def __init__()

in easycv/models/ocr/backbones/rec_mv1_enhance.py [0:0]

• 104 lines of code
• 3 McCabe index (conditional complexity)

    def __init__(self,
                 in_channels=3,
                 scale=0.5,
                 last_conv_stride=1,
                 last_pool_type='max',
                 **kwargs):
        super().__init__()
        self.scale = scale
        self.block_list = []

        self.conv1 = ConvBNLayer(
            num_channels=in_channels,
            filter_size=3,
            channels=3,
            num_filters=int(32 * scale),
            stride=2,
            padding=1)

        conv2_1 = DepthwiseSeparable(
            num_channels=int(32 * scale),
            num_filters1=32,
            num_filters2=64,
            num_groups=32,
            stride=1,
            scale=scale)
        self.block_list.append(conv2_1)

        conv2_2 = DepthwiseSeparable(
            num_channels=int(64 * scale),
            num_filters1=64,
            num_filters2=128,
            num_groups=64,
            stride=1,
            scale=scale)
        self.block_list.append(conv2_2)

        conv3_1 = DepthwiseSeparable(
            num_channels=int(128 * scale),
            num_filters1=128,
            num_filters2=128,
            num_groups=128,
            stride=1,
            scale=scale)
        self.block_list.append(conv3_1)

        conv3_2 = DepthwiseSeparable(
            num_channels=int(128 * scale),
            num_filters1=128,
            num_filters2=256,
            num_groups=128,
            stride=(2, 1),
            scale=scale)
        self.block_list.append(conv3_2)

        conv4_1 = DepthwiseSeparable(
            num_channels=int(256 * scale),
            num_filters1=256,
            num_filters2=256,
            num_groups=256,
            stride=1,
            scale=scale)
        self.block_list.append(conv4_1)

        conv4_2 = DepthwiseSeparable(
            num_channels=int(256 * scale),
            num_filters1=256,
            num_filters2=512,
            num_groups=256,
            stride=(2, 1),
            scale=scale)
        self.block_list.append(conv4_2)

        for _ in range(5):
            conv5 = DepthwiseSeparable(
                num_channels=int(512 * scale),
                num_filters1=512,
                num_filters2=512,
                num_groups=512,
                stride=1,
                dw_size=5,
                padding=2,
                scale=scale,
                use_se=False)
            self.block_list.append(conv5)

        conv5_6 = DepthwiseSeparable(
            num_channels=int(512 * scale),
            num_filters1=512,
            num_filters2=1024,
            num_groups=512,
            stride=(2, 1),
            dw_size=5,
            padding=2,
            scale=scale,
            use_se=True)
        self.block_list.append(conv5_6)

        conv6 = DepthwiseSeparable(
            num_channels=int(1024 * scale),
            num_filters1=1024,
            num_filters2=1024,
            num_groups=1024,
            stride=last_conv_stride,
            dw_size=5,
            padding=2,
            use_se=True,
            scale=scale)
        self.block_list.append(conv6)

        self.block_list = nn.Sequential(*self.block_list)
        if last_pool_type == 'avg':
            self.pool = nn.AvgPool2d(kernel_size=2, stride=2, padding=0)
        else:
            self.pool = nn.MaxPool2d(kernel_size=2, stride=2, padding=0)
        self.out_channels = int(1024 * scale)