ShufferNetV1 V2 网络+pytorch源码

最新推荐文章于 2023-01-15 10:20:26 发布

为了维护世界和平_

最新推荐文章于 2023-01-15 10:20:26 发布

阅读量573

点赞数 1

分类专栏：深度学习文章标签： pytorch 深度学习

本文链接：https://blog.csdn.net/WANGYONGZIXUE/article/details/120107729

版权

深度学习专栏收录该内容

23 篇文章 2 订阅

订阅专栏

一、ShufferNet V1

channel shuffer的思想
1）将1x1卷积替换成group卷积(ShufferNet Unit 中全是GConv和DWconv)
2）使用channel shuffer，打乱通道

在这里插入图片描述

def channel_shuffle(x: Tensor, groups: int) -> Tensor:
    batchsize, num_channels, height, width = x.size()
    channels_per_group = num_channels // groups

    # reshape
    x = x.view(batchsize, groups,
               channels_per_group, height, width)

    x = torch.transpose(x, 1, 2).contiguous()

    # flatten
    x = x.view(batchsize, -1, height, width)

    return x

channel shuffer理解

在这里插入图片描述

二、ShufferNetV2

1）FLOPs 不能准确评判模型的执行时间，还有MAC
2）不同平台（GPU 、ARM）耗时不一致

网络设计的原则
G1:当卷积层的输入特征矩阵与输出特征矩阵channel相等时MAC最小（保持FLOPs不变时）
实验数据：
在这里插入图片描述
c1:c2 输入与输出比
1）GPU x1 比值越小，处理数据量越少，并且1:1 到 1:12 对应1480,748 相差两倍而ARM x1 76.2 57.6 相差不到一倍

G2:当GConv的groups增大时（保持FLOPs不变时），MAC也会增大。

实验数据：
在这里插入图片描述
GPU上X1 列g=1,8 相差4倍，在CPU X1列，相差1倍不到

G3：网络设计的碎片化程度越高，速度越慢
碎片，就是分支。碎片化能增加准确率但是增加计算量

实验数据：
在这里插入图片描述

G4:element-wise操作带来的影响是不可忽视的
element-wise的操作有 ReLU AddTensor AddBiase 等，尽管他们的FLOPs很小，但是MAC却很大。

实验数据：
在这里插入图片描述

v1 v2网络block结构
在这里插入图片描述
(a)(b)是shuffernetv1所使用的结构（c）(d)是shuffernetv2的结构

c)block中1x1的group卷积替换成普通卷积 [满足G2]
c)利用channel split分出两个分支[满足G3]，右边的分支包含了3个卷积操作，并且其通道数都一样[满足G1],左边的分支不做任何操作
c) concat拼接输入与输出通道一致 [满足G1]
d)分支 concat channel 增加两倍

网络结构
在这里插入图片描述

三、源码

shuffernetv2网络

class InvertedResidual(nn.Module):
    def __init__(
        self,
        inp: int,
        oup: int,
        stride: int
    ) -> None:
        super(InvertedResidual, self).__init__()

        if not (1 <= stride <= 3):
            raise ValueError('illegal stride value')
        self.stride = stride

        branch_features = oup // 2
        assert (self.stride != 1) or (inp == branch_features << 1)

        if self.stride > 1:
            self.branch1 = nn.Sequential(
                self.depthwise_conv(inp, inp, kernel_size=3, stride=self.stride, padding=1),
                nn.BatchNorm2d(inp),
                nn.Conv2d(inp, branch_features, kernel_size=1, stride=1, padding=0, bias=False),
                nn.BatchNorm2d(branch_features),
                nn.ReLU(inplace=True),
            )
        else:
            self.branch1 = nn.Sequential()

        self.branch2 = nn.Sequential(
            nn.Conv2d(inp if (self.stride > 1) else branch_features,
                      branch_features, kernel_size=1, stride=1, padding=0, bias=False),
            nn.BatchNorm2d(branch_features),
            nn.ReLU(inplace=True),
            self.depthwise_conv(branch_features, branch_features, kernel_size=3, stride=self.stride, padding=1),
            nn.BatchNorm2d(branch_features),
            nn.Conv2d(branch_features, branch_features, kernel_size=1, stride=1, padding=0, bias=False),
            nn.BatchNorm2d(branch_features),
            nn.ReLU(inplace=True),
        )

    @staticmethod
    def depthwise_conv(
        i: int,
        o: int,
        kernel_size: int,
        stride: int = 1,
        padding: int = 0,
        bias: bool = False
    ) -> nn.Conv2d:
        return nn.Conv2d(i, o, kernel_size, stride, padding, bias=bias, groups=i)

    def forward(self, x: Tensor) -> Tensor:
        if self.stride == 1:
            x1, x2 = x.chunk(2, dim=1)
            out = torch.cat((x1, self.branch2(x2)), dim=1)
        else:
            out = torch.cat((self.branch1(x), self.branch2(x)), dim=1)

        out = channel_shuffle(out, 2)

        return out


class ShuffleNetV2(nn.Module):
    def __init__(
        self,
        stages_repeats: List[int],
        stages_out_channels: List[int],
        num_classes: int = 1000,
        inverted_residual: Callable[..., nn.Module] = InvertedResidual
    ) -> None:
        super(ShuffleNetV2, self).__init__()

        if len(stages_repeats) != 3:
            raise ValueError('expected stages_repeats as list of 3 positive ints')
        if len(stages_out_channels) != 5:
            raise ValueError('expected stages_out_channels as list of 5 positive ints')
        self._stage_out_channels = stages_out_channels

        input_channels = 3
        output_channels = self._stage_out_channels[0]
        self.conv1 = nn.Sequential(
            nn.Conv2d(input_channels, output_channels, 3, 2, 1, bias=False),
            nn.BatchNorm2d(output_channels),
            nn.ReLU(inplace=True),
        )
        input_channels = output_channels

        self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)

        # Static annotations for mypy
        self.stage2: nn.Sequential
        self.stage3: nn.Sequential
        self.stage4: nn.Sequential
        stage_names = ['stage{}'.format(i) for i in [2, 3, 4]]
        for name, repeats, output_channels in zip(
                stage_names, stages_repeats, self._stage_out_channels[1:]):
            seq = [inverted_residual(input_channels, output_channels, 2)]
            for i in range(repeats - 1):
                seq.append(inverted_residual(output_channels, output_channels, 1))
            setattr(self, name, nn.Sequential(*seq))
            input_channels = output_channels

        output_channels = self._stage_out_channels[-1]
        self.conv5 = nn.Sequential(
            nn.Conv2d(input_channels, output_channels, 1, 1, 0, bias=False),
            nn.BatchNorm2d(output_channels),
            nn.ReLU(inplace=True),
        )

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

    def _forward_impl(self, x: Tensor) -> Tensor:
        # See note [TorchScript super()]
        x = self.conv1(x)
        x = self.maxpool(x)
        x = self.stage2(x)
        x = self.stage3(x)
        x = self.stage4(x)
        x = self.conv5(x)
        x = x.mean([2, 3])  # globalpool
        x = self.fc(x)
        return x

    def forward(self, x: Tensor) -> Tensor:
        return self._forward_impl(x)


# def _shufflenetv2(arch: str, pretrained: bool, progress: bool, *args: Any, **kwargs: Any) -> ShuffleNetV2:
#     model = ShuffleNetV2(*args, **kwargs)
#
#     if pretrained:
#         model_url = model_urls[arch]
#         if model_url is None:
#             raise NotImplementedError('pretrained {} is not supported as of now'.format(arch))
#         else:
#             state_dict = load_state_dict_from_url(model_url, progress=progress)
#             model.load_state_dict(state_dict)
#
#     return model


def shufflenet_v2_x0_5(num_classes=1000):
    """
    Constructs a ShuffleNetV2 with 0.5x output channels, as described in
    `"ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design"
    <https://arxiv.org/abs/1807.11164>`_.
    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
    """
    # return _shufflenetv2('shufflenetv2_x0.5', pretrained, progress,
    #                      [4, 8, 4], [24, 48, 96, 192, 1024], **kwargs)

    model = ShuffleNetV2(stages_repeats=[4, 8, 4],
                         stages_out_channels=[24, 48, 96, 192, 1024],
                         num_classes=num_classes)

    return model


def shufflenet_v2_x1_0(num_classes=1000):
    """
    Constructs a ShuffleNetV2 with 1.0x output channels, as described in
    `"ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design"
    <https://arxiv.org/abs/1807.11164>`_.
    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 = ShuffleNetV2(stages_repeats=[4, 8, 4],
                         stages_out_channels=[24, 116, 232, 464, 1024],
                         num_classes=num_classes)

    return model