REPVGG.py

import torch.nn as nn
import numpy as np
import torch
import copy

def conv_bn(in_channels, out_channels, kernel_size, stride, padding, groups=1):
    result = nn.Sequential()
    result.add_module('conv', nn.Conv2d(in_channels=in_channels, out_channels=out_channels,
                                                  kernel_size=kernel_size, stride=stride, padding=padding, groups=groups, bias=False))
    result.add_module('bn', nn.BatchNorm2d(num_features=out_channels))
    return result


class SiLU(nn.Module):
    @staticmethod
    def forward(x):
        return x * torch.sigmoid(x)

def get_activation(name="silu", inplace=True):
    if name == "silu":
        module = SiLU()
    elif name == "relu":
        module = nn.ReLU(inplace=inplace)
    elif name == "lrelu":
        module = nn.LeakyReLU(0.1, inplace=inplace)
    else:
        raise AttributeError("Unsupported act type: {}".format(name))
    return module


class RepVGGBlock(nn.Module):

    def __init__(self, in_channels, out_channels, kernel_size,
                 stride=1, padding=1, dilation=1, groups=1, padding_mode='zeros', deploy=False, use_se=False, act="relu"):
        super(RepVGGBlock, self).__init__()
        self.deploy = deploy
        self.groups = groups
        self.in_channels = in_channels
        self.act = get_activation(act, inplace=True)
        self.se = use_se

        assert kernel_size == 3
        assert padding == 1

        padding_11 = padding - kernel_size // 2


        # if use_se:
        #     self.se = SEBlock(out_channels, internal_neurons=out_channels // 16)
        # else:
        #     self.se = nn.Identity()

        if deploy:
            self.rbr_reparam = nn.Conv2d(in_channels=in_channels, out_channels=out_channels, kernel_size=kernel_size, stride=stride,
                                      padding=padding, dilation=dilation, groups=groups, bias=True, padding_mode=padding_mode)

        else:
            self.rbr_identity = nn.BatchNorm2d(num_features=in_channels) if out_channels == in_channels and stride == 1 else None
            self.rbr_dense = conv_bn(in_channels=in_channels, out_channels=out_channels, kernel_size=kernel_size, stride=stride, padding=padding, groups=groups)
            self.rbr_1x1 = conv_bn(in_channels=in_channels, out_channels=out_channels, kernel_size=1, stride=stride, padding=padding_11, groups=groups)
            print('RepVGG Block, identity = ', self.rbr_identity)


    def forward(self, inputs):
        if hasattr(self, 'rbr_reparam'):
            return self.act(self.rbr_reparam(inputs))

        if self.rbr_identity is None:
            id_out = 0
        else:
            id_out = self.rbr_identity(inputs)

        # return self.act(self.se(self.rbr_dense(inputs) + self.rbr_1x1(inputs) + id_out))
        return self.act(self.rbr_dense(inputs) + self.rbr_1x1(inputs) + id_out)


    #   Optional. This improves the accuracy and facilitates quantization.
    #   1.  Cancel the original weight decay on rbr_dense.conv.weight and rbr_1x1.conv.weight.
    #   2.  Use like this.
    #       loss = criterion(....)
    #       for every RepVGGBlock blk:
    #           loss += weight_decay_coefficient * 0.5 * blk.get_cust_L2()
    #       optimizer.zero_grad()
    #       loss.backward()
    def get_custom_L2(self):
        K3 = self.rbr_dense.conv.weight
        K1 = self.rbr_1x1.conv.weight
        t3 = (self.rbr_dense.bn.weight / ((self.rbr_dense.bn.running_var + self.rbr_dense.bn.eps).sqrt())).reshape(-1, 1, 1, 1).detach()
        t1 = (self.rbr_1x1.bn.weight / ((self.rbr_1x1.bn.running_var + self.rbr_1x1.bn.eps).sqrt())).reshape(-1, 1, 1, 1).detach()

        l2_loss_circle = (K3 ** 2).sum() - (K3[:, :, 1:2, 1:2] ** 2).sum()      # The L2 loss of the "circle" of weights in 3x3 kernel. Use regular L2 on them.
        eq_kernel = K3[:, :, 1:2, 1:2] * t3 + K1 * t1                           # The equivalent resultant central point of 3x3 kernel.
        l2_loss_eq_kernel = (eq_kernel ** 2 / (t3 ** 2 + t1 ** 2)).sum()        # Normalize for an L2 coefficient comparable to regular L2.
        return l2_loss_eq_kernel + l2_loss_circle



#   This func derives the equivalent kernel and bias in a DIFFERENTIABLE way.
#   You can get the equivalent kernel and bias at any time and do whatever you want,
    #   for example, apply some penalties or constraints during training, just like you do to the other models.
#   May be useful for quantization or pruning.
    def get_equivalent_kernel_bias(self):
        kernel3x3, bias3x3 = self._fuse_bn_tensor(self.rbr_dense)
        kernel1x1, bias1x1 = self._fuse_bn_tensor(self.rbr_1x1)
        kernelid, biasid = self._fuse_bn_tensor(self.rbr_identity)
        return kernel3x3 + self._pad_1x1_to_3x3_tensor(kernel1x1) + kernelid, bias3x3 + bias1x1 + biasid

    def _pad_1x1_to_3x3_tensor(self, kernel1x1):
        if kernel1x1 is None:
            return 0
        else:
            return torch.nn.functional.pad(kernel1x1, [1,1,1,1])

    def _fuse_bn_tensor(self, branch):
        if branch is None:
            return 0, 0
        if isinstance(branch, nn.Sequential):
            kernel = branch.conv.weight
            running_mean = branch.bn.running_mean
            running_var = branch.bn.running_var
            gamma = branch.bn.weight
            beta = branch.bn.bias
            eps = branch.bn.eps
        else:
            assert isinstance(branch, nn.BatchNorm2d)
            if not hasattr(self, 'id_tensor'):
                input_dim = self.in_channels // self.groups
                kernel_value = np.zeros((self.in_channels, input_dim, 3, 3), dtype=np.float32)
                for i in range(self.in_channels):
                    kernel_value[i, i % input_dim, 1, 1] = 1
                self.id_tensor = torch.from_numpy(kernel_value).to(branch.weight.device)
            kernel = self.id_tensor
            running_mean = branch.running_mean
            running_var = branch.running_var
            gamma = branch.weight
            beta = branch.bias
            eps = branch.eps
        std = (running_var + eps).sqrt()
        t = (gamma / std).reshape(-1, 1, 1, 1)
        return kernel * t, beta - running_mean * gamma / std

    def switch_to_deploy(self):
        if hasattr(self, 'rbr_reparam'):
            return
        kernel, bias = self.get_equivalent_kernel_bias()
        self.rbr_reparam = nn.Conv2d(in_channels=self.rbr_dense.conv.in_channels, out_channels=self.rbr_dense.conv.out_channels,
                                     kernel_size=self.rbr_dense.conv.kernel_size, stride=self.rbr_dense.conv.stride,
                                     padding=self.rbr_dense.conv.padding, dilation=self.rbr_dense.conv.dilation, groups=self.rbr_dense.conv.groups, bias=True)
        self.rbr_reparam.weight.data = kernel
        self.rbr_reparam.bias.data = bias
        for para in self.parameters():
            para.detach_()
        self.__delattr__('rbr_dense')
        self.__delattr__('rbr_1x1')
        if hasattr(self, 'rbr_identity'):
            self.__delattr__('rbr_identity')
        if hasattr(self, 'id_tensor'):
            self.__delattr__('id_tensor')
        self.deploy = True

class eca_layer(nn.Module):
    """Constructs a ECA module.

    Args:
        channel: Number of channels of the input feature map
        k_size: Adaptive selection of kernel size
    """
    def __init__(self, channel, k_size=3):
        super(eca_layer, self).__init__() # super类的作用是继承的时候，调用含super的哥哥的基类__init__函数。
        self.avg_pool = nn.AdaptiveAvgPool2d(1) # 全局平均池化
        self.conv = nn.Conv1d(1, 1, kernel_size=k_size,
                              padding=(k_size - 1) // 2, bias=False) # 一维卷积
        self.sigmoid = nn.Sigmoid()

    def forward(self, x):
        # x: input features with shape [b, c, h, w]
        b, c, h, w = x.size() # b代表b个样本，c为通道数，h为高度，w为宽度

        # feature descriptor on the global spatial information
        y = self.avg_pool(x)

        # Two different branches of ECA module
        # torch.squeeze()这个函数主要对数据的维度进行压缩,torch.unsqueeze()这个函数 主要是对数据维度进行扩充
        y = self.conv(y.squeeze(-1).transpose(-1, -2)).transpose(-1, -2).unsqueeze(-1)

        # Multi-scale information fusion多尺度信息融合
        y = self.sigmoid(y)
        # 原网络中克罗内克积，也叫张量积，为两个任意大小矩阵间的运算
        return x * y.expand_as(x)

# class ChannelAttentionModule(nn.Module):
#     def __init__(self, channel, ratio=16):
#         super(ChannelAttentionModule, self).__init__()
#         self.avg_pool = nn.AdaptiveAvgPool2d(1)
#         self.max_pool = nn.AdaptiveMaxPool2d(1)
#
#         self.shared_MLP = nn.Sequential(
#             nn.Conv2d(channel, channel // ratio, 1, bias=False),
#             nn.ReLU(),
#             nn.Conv2d(channel // ratio, channel, 1, bias=False)
#         )
#         self.sigmoid = nn.Sigmoid()
#
#     def forward(self, x):
#         avgout = self.shared_MLP(self.avg_pool(x))
#         # print(avgout.shape)
#         maxout = self.shared_MLP(self.max_pool(x))
#         return self.sigmoid(avgout + maxout)
#
# class SpatialAttentionModule(nn.Module):
#     def __init__(self):
#         super(SpatialAttentionModule, self).__init__()
#         self.conv2d = nn.Conv2d(in_channels=2, out_channels=1, kernel_size=7, stride=1, padding=3)
#         self.sigmoid = nn.Sigmoid()
#
#     def forward(self, x):
#         avgout = torch.mean(x, dim=1, keepdim=True)
#         maxout, _ = torch.max(x, dim=1, keepdim=True)
#         out = torch.cat([avgout, maxout], dim=1)
#         out = self.sigmoid(self.conv2d(out))
#         return out
#
# class CBAM(nn.Module):
#     def __init__(self, channel):
#         super(CBAM, self).__init__()
#         self.channel_attention = ChannelAttentionModule(channel)
#         self.spatial_attention = SpatialAttentionModule()
#
#     def forward(self, x):
#         out = self.channel_attention(x) * x
#         # print('outchannels:{}'.format(out.shape))
#         out = self.spatial_attention(out) * out
#         return out


class RepVGG(nn.Module):

    def __init__(self, num_blocks, num_classes=3, width_multiplier=None, override_groups_map=None, deploy=False, use_se=False):
        super(RepVGG, self).__init__()

        assert len(width_multiplier) == 4

        self.deploy = deploy
        self.override_groups_map = override_groups_map or dict()
        self.use_se = use_se
        self.eca = eca_layer(384)

        assert 0 not in self.override_groups_map

        self.in_planes = min(64, int(64 * width_multiplier[0]))

        self.stage0 = RepVGGBlock(in_channels=3, out_channels=self.in_planes, kernel_size=3, stride=2, padding=1, deploy=self.deploy, use_se=self.use_se)
        self.cur_layer_idx = 1
        self.stage1 = self._make_stage(int(64 * width_multiplier[0]), num_blocks[0], stride=2)
        self.stage2 = self._make_stage(int(128 * width_multiplier[1]), num_blocks[1], stride=2)
        self.stage3 = self._make_stage(int(256 * width_multiplier[2]), num_blocks[2], stride=2)
        self.stage4 = self._make_stage(int(512 * width_multiplier[3]), num_blocks[3], stride=2)

    def _make_stage(self, planes, num_blocks, stride):
        strides = [stride] + [1]*(num_blocks-1)
        blocks = []
        for stride in strides:
            cur_groups = self.override_groups_map.get(self.cur_layer_idx, 1)
            blocks.append(RepVGGBlock(in_channels=self.in_planes, out_channels=planes, kernel_size=3,
                                      stride=stride, padding=1, groups=cur_groups, deploy=self.deploy, use_se=self.use_se))
            self.in_planes = planes
            self.cur_layer_idx += 1
        return nn.Sequential(*blocks)

    def forward(self, x):
        outputs = {}
        x = self.stage0(x)
        # print(x.shape)
        x = self.stage1(x)
        # print(x.shape)
        x = self.stage2(x)
        # print(x.shape)
        outputs["dark3"] = x
        x = self.stage3(x)
        # print(x.shape)
        outputs["dark4"] = x
        x = self.stage4(x)
        x = self.eca(x)
        # print(x.shape)
        outputs["dark5"] = x
        return {k: v for k, v in outputs.items()}


optional_groupwise_layers = [2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26]
g2_map = {l: 2 for l in optional_groupwise_layers}
g4_map = {l: 4 for l in optional_groupwise_layers}

def create_RepVGG_A0(deploy=False):
    return RepVGG(num_blocks=[2, 4, 14, 1], num_classes=3,
                  width_multiplier=[1, 0.75, 0.75, 0.75], override_groups_map=None, deploy=deploy)

def create_RepVGG_A1(deploy=False):
    return RepVGG(num_blocks=[2, 4, 14, 1], num_classes=1000,
                  width_multiplier=[1, 1, 1, 2.5], override_groups_map=None, deploy=deploy)

def create_RepVGG_A2(deploy=False):
    return RepVGG(num_blocks=[2, 4, 14, 1], num_classes=1000,
                  width_multiplier=[1.5, 1.5, 1.5, 2.75], override_groups_map=None, deploy=deploy)

def create_RepVGG_B0(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[1, 1, 1, 2.5], override_groups_map=None, deploy=deploy)

def create_RepVGG_B1(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[2, 2, 2, 4], override_groups_map=None, deploy=deploy)

def create_RepVGG_B1g2(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[2, 2, 2, 4], override_groups_map=g2_map, deploy=deploy)

def create_RepVGG_B1g4(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[2, 2, 2, 4], override_groups_map=g4_map, deploy=deploy)


def create_RepVGG_B2(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[2.5, 2.5, 2.5, 5], override_groups_map=None, deploy=deploy)

def create_RepVGG_B2g2(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[2.5, 2.5, 2.5, 5], override_groups_map=g2_map, deploy=deploy)

def create_RepVGG_B2g4(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[2.5, 2.5, 2.5, 5], override_groups_map=g4_map, deploy=deploy)


def create_RepVGG_B3(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[3, 3, 3, 5], override_groups_map=None, deploy=deploy)

def create_RepVGG_B3g2(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[3, 3, 3, 5], override_groups_map=g2_map, deploy=deploy)

def create_RepVGG_B3g4(deploy=False):
    return RepVGG(num_blocks=[4, 6, 16, 1], num_classes=1000,
                  width_multiplier=[3, 3, 3, 5], override_groups_map=g4_map, deploy=deploy)

def create_RepVGG_D2se(deploy=False):
    return RepVGG(num_blocks=[8, 14, 24, 1], num_classes=1000,
                  width_multiplier=[2.5, 2.5, 2.5, 5], override_groups_map=None, deploy=deploy, use_se=True)


func_dict = {
'RepVGG-A0': create_RepVGG_A0,
'RepVGG-A1': create_RepVGG_A1,
'RepVGG-A2': create_RepVGG_A2,
'RepVGG-B0': create_RepVGG_B0,
'RepVGG-B1': create_RepVGG_B1,
'RepVGG-B1g2': create_RepVGG_B1g2,
'RepVGG-B1g4': create_RepVGG_B1g4,
'RepVGG-B2': create_RepVGG_B2,
'RepVGG-B2g2': create_RepVGG_B2g2,
'RepVGG-B2g4': create_RepVGG_B2g4,
'RepVGG-B3': create_RepVGG_B3,
'RepVGG-B3g2': create_RepVGG_B3g2,
'RepVGG-B3g4': create_RepVGG_B3g4,
'RepVGG-D2se': create_RepVGG_D2se,      #   Updated at April 25, 2021. This is not reported in the CVPR paper.
}
def get_RepVGG_func_by_name(name):
    return func_dict[name]



#   Use this for converting a RepVGG model or a bigger model with RepVGG as its component
#   Use like this
#   model = create_RepVGG_A0(deploy=False)
#   train model or load weights
#   repvgg_model_convert(model, save_path='repvgg_deploy.pth')
#   If you want to preserve the original model, call with do_copy=True

#   ====================== for using RepVGG as the backbone of a bigger model, e.g., PSPNet, the pseudo code will be like
#   train_backbone = create_RepVGG_B2(deploy=False)
#   train_backbone.load_state_dict(torch.load('RepVGG-B2-train.pth'))
#   train_pspnet = build_pspnet(backbone=train_backbone)
#   segmentation_train(train_pspnet)
#   deploy_pspnet = repvgg_model_convert(train_pspnet)
#   segmentation_test(deploy_pspnet)
#   =====================   example_pspnet.py shows an example

def repvgg_model_convert(model:torch.nn.Module, save_path=None, do_copy=True):
    if do_copy:
        model = copy.deepcopy(model)
    for module in model.modules():
        if hasattr(module, 'switch_to_deploy'):
            module.switch_to_deploy()
    if save_path is not None:
        torch.save(model.state_dict(), save_path)
    return model