YOLO-Tutorial-v2/yolo/models/yolov6/modules.py

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


# --------------------- Basic modules ---------------------
def get_activation(act_type=None):
    if act_type == 'relu':
        return nn.ReLU(inplace=True)
    elif act_type == 'lrelu':
        return nn.LeakyReLU(0.1, inplace=True)
    elif act_type == 'mish':
        return nn.Mish(inplace=True)
    elif act_type == 'silu':
        return nn.SiLU(inplace=True)
    elif act_type is None:
        return nn.Identity()
    else:
        raise NotImplementedError

class ConvModule(nn.Module):
    def __init__(self, 
                 in_dim,                   # in channels
                 out_dim,                  # out channels 
                 kernel_size=1,            # kernel size
                 padding=0,                # padding
                 stride=1,                 # padding
                 act_type  :str = 'lrelu', # activation
                ):
        super(ConvModule, self).__init__()
        self.conv = nn.Conv2d(in_dim, out_dim, kernel_size=kernel_size, padding=padding, stride=stride, bias=False)
        self.norm = nn.BatchNorm2d(out_dim)
        self.act  = get_activation(act_type)

    def forward(self, x):
        return self.act(self.norm(self.conv(x)))

class RepVGGBlock(nn.Module):
    def __init__(self,
                 in_channels,
                 out_channels,
                 kernel_size=3,
                 stride=1,
                 padding=1,
                 dilation=1,
                 groups=1,
                 deploy=False,
                 ):
        super(RepVGGBlock, self).__init__()
        assert kernel_size == 3
        assert padding == 1
        # --------- Basic parameters ---------
        self.deploy = deploy
        self.groups = groups
        self.in_channels = in_channels
        self.out_channels = out_channels
        padding_11 = padding - kernel_size // 2
        # --------- Model parameters ---------
        if deploy:
            self.rbr_reparam = nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size, stride=stride,
                                         padding=padding, dilation=dilation, groups=groups, bias=True)
        else:
            self.rbr_identity = nn.BatchNorm2d(num_features=in_channels) if out_channels == in_channels and stride == 1 else None
            self.rbr_dense    = ConvModule(in_channels, out_channels, kernel_size=kernel_size, padding=padding, stride=stride, act_type=None)
            self.rbr_1x1      = ConvModule(in_channels, out_channels, kernel_size=1, padding=padding_11, stride=stride, act_type=None)
        self.nonlinearity = nn.ReLU()

    def forward(self, inputs):
        '''Forward process'''
        if hasattr(self, 'rbr_reparam'):
            return self.nonlinearity(self.rbr_reparam(inputs))

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

        return self.nonlinearity(self.rbr_dense(inputs) + self.rbr_1x1(inputs) + id_out)

    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 _avg_to_3x3_tensor(self, avgp):
        channels = self.in_channels
        groups = self.groups
        kernel_size = avgp.kernel_size
        input_dim = channels // groups
        k = torch.zeros((channels, input_dim, kernel_size, kernel_size))
        k[np.arange(channels), np.tile(np.arange(input_dim), groups), :, :] = 1.0 / kernel_size ** 2
        return k

    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, ConvModule):
            kernel = branch.conv.weight
            bias   = branch.conv.bias
            return kernel, bias
        elif 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


# ---------------------------- Basic Modules ----------------------------
class RepBlock(nn.Module):
    def __init__(self, in_channels, out_channels, num_blocks=1, block=RepVGGBlock):
        super().__init__()
        self.conv1 = block(in_channels, out_channels)
        self.block = nn.Sequential(*(block(out_channels, out_channels)
                                     for _ in range(num_blocks - 1))) if num_blocks > 1 else nn.Identity()
        if block == BottleRep:
            self.conv1 = BottleRep(in_channels, out_channels, weight=True)
            num_blocks = num_blocks // 2
            self.block = nn.Sequential(*(BottleRep(out_channels, out_channels, weight=True)
                                         for _ in range(num_blocks - 1))) if num_blocks > 1 else None

    def forward(self, x):
        x = self.conv1(x)
        if self.block is not None:
            x = self.block(x)

        return x

class BottleRep(nn.Module):

    def __init__(self, in_channels, out_channels, weight=False):
        super().__init__()
        self.conv1 = RepVGGBlock(in_channels, out_channels, kernel_size=3, padding=1, stride=1)
        self.conv2 = RepVGGBlock(out_channels, out_channels, kernel_size=3, padding=1, stride=1)
        if in_channels != out_channels:
            self.shortcut = False
        else:
            self.shortcut = True
        if weight:
            self.alpha = nn.Parameter(torch.ones(1))
        else:
            self.alpha = 1.0

    def forward(self, x):
        outputs = self.conv1(x)
        outputs = self.conv2(outputs)

        return outputs + self.alpha * x if self.shortcut else outputs

class RepCSPBlock(nn.Module):
    def __init__(self, in_channels, out_channels, num_blocks=1, expansion=0.5):
        super().__init__()
        inter_dim = round(out_channels * expansion)  # hidden channels
        self.cv1 = ConvModule(in_channels, inter_dim, kernel_size=1, act_type='relu')
        self.cv2 = ConvModule(in_channels, inter_dim, kernel_size=1, act_type='relu')
        self.cv3 = ConvModule(2 * inter_dim, out_channels, kernel_size=1, act_type='relu')

        self.module = RepBlock(inter_dim, inter_dim, num_blocks, block=BottleRep)

    def forward(self, x):
        x1 = self.cv1(x)
        x2 = self.module(self.cv2(x))
        out = self.cv3(torch.cat((x1, x2), dim=1))

        return out