = data_set.map(operations=trans,
                            input_columns='image',
                            num_parallel_workers=workers)

    data_set = data_set.map(operations=target_trans,
                            input_columns='label',
                            num_parallel_workers=workers)

    # 批量操作
    data_set = data_set.batch(batch_size)

    return data_set


# 获取处理后的训练与测试数据集

dataset_train = create_dataset_cifar10(dataset_dir=data_dir,
                                       usage="train",
                                       resize=image_size,
                                       batch_size=batch_size,
                                       workers=workers)
step_size_train = dataset_train.get_dataset_size()

dataset_val = create_dataset_cifar10(dataset_dir=data_dir,
                                     usage="test",
                                     resize=image_size,
                                     batch_size=batch_size,
                                     workers=workers)
step_size_val = dataset_val.get_dataset_size()

模型构建

残差网络的类别

from typing import Type, Union, List, Optional
import mindspore.nn as nn
from mindspore.common.initializer import Normal

# 初始化卷积层与BatchNorm的参数，适用于较浅的网络
weight_init = Normal(mean=0, sigma=0.02)
gamma_init = Normal(mean=1, sigma=0.02)
#定义ResidualBlockBase类实现Building Block结构
class ResidualBlockBase(nn.Cell):
    expansion: int = 1  # 最后一个卷积核数量与第一个卷积核数量相等

    def __init__(self, in_channel: int, out_channel: int,
                 stride: int = 1, norm: Optional[nn.Cell] = None,
                 down_sample: Optional[nn.Cell] = None) -> None:
        super(ResidualBlockBase, self).__init__()
        if not norm:
            self.norm = nn.BatchNorm2d(out_channel)
        else:
            self.norm = norm

        self.conv1 = nn.Conv2d(in_channel, out_channel,
                               kernel_size=3, stride=stride,
                               weight_init=weight_init)
        self.conv2 = nn.Conv2d(in_channel, out_channel,
                               kernel_size=3, weight_init=weight_init)
        self.relu = nn.ReLU()
        self.down_sample = down_sample

    def construct(self, x):
        """ResidualBlockBase construct."""
        identity = x  # shortcuts分支

        out = self.conv1(x)  # 主分支第一层：3*3卷积层
        out = self.norm(out)
        out = self.relu(out)
        out = self.conv2(out)  # 主分支第二层：3*3卷积层
        out = self.norm(out)

        if self.down_sample is not None:
            identity = self.down_sample(x)
        out += identity  # 输出为主分支与shortcuts之和
        out = self.relu(out)

        return out
#定义ResidualBlock类实现Bottleneck结构，适用于深层网络
class ResidualBlock(nn.Cell):
    expansion = 4  # 最后一个卷积核的数量是第一个卷积核数量的4倍

    def __init__(self, in_channel: int, out_channel: int,
                 stride: int = 1, down_sample: Optional[nn.Cell] = None) -> None:
        super(ResidualBlock, self).__init__()

        # 第一层：1x1 卷积，用于减少或增加通道数
        self.conv1 = nn.Conv2d(in_channel, out_channel,
                               kernel_size=1, weight_init=weight_init)
        self.norm1 = nn.BatchNorm2d(out_channel)

        # 第二层：3x3 卷积，stride 可以控制是否减小 feature map 尺寸
        self.conv2 = nn.Conv2d(out_channel, out_channel,
                               kernel_size=3, stride=stride,
                               weight_init=weight_init)
        self.norm2 = nn.BatchNorm2d(out_channel)

        # 第三层：1x1 卷积，用于恢复通道数，并扩展到 expansion 倍
        self.conv3 = nn.Conv2d(out_channel, out_channel * self.expansion,
                               kernel_size=1, weight_init=weight_init)
        self.norm3 = nn.BatchNorm2d(out_channel * self.expansion)

        self.relu = nn.ReLU()
        self.down_sample = down_sample  # 如果需要降采样，则使用的 down_sample 操作

    def construct(self, x):
        identity = x  # shortcuts 分支，即输入 x 本身

        out = self.conv1(x)  # 主分支第一层：1x1 卷积层
        out = self.norm1(out)
        out = self.relu(out)

        out = self.conv2(out)  # 主分支第二层：3x3 卷积层
        out = self.norm2(out)
        out = self.relu(out)

        out = self.conv3(out)  # 主分支第三层：1x1 卷积层，扩展通道数到 out_channel * expansion
        out = self.norm3(out)

        # 如果需要降采样，则对输入 x 进行降采样操作
        if self.down_sample is not None:
            identity = self.down_sample(x)

        out += identity  # 输出为主分支与 shortcuts 分支的和
        out = self.relu(out)

        return out

Building Block（resnet-50中没用到）

1. expansion: int = 1：意味着输出通道数和输入通道数相同

Bottleneck（后续残差网络就是用这个结构）

1. expansion: int = 4：最后一个卷积层的输出通道数是第一个卷积层输出通道数的4倍，（nn.BatchNorm2d(out_channel * self.expansion)）
2. 由于输入和输出通道数不匹配，需要在if语句中进行降采样

实现残差块的构建

def make_layer(last_out_channel, block: Type[Union[ResidualBlockBase, ResidualBlock]],
               channel: int, block_nums: int, stride: int = 1):
    down_sample = None  # shortcuts分支
	#对于**Bottleneck**的特殊处理
    if stride != 1 or last_out_channel != channel * block.expansion:

        down_sample = nn.SequentialCell([
            nn.Conv2d(last_out_channel, channel * block.expansion,
                      kernel_size=1, stride=stride, weight_init=weight_init),
            nn.BatchNorm2d(channel * block.expansion, gamma_init=gamma_init)
        ])

    layers = []
    layers.append(block(last_out_channel, channel, stride=stride, down_sample=down_sample))

    in_channel = channel * block.expansion
    # 堆叠残差网络
    for _ in range(1, block_nums):

        layers.append(block(in_channel, channel))

    return nn.SequentialCell(layers)

from mindspore import load_checkpoint, load_param_into_net


class ResNet(nn.Cell):
    def __init__(self, block: Type[Union[ResidualBlockBase, ResidualBlock]],
                 layer_nums: List[int], num_classes: int, input_channel: int) -> None:
        super(ResNet, self).__init__()

        self.relu = nn.ReLU()
        # 第一个卷积层，输入channel为3（彩色图像），输出channel为64
        self.conv1 = nn.Conv2d(3, 64, kernel_size=7, stride=2, weight_init=weight_init)
        self.norm = nn.BatchNorm2d(64)
        # 最大池化层，缩小图片的尺寸
        self.max_pool = nn.MaxPool2d(kernel_size=3, stride=2, pad_mode='same')
        # 各个残差网络结构块定义
        self.layer1 = make_layer(64, block, 64, layer_nums[0])
        self.layer2 = make_layer(64 * block.expansion, block, 128, layer_nums[1], stride=2)
        self.layer3 = make_layer(128 * block.expansion, block, 256, layer_nums[2], stride=2)
        self.layer4 = make_layer(256 * block.expansion, block, 512, layer_nums[3], stride=2)
        # 平均池化层
        self.avg_pool = nn.AvgPool2d()
        # flattern层
        self.flatten = nn.Flatten()
        # 全连接层
        self.fc = nn.Dense(in_channels=input_channel, out_channels=num_classes)

    def construct(self, x):

        x = self.conv1(x)
        x = self.norm(x)
        x = self.relu(x)
        x = self.max_pool(x)

        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)

        x = self.avg_pool(x)
        x = self.flatten(x)
        x = self.fc(x)

        return x
def _resnet(model_url: str, block: Type[Union[ResidualBlockBase, ResidualBlock]],
            layers: List[int], num_classes: int, pretrained: bool, pretrained_ckpt: str,
            input_channel: int):
    model = ResNet(block, layers, num_classes, input_channel)

    if pretrained:
        # 加载预训练模型
        download(url=model_url, path=pretrained_ckpt, replace=True)
        param_dict = load_checkpoint(pretrained_ckpt)
        load_param_into_net(model, param_dict)

    return model


def resnet50(num_classes: int = 1000, pretrained: bool = False):
    """ResNet50模型"""
    resnet50_url = "https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/models/application/resnet50_224_new.ckpt"
    resnet50_ckpt = "./LoadPretrainedModel/resnet50_224_new.ckpt"
    return _resnet(resnet50_url, ResidualBlock, [3, 4, 6, 3], num_classes,
                   pretrained, resnet50_ckpt, 2048)
def _resnet(model_url: str, block: Type[Union[ResidualBlockBase, ResidualBlock]],
            layers: List[int], num_classes: int, pretrained: bool, pretrained_ckpt: str,
            input_channel: int):
    model = ResNet(block, layers, num_classes, input_channel)

    if pretrained:
        # 加载预训练模型
        download(url=model_url, path=pretrained_ckpt, replace=True)
        param_dict = load_checkpoint(pretrained_ckpt)
        load_param_into_net(model, param_dict)

    return model


def resnet50(num_classes: int = 1000, pretrained: bool = False):
    """ResNet50模型"""
    resnet50_url = "https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/models/application/resnet50_224_new.ckpt"
    resnet50_ckpt = "./LoadPretrainedModel/resnet50_224_new.ckpt"
    return _resnet(resnet50_url, ResidualBlock, [3, 4, 6, 3], num_classes,
                   pretrained, resnet50_ckpt, 2048)

1. model = ResNet(block, layers, num_classes, input_channel)初始化的模型是按照1000类别进行的（这样才能预加载），后续加载完后需要重置全连接层，使其out_channels=10

2. 数据在每个模块后的shape实例

   假设输入图像的大小为 224x224x3（宽度224，高度224，RGB通道3）。

   1. conv1:
      • 输入: 224x224x3
      • 输出: 112x112x64
      • 解释: 使用7x7卷积核，步长为2，输出大小为 (224-7)/2 + 1 = 112, 通道为64
   2. norm:
      • 输入: 112x112x64
      • 输出: 112x112x64
   3. max_pool:
      • 输入: 112x112x64
      • 输出: 56x56x64
      • 解释: 使用3x3池化核，步长为2，输出大小为 (112-3)/2 + 1 = 56
   4. layer1:
      • 输入: 56x56x64
      • 输出: 56x56x256 (假设使用的是 ResidualBlock，expansion=4)
   5. layer2:
      • 输入: 56x56x256
      • 输出: 28x28x512
      • 解释: 使用的步长为2，输出大小减半，通道数增加
   6. layer3:
      • 输入: 28x28x512
      • 输出: 14x14x1024
      • 解释: 使用的步长为2，输出大小减半，通道数增加
   7. layer4:
      • 输入: 14x14x1024
      • 输出: 7x7x2048
      • 解释: 使用的步长为2，输出大小减半，通道数增加
   8. avg_pool:
      • 输入: 7x7x2048
      • 输出: 1x1x2048
      • 解释: 全局平均池化，将每个通道的7x7区域的值取平均
   9. flatten:
      • 输入: 1x1x2048
      • 输出: 2048
      • 解释: 展平成一维张量
   10. fc:
       • 输入: 2048
       • 输出: num_classes（例如1000，后面通过重置改为10）

训练和评估

# 定义ResNet50网络
network = resnet50(pretrained=True)

# 全连接层输入层的大小
in_channel = network.fc.in_channels
fc = nn.Dense(in_channels=in_channel, out_channels=10)
# 重置全连接层
network.fc = fc

# 设置学习率
num_epochs = 5
lr = nn.cosine_decay_lr(min_lr=0.00001, max_lr=0.001, total_step=step_size_train * num_epochs,
                        step_per_epoch=step_size_train, decay_epoch=num_epochs)
# 定义优化器和损失函数
opt = nn.Momentum(params=network.trainable_params(), learning_rate=lr, momentum=0.9)
loss_fn = nn.SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')

#定义前向传播
def forward_fn(inputs, targets):
    logits = network(inputs)
    loss = loss_fn(logits, targets)
    return loss


grad_fn = ms.value_and_grad(forward_fn, None, opt.parameters)


def train_step(inputs, targets):
    loss, grads = grad_fn(inputs, targets)
    opt(grads)
    return loss
import os

# 创建迭代器
data_loader_train = dataset_train.create_tuple_iterator(num_epochs=num_epochs)
data_loader_val = dataset_val.create_tuple_iterator(num_epochs=num_epochs)

# 最佳模型存储路径
best_acc = 0
best_ckpt_dir = "./BestCheckpoint"
best_ckpt_path = "./BestCheckpoint/resnet50-best.ckpt"

if not os.path.exists(best_ckpt_dir):
    os.mkdir(best_ckpt_dir)
import mindspore.ops as ops


def train(data_loader, epoch):
    """模型训练"""
    losses = []
    network.set_train(True)

    for i, (images, labels) in enumerate(data_loader):
        loss = train_step(images, labels)
        if i % 100 == 0 or i == step_size_train - 1:
            print('Epoch: [%3d/%3d], Steps: [%3d/%3d], Train Loss: [%5.3f]' %
                  (epoch + 1, num_epochs, i + 1, step_size_train, loss))
        losses.append(loss)

    return sum(losses) / len(losses)


def evaluate(data_loader):
    """模型验证"""
    network.set_train(False)

    correct_num = 0.0  # 预测正确个数
    total_num = 0.0  # 预测总数

    for images, labels in data_loader:
        logits = network(images)
        pred = logits.argmax(axis=1)  # 预测结果
        correct = ops.equal(pred, labels).reshape((-1, ))
        correct_num += correct.sum().asnumpy()
        total_num += correct.shape[0]

    acc = correct_num / total_num  # 准确率

    return acc    

# 开始循环训练
print("Start Training Loop ...")

for epoch in range(num_epochs):
    curr_loss = train(data_loader_train, epoch)
    curr_acc = evaluate(data_loader_val)

    print("-" * 50)
    print("Epoch: [%3d/%3d], Average Train Loss: [%5.3f], Accuracy: [%5.3f]" % (
        epoch+1, num_epochs, curr_loss, curr_acc
    ))
    print("-" * 50)

    # 保存当前预测准确率最高的模型
    if curr_acc > best_acc:
        best_acc = curr_acc
        ms.save_checkpoint(network, best_ckpt_path)

print("=" * 80)
print(f"End of validation the best Accuracy is: {best_acc: 5.3f}, "
      f"save the best ckpt file in {best_ckpt_path}", flush=True)

1. loss是从训练集中计算
2. acc要从测试集中计算

加载和预测

import matplotlib.pyplot as plt


def visualize_model(best_ckpt_path, dataset_val):
    num_class = 10  # 对狼和狗图像进行二分类
    net = resnet50(num_class)
    # 加载模型参数
    param_dict = ms.load_checkpoint(best_ckpt_path)
    ms.load_param_into_net(net, param_dict)
    # 加载验证集的数据进行验证
    data = next(dataset_val.create_dict_iterator())
    images = data["image"]
    labels = data["label"]
    # 预测图像类别
    output = net(data['image'])
    pred = np.argmax(output.asnumpy(), axis=1)

    # 图像分类
    classes = []

    with open(data_dir + "/batches.meta.txt", "r") as f:
        for line in f:
            line = line.rstrip()
            if line:
                classes.append(line)

    # 显示图像及图像的预测值
    plt.figure()
    for i in range(6):
        plt.subplot(2, 3, i + 1)
        # 若预测正确，显示为蓝色；若预测错误，显示为红色
        color = 'blue' if pred[i] == labels.asnumpy()[i] else 'red'
        plt.title('predict:{}'.format(classes[pred[i]]), color=color)
        picture_show = np.transpose(images.asnumpy()[i], (1, 2, 0))
        mean = np.array([0.4914, 0.4822, 0.4465])
        std = np.array([0.2023, 0.1994, 0.2010])
        picture_show = std * picture_show + mean
        picture_show = np.clip(picture_show, 0, 1)
        plt.imshow(picture_show)
        plt.axis('off')

    plt.show()


# 使用测试数据集进行验证
visualize_model(best_ckpt_path=best_ckpt_path, dataset_val=dataset_val)