pytorch中MobileNetV2分类模型的源码注解

ゝ一世哀愁。 2022-12-22 00:00 230阅读 0赞

上一篇[《pytorch卷积操作nn.Conv中的groups参数用法解释》][pytorch_nn.Conv_groups]中简单介绍了MobileNet中使用的深度可分离卷积，以及pytorch中在实现深度可分离卷积时使用的nn.Conv模块的groups参数。本篇通过代码注释的方式解释一下pytorch中MobileNetV2网络的具体实现过程。

from torch import nn
    from .utils import load_state_dict_from_url
    
    
    __all__ = ['MobileNetV2', 'mobilenet_v2']
    
    
    model_urls = {
        'mobilenet_v2': 'https://download.pytorch.org/models/mobilenet_v2-b0353104.pth',
    }
    
    
    def _make_divisible(v, divisor, min_value=None):
        """
        This function is taken from the original tf repo.
        It ensures that all layers have a channel number that is divisible by 8
        It can be seen here:
        https://github.com/tensorflow/models/blob/master/research/slim/nets/mobilenet/mobilenet.py
        :param v:
        :param divisor:
        :param min_value:
        :return:
        """
        if min_value is None:
            min_value = divisor
        new_v = max(min_value, int(v + divisor / 2) // divisor * divisor)
        # Make sure that round down does not go down by more than 10%.
        if new_v < 0.9 * v:
            new_v += divisor
        return new_v
    
    
    class ConvBNReLU(nn.Sequential):
        """
            深度可分离卷积：
            1、convolution on each input channel
            2、Batch Normalize
            3、Relu6
        """
        def __init__(self, in_planes, out_planes, kernel_size=3, stride=1, groups=1, norm_layer=None):
            padding = (kernel_size - 1) // 2
            if norm_layer is None:
                norm_layer = nn.BatchNorm2d
            # 深度可分离卷积，卷积的groups等于卷积输入的通道数，实现每个通道单独进行卷积的目的
            super(ConvBNReLU, self).__init__(
                nn.Conv2d(in_planes, out_planes, kernel_size, stride, padding, groups=groups, bias=False),
                norm_layer(out_planes),
                nn.ReLU6(inplace=True)
            )
    
    
    class InvertedResidual(nn.Module):
        def __init__(self, inp, oup, stride, expand_ratio, norm_layer=None):
            super(InvertedResidual, self).__init__()
            self.stride = stride
            # 有两种卷积stride模式，stride=1和stride=2
            assert stride in [1, 2]
    
            if norm_layer is None:
                norm_layer = nn.BatchNorm2d
    
            hidden_dim = int(round(inp * expand_ratio))
            # 当stride=1且输入、输出通道数相等时，使用short-cut残差连接
            # 当stride=2时不使用short-cut残差连接
            self.use_res_connect = self.stride == 1 and inp == oup
    
            layers = []
            if expand_ratio != 1:
                # pw 使用扩展因子提升通道数
                layers.append(ConvBNReLU(inp, hidden_dim, kernel_size=1, norm_layer=norm_layer))
            layers.extend([
                # dw， 深度可分离卷积，卷积groups等于输入通道数
                ConvBNReLU(hidden_dim, hidden_dim, stride=stride, groups=hidden_dim, norm_layer=norm_layer),
                # pw-linear， 使用1x1卷积降维
                nn.Conv2d(hidden_dim, oup, 1, 1, 0, bias=False),
                norm_layer(oup),
            ])
            self.conv = nn.Sequential(*layers)
    
        def forward(self, x):
            if self.use_res_connect:
                # stride=1且输入输出维度相等，使用残差连接
                return x + self.conv(x)
            else:
                # 不使用残差连接
                return self.conv(x)
    
    
    class MobileNetV2(nn.Module):
        def __init__(self,
                     num_classes=1000,
                     width_mult=1.0,
                     inverted_residual_setting=None,
                     round_nearest=8,
                     block=None,
                     norm_layer=None):
            """
            MobileNet V2 main class
    
            Args:
                num_classes (int): Number of classes
                width_mult (float): Width multiplier - adjusts number of channels in each layer by this amount
                inverted_residual_setting: Network structure
                round_nearest (int): Round the number of channels in each layer to be a multiple of this number
                Set to 1 to turn off rounding
                block: Module specifying inverted residual building block for mobilenet
                norm_layer: Module specifying the normalization layer to use
    
            """
            super(MobileNetV2, self).__init__()
    
            if block is None:
                block = InvertedResidual
    
            if norm_layer is None:
                norm_layer = nn.BatchNorm2d
    
            input_channel = 32
            last_channel = 1280
    
            if inverted_residual_setting is None:
                inverted_residual_setting = [
                    # t, c, n, s
                    [1, 16, 1, 1],
                    [6, 24, 2, 2],
                    [6, 32, 3, 2],
                    [6, 64, 4, 2],
                    [6, 96, 3, 1],
                    [6, 160, 3, 2],
                    [6, 320, 1, 1],
                ]
    
            # only check the first element, assuming user knows t,c,n,s are required
            if len(inverted_residual_setting) == 0 or len(inverted_residual_setting[0]) != 4:
                raise ValueError("inverted_residual_setting should be non-empty "
                                 "or a 4-element list, got {}".format(inverted_residual_setting))
    
            # building first layer
            input_channel = _make_divisible(input_channel * width_mult, round_nearest)
            self.last_channel = _make_divisible(last_channel * max(1.0, width_mult), round_nearest)
            # first 3x3 convolution
            features = [ConvBNReLU(3, input_channel, stride=2, norm_layer=norm_layer)]
            # building inverted residual blocks
            for t, c, n, s in inverted_residual_setting:
                output_channel = _make_divisible(c * width_mult, round_nearest)
                for i in range(n):
                    # 在每个反转卷积中，除第一个卷积外其他卷积操作均使用stride=1
                    stride = s if i == 0 else 1
                    features.append(block(input_channel, output_channel, stride, expand_ratio=t, norm_layer=norm_layer))
                    input_channel = output_channel
            # building last several layers, 使用1x1卷积降低通道数
            features.append(ConvBNReLU(input_channel, self.last_channel, kernel_size=1, norm_layer=norm_layer))
            # make it nn.Sequential
            # 底层CNN进行特征提取
            self.features = nn.Sequential(*features)
    
            # building classifier
            # 构建上层分类器
            self.classifier = nn.Sequential(
                nn.Dropout(0.2),
                nn.Linear(self.last_channel, num_classes),
            )
    
            # weight initialization
            # 权重参数初始化
            for m in self.modules():
                if isinstance(m, nn.Conv2d):
                    nn.init.kaiming_normal_(m.weight, mode='fan_out')
                    if m.bias is not None:
                        nn.init.zeros_(m.bias)
                elif isinstance(m, (nn.BatchNorm2d, nn.GroupNorm)):
                    nn.init.ones_(m.weight)
                    nn.init.zeros_(m.bias)
                elif isinstance(m, nn.Linear):
                    nn.init.normal_(m.weight, 0, 0.01)
                    nn.init.zeros_(m.bias)
    
        def _forward_impl(self, x):
            # This exists since TorchScript doesn't support inheritance, so the superclass method
            # (this one) needs to have a name other than `forward` that can be accessed in a subclass
            x = self.features(x)
            # Cannot use "squeeze" as batch-size can be 1 => must use reshape with x.shape[0]
            # 通常在卷积网络中，需要把最后一个卷积的结果进行flatten压平，然后使用全连接层，并最终输出与分类类别数一致的输出数量
            # 在MobileNetV2中 在最后一个卷积结果之后使用自适应平均池化，将卷积结果[N, C, H, W]自适应平均池化后得到[N, C, 1, 1],起到
            # 代替全连接层的作用，使得网络可以自动处理任意大小的图片
            # 原始论文中使用的训练数据是224x224，但是我实际使用中使用的是40x40大小，只需要把self.classifier部分的分类器头换掉即可
            x = nn.functional.adaptive_avg_pool2d(x, 1).reshape(x.shape[0], -1)
            x = self.classifier(x)
            return x
    
        def forward(self, x):
            return self._forward_impl(x)
    
    
    def mobilenet_v2(pretrained=False, progress=True, **kwargs):
        """
        Constructs a MobileNetV2 architecture from
        `"MobileNetV2: Inverted Residuals and Linear Bottlenecks" <https://arxiv.org/abs/1801.04381>`_.
    
        Args:
            pretrained (bool): If True, returns a model pre-trained on ImageNet
            progress (bool): If True, displays a progress bar of the download to stderr
        """
        model = MobileNetV2(**kwargs)
        if pretrained:
            # 加载在ImageNet上预训练的参数
            state_dict = load_state_dict_from_url(model_urls['mobilenet_v2'],
                                                  progress=progress)
            model.load_state_dict(state_dict)
        return model

[pytorch_nn.Conv_groups]: https://blog.csdn.net/cxx654/article/details/109681004