mmclassification/mmcls/models/backbones/seresnext.py

import math

from mmcv.cnn import build_conv_layer, build_norm_layer

from ..builder import BACKBONES
from .resnet import ResLayer
from .seresnet import SEBottleneck as _SEBottleneck
from .seresnet import SEResNet


class SEBottleneck(_SEBottleneck):
    """SEBottleneck block for SEResNeXt.

    Args:
        inplanes (int): inplanes of block.
        planes (int): planes of block.
        groups (int): group of convolution.
        base_width (int): Base width of resnext.
        base_channels (int): Number of base channels of hidden layer.
        stride (int): stride of the block. Default: 1
        dilation (int): dilation of convolution. Default: 1
        dbownsample (nn.Module): downsample operation on identity branch.
            Default: None
        se_ratio (int): Squeeze ratio in SELayer. Default: 16
        style (str): `pytorch` or `caffe`. If set to "pytorch", the stride-two
            layer is the 3x3 conv layer, otherwise the stride-two layer is
            the first 1x1 conv layer.
        conv_cfg (dict): dictionary to construct and config conv layer.
            Default: None
        norm_cfg (dict): dictionary to construct and config norm layer.
            Default: dict(type='BN')
        with_cp (bool): Use checkpoint or not. Using checkpoint will save some
            memory while slowing down the training speed.
    """

    expansion = 4

    def __init__(self,
                 inplanes,
                 planes,
                 groups=1,
                 base_width=4,
                 base_channels=64,
                 se_ratio=16,
                 **kwargs):
        super(SEBottleneck, self).__init__(inplanes, planes, se_ratio,
                                           **kwargs)

        if groups == 1:
            width = self.planes
        else:
            width = math.floor(self.planes *
                               (base_width / base_channels)) * groups

        self.norm1_name, norm1 = build_norm_layer(
            self.norm_cfg, width, postfix=1)
        self.norm2_name, norm2 = build_norm_layer(
            self.norm_cfg, width, postfix=2)
        self.norm3_name, norm3 = build_norm_layer(
            self.norm_cfg, self.planes * self.expansion, postfix=3)

        self.conv1 = build_conv_layer(
            self.conv_cfg,
            self.inplanes,
            width,
            kernel_size=1,
            stride=self.conv1_stride,
            bias=False)
        self.add_module(self.norm1_name, norm1)
        self.conv2 = build_conv_layer(
            self.conv_cfg,
            width,
            width,
            kernel_size=3,
            stride=self.conv2_stride,
            padding=self.dilation,
            dilation=self.dilation,
            groups=groups,
            bias=False)

        self.add_module(self.norm2_name, norm2)
        self.conv3 = build_conv_layer(
            self.conv_cfg,
            width,
            self.planes * self.expansion,
            kernel_size=1,
            bias=False)
        self.add_module(self.norm3_name, norm3)


@BACKBONES.register_module()
class SEResNeXt(SEResNet):
    """SEResNeXt backbone.

    Args:
        groups (int): Group of seresnext.
        base_width (int): Base width of resnext.
        depth (int): Depth of resnext, from {50, 101, 152}.
        in_channels (int): Number of input image channels. Default: 3.
        base_channels (int): Number of base channels of hidden layer.
        num_stages (int): Resnet stages. Default: 4.
        strides (Sequence[int]): Strides of the first block of each stage.
        dilations (Sequence[int]): Dilation of each stage.
        out_indices (Sequence[int]): Output from which stages.
        se_ratio (int): Squeeze ratio in SELayer. Default: 16
        style (str): `pytorch` or `caffe`. If set to "pytorch", the stride-two
            layer is the 3x3 conv layer, otherwise the stride-two layer is
            the first 1x1 conv layer.
        frozen_stages (int): Stages to be frozen (all param fixed). -1 means
            not freezing any parameters.
        norm_cfg (dict): dictionary to construct and config norm layer.
        norm_eval (bool): Whether to set norm layers to eval mode, namely,
            freeze running stats (mean and var). Note: Effect on Batch Norm
            and its variants only.
        with_cp (bool): Use checkpoint or not. Using checkpoint will save some
            memory while slowing down the training speed.
        zero_init_residual (bool): whether to use zero init for last norm layer
            in resblocks to let them behave as identity.
    """

    arch_settings = {
        50: (SEBottleneck, (3, 4, 6, 3)),
        101: (SEBottleneck, (3, 4, 23, 3)),
        152: (SEBottleneck, (3, 8, 36, 3))
    }

    def __init__(self, groups=1, base_width=4, **kwargs):
        self.groups = groups
        self.base_width = base_width
        super(SEResNeXt, self).__init__(**kwargs)

    def make_res_layer(self, **kwargs):
        return ResLayer(
            groups=self.groups,
            base_width=self.base_width,
            base_channels=self.base_channels,
            **kwargs)