添加DBnet代码

models/backbone/MobilenetV3.py

+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+from torch import nn
+import torch.nn.functional as F
+
+
+class HSwish(nn.Module):
+    def forward(self, x):
+        out = x * F.relu6(x + 3, inplace=True) / 6
+        return out
+
+
+class HardSigmoid(nn.Module):
+    def __init__(self, slope=.2, offset=.5):
+        super().__init__()
+        self.slope = slope
+        self.offset = offset
+
+    def forward(self, x):
+        x = (self.slope * x) + self.offset
+        x = F.threshold(-x, -1, -1)
+        x = F.threshold(-x, 0, 0)
+        return x
+
+
+class ConvBNACT(nn.Module):
+    def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0, groups=1, act=None):
+        super().__init__()
+        self.conv = nn.Conv2d(in_channels=in_channels, out_channels=out_channels, kernel_size=kernel_size,
+                              stride=stride, padding=padding, groups=groups,
+                              bias=False)
+        self.bn = nn.BatchNorm2d(out_channels)
+        if act == 'relu':
+            self.act = nn.ReLU()
+        elif act == 'hard_swish':
+            self.act = HSwish()
+        elif act is None:
+            self.act = None
+
+    def forward(self, x):
+        x = self.conv(x)
+        x = self.bn(x)
+        if self.act is not None:
+            x = self.act(x)
+        return x
+
+
+class SEBlock(nn.Module):
+    def __init__(self, in_channels, out_channels, ratio=4):
+        super().__init__()
+        num_mid_filter = out_channels // ratio
+        self.pool = nn.AdaptiveAvgPool2d(1)
+        self.conv1 = nn.Conv2d(in_channels=in_channels, out_channels=num_mid_filter, kernel_size=1, bias=True)
+        self.relu1 = nn.ReLU()
+        self.conv2 = nn.Conv2d(in_channels=num_mid_filter, kernel_size=1, out_channels=out_channels, bias=True)
+        self.relu2 = HardSigmoid()
+
+    def forward(self, x):
+        attn = self.pool(x)
+        attn = self.conv1(attn)
+        attn = self.relu1(attn)
+        attn = self.conv2(attn)
+        attn = self.relu2(attn)
+        return x * attn
+
+
+class ResidualUnit(nn.Module):
+    def __init__(self, num_in_filter, num_mid_filter, num_out_filter, stride, kernel_size, act=None, use_se=False):
+        super().__init__()
+        self.conv0 = ConvBNACT(in_channels=num_in_filter, out_channels=num_mid_filter, kernel_size=1, stride=1,
+                               padding=0, act=act)
+
+        self.conv1 = ConvBNACT(in_channels=num_mid_filter, out_channels=num_mid_filter, kernel_size=kernel_size,
+                               stride=stride,
+                               padding=int((kernel_size - 1) // 2), act=act, groups=num_mid_filter)
+        if use_se:
+            self.se = SEBlock(in_channels=num_mid_filter, out_channels=num_mid_filter)
+        else:
+            self.se = None
+
+        self.conv2 = ConvBNACT(in_channels=num_mid_filter, out_channels=num_out_filter, kernel_size=1, stride=1,
+                               padding=0)
+        self.not_add = num_in_filter != num_out_filter or stride != 1
+
+    def forward(self, x):
+        y = self.conv0(x)
+        y = self.conv1(y)
+        if self.se is not None:
+            y = self.se(y)
+        y = self.conv2(y)
+        if not self.not_add:
+            y = x + y
+        return y
+
+
+class MobileNetV3(nn.Module):
+    def __init__(self, in_channels=3, **kwargs):
+        """
+        the MobilenetV3 backbone network for detection module.
+        Args:
+            params(dict): the super parameters for build network
+        """
+        super().__init__()
+        self.scale = kwargs.get('scale', 0.5)
+        model_name = kwargs.get('model_name', 'large')
+        self.inplanes = 16
+        if model_name == "large":
+            self.cfg = [
+                # k, exp, c,  se,     nl,  s,
+                [3, 16, 16, False, 'relu', 1],
+                [3, 64, 24, False, 'relu', 2],
+                [3, 72, 24, False, 'relu', 1],
+                [5, 72, 40, True, 'relu', 2],
+                [5, 120, 40, True, 'relu', 1],
+                [5, 120, 40, True, 'relu', 1],
+                [3, 240, 80, False, 'hard_swish', 2],
+                [3, 200, 80, False, 'hard_swish', 1],
+                [3, 184, 80, False, 'hard_swish', 1],
+                [3, 184, 80, False, 'hard_swish', 1],
+                [3, 480, 112, True, 'hard_swish', 1],
+                [3, 672, 112, True, 'hard_swish', 1],
+                [5, 672, 160, True, 'hard_swish', 2],
+                [5, 960, 160, True, 'hard_swish', 1],
+                [5, 960, 160, True, 'hard_swish', 1],
+            ]
+            self.cls_ch_squeeze = 960
+            self.cls_ch_expand = 1280
+        elif model_name == "small":
+            self.cfg = [
+                # k, exp, c,  se,     nl,  s,
+                [3, 16, 16, True, 'relu', 2],
+                [3, 72, 24, False, 'relu', 2],
+                [3, 88, 24, False, 'relu', 1],
+                [5, 96, 40, True, 'hard_swish', 2],
+                [5, 240, 40, True, 'hard_swish', 1],
+                [5, 240, 40, True, 'hard_swish', 1],
+                [5, 120, 48, True, 'hard_swish', 1],
+                [5, 144, 48, True, 'hard_swish', 1],
+                [5, 288, 96, True, 'hard_swish', 2],
+                [5, 576, 96, True, 'hard_swish', 1],
+                [5, 576, 96, True, 'hard_swish', 1],
+            ]
+            self.cls_ch_squeeze = 576
+            self.cls_ch_expand = 1280
+        else:
+            raise NotImplementedError("mode[" + model_name +
+                                      "_model] is not implemented!")
+
+        supported_scale = [0.35, 0.5, 0.75, 1.0, 1.25]
+        assert self.scale in supported_scale, \
+            "supported scale are {} but input scale is {}".format(supported_scale, self.scale)
+
+        scale = self.scale
+        inplanes = self.inplanes
+        cfg = self.cfg
+        cls_ch_squeeze = self.cls_ch_squeeze
+        # conv1
+        self.conv1 = ConvBNACT(in_channels=in_channels,
+                               out_channels=self.make_divisible(inplanes * scale),
+                               kernel_size=3,
+                               stride=2,
+                               padding=1,
+                               groups=1,
+                               act='hard_swish')
+        i = 0
+        inplanes = self.make_divisible(inplanes * scale)
+        self.stages = nn.ModuleList()
+        block_list = []
+        self.out_channels = []
+        for layer_cfg in cfg:
+            if layer_cfg[5] == 2 and i > 2:
+                self.out_channels.append(inplanes)
+                self.stages.append(nn.Sequential(*block_list))
+                block_list = []
+            block = ResidualUnit(num_in_filter=inplanes,
+                                 num_mid_filter=self.make_divisible(scale * layer_cfg[1]),
+                                 num_out_filter=self.make_divisible(scale * layer_cfg[2]),
+                                 act=layer_cfg[4],
+                                 stride=layer_cfg[5],
+                                 kernel_size=layer_cfg[0],
+                                 use_se=layer_cfg[3])
+            block_list.append(block)
+            inplanes = self.make_divisible(scale * layer_cfg[2])
+            i += 1
+        self.stages.append(nn.Sequential(*block_list))
+        self.conv2 = ConvBNACT(
+            in_channels=inplanes,
+            out_channels=self.make_divisible(scale * cls_ch_squeeze),
+            kernel_size=1,
+            stride=1,
+            padding=0,
+            groups=1,
+            act='hard_swish')
+        self.out_channels.append(self.make_divisible(scale * cls_ch_squeeze))
+
+    def make_divisible(self, v, divisor=8, min_value=None):
+        if min_value is None:
+            min_value = divisor
+        new_v = max(min_value, int(v + divisor / 2) // divisor * divisor)
+        if new_v < 0.9 * v:
+            new_v += divisor
+        return new_v
+
+    def forward(self, x):
+        x = self.conv1(x)
+        out = []
+        for stage in self.stages:
+            x = stage(x)
+            out.append(x)
+        out[-1] = self.conv2(out[-1])
+        return out

models/backbone/__init__.py

+# -*- coding: utf-8 -*-
+# @Time    : 2019/8/23 21:54
+# @Author  : zhoujun
+
+from .resnet import *
+from .resnest import *
+from .shufflenetv2 import *
+from .MobilenetV3 import MobileNetV3
+
+__all__ = ['build_backbone']
+
+support_backbone = ['resnet18', 'deformable_resnet18', 'deformable_resnet50',
+                    'resnet50', 'resnet34', 'resnet101', 'resnet152',
+                    'resnest50', 'resnest101', 'resnest200', 'resnest269',
+                    'shufflenet_v2_x0_5', 'shufflenet_v2_x1_0', 'shufflenet_v2_x1_5', 'shufflenet_v2_x2_0',
+                    'MobileNetV3']
+
+
+def build_backbone(backbone_name, **kwargs):
+    assert backbone_name in support_backbone, f'all support backbone is {support_backbone}'
+    backbone = eval(backbone_name)(**kwargs)
+    return backbone

models/backbone/resnest/__init__.py

+from .resnest import *
\ No newline at end of file

models/backbone/resnest/ablation.py

+##+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+## Created by: Hang Zhang
+## Email: zhanghang0704@gmail.com
+## Copyright (c) 2020
+##
+## LICENSE file in the root directory of this source tree 
+##+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+"""ResNeSt ablation study models"""
+
+import torch
+from .resnet import ResNet, Bottleneck
+
+__all__ = ['resnest50_fast_1s1x64d', 'resnest50_fast_2s1x64d', 'resnest50_fast_4s1x64d',
+           'resnest50_fast_1s2x40d', 'resnest50_fast_2s2x40d', 'resnest50_fast_4s2x40d',
+           'resnest50_fast_1s4x24d']
+
+_url_format = 'https://hangzh.s3.amazonaws.com/encoding/models/{}-{}.pth'
+
+_model_sha256 = {name: checksum for checksum, name in [
+    ('d8fbf808', 'resnest50_fast_1s1x64d'),
+    ('44938639', 'resnest50_fast_2s1x64d'),
+    ('f74f3fc3', 'resnest50_fast_4s1x64d'),
+    ('32830b84', 'resnest50_fast_1s2x40d'),
+    ('9d126481', 'resnest50_fast_2s2x40d'),
+    ('41d14ed0', 'resnest50_fast_4s2x40d'),
+    ('d4a4f76f', 'resnest50_fast_1s4x24d'),
+    ]}
+
+def short_hash(name):
+    if name not in _model_sha256:
+        raise ValueError('Pretrained model for {name} is not available.'.format(name=name))
+    return _model_sha256[name][:8]
+
+resnest_model_urls = {name: _url_format.format(name, short_hash(name)) for
+    name in _model_sha256.keys()
+}
+
+def resnest50_fast_1s1x64d(pretrained=False, root='~/.encoding/models', **kwargs):
+    model = ResNet(Bottleneck, [3, 4, 6, 3],
+                   radix=1, groups=1, bottleneck_width=64,
+                   deep_stem=True, stem_width=32, avg_down=True,
+                   avd=True, avd_first=True, **kwargs)
+    if pretrained:
+        model.load_state_dict(torch.hub.load_state_dict_from_url(
+            resnest_model_urls['resnest50_fast_1s1x64d'], progress=True, check_hash=True))
+    return model
+
+def resnest50_fast_2s1x64d(pretrained=False, root='~/.encoding/models', **kwargs):
+    model = ResNet(Bottleneck, [3, 4, 6, 3],
+                   radix=2, groups=1, bottleneck_width=64,
+                   deep_stem=True, stem_width=32, avg_down=True,
+                   avd=True, avd_first=True, **kwargs)
+    if pretrained:
+        model.load_state_dict(torch.hub.load_state_dict_from_url(
+            resnest_model_urls['resnest50_fast_2s1x64d'], progress=True, check_hash=True))
+    return model
+
+def resnest50_fast_4s1x64d(pretrained=False, root='~/.encoding/models', **kwargs):
+    model = ResNet(Bottleneck, [3, 4, 6, 3],
+                   radix=4, groups=1, bottleneck_width=64,
+                   deep_stem=True, stem_width=32, avg_down=True,
+                   avd=True, avd_first=True, **kwargs)
+    if pretrained:
+        model.load_state_dict(torch.hub.load_state_dict_from_url(
+            resnest_model_urls['resnest50_fast_4s1x64d'], progress=True, check_hash=True))
+    return model
+
+def resnest50_fast_1s2x40d(pretrained=False, root='~/.encoding/models', **kwargs):
+    model = ResNet(Bottleneck, [3, 4, 6, 3],
+                   radix=1, groups=2, bottleneck_width=40,
+                   deep_stem=True, stem_width=32, avg_down=True,
+                   avd=True, avd_first=True, **kwargs)
+    if pretrained:
+        model.load_state_dict(torch.hub.load_state_dict_from_url(
+            resnest_model_urls['resnest50_fast_1s2x40d'], progress=True, check_hash=True))
+    return model
+
+def resnest50_fast_2s2x40d(pretrained=False, root='~/.encoding/models', **kwargs):
+    model = ResNet(Bottleneck, [3, 4, 6, 3],
+                   radix=2, groups=2, bottleneck_width=40,
+                   deep_stem=True, stem_width=32, avg_down=True,
+                   avd=True, avd_first=True, **kwargs)
+    if pretrained:
+        model.load_state_dict(torch.hub.load_state_dict_from_url(
+            resnest_model_urls['resnest50_fast_2s2x40d'], progress=True, check_hash=True))
+    return model
+
+def resnest50_fast_4s2x40d(pretrained=False, root='~/.encoding/models', **kwargs):
+    model = ResNet(Bottleneck, [3, 4, 6, 3],
+                   radix=4, groups=2, bottleneck_width=40,
+                   deep_stem=True, stem_width=32, avg_down=True,
+                   avd=True, avd_first=True, **kwargs)
+    if pretrained:
+        model.load_state_dict(torch.hub.load_state_dict_from_url(
+            resnest_model_urls['resnest50_fast_4s2x40d'], progress=True, check_hash=True))
+    return model
+
+def resnest50_fast_1s4x24d(pretrained=False, root='~/.encoding/models', **kwargs):
+    model = ResNet(Bottleneck, [3, 4, 6, 3],
+                   radix=1, groups=4, bottleneck_width=24,
+                   deep_stem=True, stem_width=32, avg_down=True,
+                   avd=True, avd_first=True, **kwargs)
+    if pretrained:
+        model.load_state_dict(torch.hub.load_state_dict_from_url(
+            resnest_model_urls['resnest50_fast_1s4x24d'], progress=True, check_hash=True))
+    return model