主干目录：
【YOLOV5-6.x 版本讲解】整体项目代码注释导航现在YOLOV5已经更新到6.X版本，现在网上很多还停留在5.X的源码注释上，因此特开一贴传承开源精神！5.X版本的可以看其他大佬的帖子本文章主要从6.X版本出发，主要解决6.X版本的项目注释与代码分析！......https://blog.csdn.net/qq_39237205/article/details/125729662
以下内容为本栏目的一部分，更多关注以上链接目录，查找YOLOV5的更多信息
祝福你朋友早日发表sci！
# YOLOv5 🚀 by Ultralytics, GPL-3.0 license# https://blog.csdn.net/qq_39237205/category_11911202.html"""
YOLO-specific modules这个模块是yolov5的模型搭建模块，非常的重要，不过代码量并不大，不是很难，只是yolov5的作者把封装的太好了，模型扩展了很多的额外的功能，导致看起来很难，其实真正有用的代码不多的。重点是抓住三个函数是在哪里调用的，谁调用谁的。Usage:$ python path/to/models/yolo.py --cfg yolov5s.yaml
"""import argparse # 解析命令行参数模块
import sys # sys系统模块 包含了与Python解释器和它的环境有关的函数
from copy import deepcopy # 数据拷贝模块 深拷贝
from pathlib import Path # Path将str转换为Path对象 使字符串路径易于操作的模块FILE = Path(__file__).resolve()
ROOT = FILE.parents[1] # YOLOv5 root directory
if str(ROOT) not in sys.path:sys.path.append(str(ROOT)) # add ROOT to PATH
# ROOT = ROOT.relative_to(Path.cwd()) # relativefrom models.common import *
from models.experimental import *
from utils.autoanchor import check_anchor_order
from utils.general import LOGGER, check_version, check_yaml, make_divisible, print_args
from utils.plots import feature_visualization
from utils.torch_utils import fuse_conv_and_bn, initialize_weights, model_info, scale_img, select_device, time_sync# 导入thop包 用于计算FLOPs
try:import thop # for FLOPs computation
except ImportError:thop = Noneclass Detect(nn.Module):"""Detect模块是用来构建Detect层的，将输入feature map 通过一个卷积操作和公式计算到我们想要的shape, 为后面的计算损失或者NMS作准备"""stride = None # strides computed during buildonnx_dynamic = False # ONNX export parameter 再export中这个参数会重新设为Truedef __init__(self, nc=80, anchors=(), ch=(), inplace=True): # detection layersuper().__init__()"""detection layer 相当于yolov3中的YOLOLayer层:params nc: number of classes:params anchors: 传入3个feature map上的所有anchor的大小（P3、P4、P5）:params ch: [128, 256, 512] 3个输出feature map的channel"""self.nc = nc # number of classes，若是VOC，则类别为20self.no = nc + 5 # number of outputs per anchor。 若是VOC: 5+20=25 该数为：xywhc+classesself.nl = len(anchors) # number of detection layers Detect的个数 3self.na = len(anchors[0]) // 2 # number of anchors 每个feature map的anchor个数 3self.grid = [torch.zeros(1)] * self.nl # init grid {list: 3} tensor([0.]) X 3self.anchor_grid = [torch.zeros(1)] * self.nl # init anchor grid# a=[3, 3, 2] anchors以[w, h]对的形式存储 3个feature map 每个feature map上有三个anchor（w,h）# a = torch.tensor(anchors).float().view(self.nl, -1, 2)# register_buffer# 模型中需要保存的参数一般有两种：# 一种是反向传播需要被optimizer更新的，即参与训练的参数称为parameter，optim.step只能更新nn.parameter类型的参数# 另一种不要被更新，即不参与训练的参数称为buffer，buffer的参数更新是在forward中。# shape(nl,na,2)# self.register_buffer('anchors', a)self.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2)) # shape(nl,na,2)# output conv 对每个输出的feature map都要调用一次conv1x1self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv# use in-place ops (e.g. slice assignment) 一般都是True 默认不使用AWS Inferentia加速self.inplace = inplace # use in-place ops (e.g. slice assignment)def forward(self, x): # x：[[],[],[]]分别对应1/8 1/16 1/32 三个维度大小的宽高输入# forward函数在Model类的forward_once中调用""":returntrain: 一个tensor list 存放三个元素 [bs, anchor_num, grid_w, grid_h, xywh+c+20classes]分别是 [1, 3, 80, 80, 25] [1, 3, 40, 40, 25] [1, 3, 20, 20, 25]inference: 0 [1, 19200+4800+1200, 25] = [bs, anchor_num*grid_w*grid_h, xywh+c+20classes]1 一个tensor list 存放三个元素 [bs, anchor_num, grid_w, grid_h, xywh+c+20classes][1, 3, 80, 80, 25] [1, 3, 40, 40, 25] [1, 3, 20, 20, 25]"""z = [] # inference outputfor i in range(self.nl): # 对三个feature map分别进行处理，遍历一共多少层x[i] = self.m[i](x[i]) # conv xi[bs, 128/256/512, 80, 80] to [bs, 75, 80, 80]bs, _, ny, nx = x[i].shape # x(bs,255,20,20) to x(bs,3,20,20,85)x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()# inference，预测部分if not self.training: # inference# 构造网格# 因为推理返回的不是归一化后的网格偏移量 需要再加上网格的位置 得到最终的推理坐标 再送入nms# 所以这里构建网格就是为了记录每个grid的网格坐标 方面后面使用if self.onnx_dynamic or self.grid[i].shape[2:4] != x[i].shape[2:4]: # 第一次运行时候，会实例化这两个属性self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i) # 拿到左上角的坐标y = x[i].sigmoid() # 将每一层的特征归一化到0到1之间if self.inplace:# 默认执行 不使用AWS Inferentia# 这里的公式和yolov3、v4中使用的不一样 是yolov5作者自己用的效果更好，边框预测公式，ppt有# 计算中心点坐标，将0到1之间处理到原图大小的区间y[..., 0:2] = (y[..., 0:2] * 2 - 0.5 + self.grid[i]) * self.stride[i] # xy # xy||||| × self.stride[i]是为了放大到原图# 计算宽高，将0到1之间处理到原图大小的区间y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # whelse: # for YOLOv5 on AWS Inferentia https://github.com/ultralytics/yolov5/pull/2953xy = (y[..., 0:2] * 2 - 0.5 + self.grid[i]) * self.stride[i] # xywh = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # why = torch.cat((xy, wh, y[..., 4:]), -1)# z是一个tensor list 三个元素 分别是[1, 19200, 25] [1, 4800, 25] [1, 1200, 25]z.append(y.view(bs, -1, self.no))return x if self.training else (torch.cat(z, 1), x)def _make_grid(self, nx=20, ny=20, i=0):"""构造网格"""d = self.anchors[i].deviceif check_version(torch.__version__, '1.10.0'): # torch>=1.10.0 meshgrid workaround for torch>=0.7 compatibilityyv, xv = torch.meshgrid([torch.arange(ny, device=d), torch.arange(nx, device=d)], indexing='ij')else:yv, xv = torch.meshgrid([torch.arange(ny, device=d), torch.arange(nx, device=d)])grid = torch.stack((xv, yv), 2).expand((1, self.na, ny, nx, 2)).float()anchor_grid = (self.anchors[i].clone() * self.stride[i]) \.view((1, self.na, 1, 1, 2)).expand((1, self.na, ny, nx, 2)).float()return grid, anchor_gridclass Model(nn.Module):def __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None): # model, input channels, number of classes"""Model主要包含模型的搭建与扩展功能，yolov5的作者将这个模块的功能写的很全，扩展功能如：特征可视化，打印模型信息、TTA推理增强、融合Conv+Bn加速推理、模型搭载nms功能、autoshape函数：模型搭建包含前处理、推理、后处理的模块(预处理 + 推理 + nms)。感兴趣的可以仔细看看，不感兴趣的可以直接看__init__和__forward__两个函数即可。:params cfg:模型配置文件:params ch: input img channels 一般是3 RGB文件:params nc: number of classes 数据集的类别个数:anchors: 一般是None"""super().__init__()if isinstance(cfg, dict):self.yaml = cfg # model dictelse: # is *.yaml# is *.yaml 一般执行这里import yaml # for torch hubself.yaml_file = Path(cfg).name # cfg file name = yolov5s.yaml# 如果配置文件中有中文，打开时要加encoding参数with open(cfg, encoding='ascii', errors='ignore') as f:# model dict 取到配置文件中每条的信息（没有注释内容）self.yaml = yaml.safe_load(f) # model dict# Define modelch = self.yaml['ch'] = self.yaml.get('ch', ch) # input channels# 设置类别数 一般不执行, 因为nc=self.yaml['nc']恒成立if nc and nc != self.yaml['nc']:LOGGER.info(f"Overriding model.yaml nc={self.yaml['nc']} with nc={nc}")self.yaml['nc'] = nc # override yaml value# 重写anchor，一般不执行, 因为传进来的anchors一般都是Noneif anchors:LOGGER.info(f'Overriding model.yaml anchors with anchors={anchors}')self.yaml['anchors'] = round(anchors) # override yaml value# 创建网络模型# self.model: 初始化的整个网络模型(包括Detect层结构)# self.save: 所有层结构中from不等于-1的序号，并排好序 [4, 6, 10, 14, 17, 20, 23]self.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch]) # model, savelist# default class names ['0', '1', '2',..., '19']self.names = [str(i) for i in range(self.yaml['nc'])] # default names# self.inplace=True 默认True 不使用加速推理# AWS Inferentia Inplace compatiability# https://github.com/ultralytics/yolov5/pull/2953self.inplace = self.yaml.get('inplace', True)# 获取Detect模块的stride(相对输入图像的下采样率)和anchors在当前Detect输出的feature map的尺度# Build strides, anchorsm = self.model[-1] # Detect()if isinstance(m, Detect):s = 256 # 2x min stridem.inplace = self.inplace# 计算三个feature map下采样的倍率 [8, 16, 32]# 假设640X640的图片大小，在最后三层时分别乘1/8 1/16 1/32，得到80，40，20m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))]) # 前向传播的处理，为了得到最后输出的stride的大小 # forward# 将当前图片的大小处理成相对当前feature map的anchor大小 如[10, 13]/8 -> [1.25, 1.625]m.anchors /= m.stride.view(-1, 1, 1)# 检查anchor顺序与stride顺序是否一致check_anchor_order(m)self.stride = m.strideself._initialize_biases() # only run once # only run once 初始化偏置# logger.info('Strides: %s' % m.stride.tolist())# Init weights, biasesinitialize_weights(self) # 调用torch_utils.py下initialize_weights初始化模型权重self.info() # 打印模型信息LOGGER.info('')def forward(self, x, augment=False, profile=False, visualize=False):# augmented inference, None 上下flip/左右flip# 是否在测试时也使用数据增强 Test Time Augmentation(TTA)if augment:return self._forward_augment(x) # augmented inference, None# 默认执行 正常前向推理# single-scale inference, trainreturn self._forward_once(x, profile, visualize) # single-scale inference, traindef _forward_augment(self, x):img_size = x.shape[-2:] # height, widths = [1, 0.83, 0.67] # scalesf = [None, 3, None] # flips (2-ud, 3-lr)y = [] # outputsfor si, fi in zip(s, f):# scale_img缩放图片尺寸xi = scale_img(x.flip(fi) if fi else x, si, gs=int(self.stride.max()))yi = self._forward_once(xi)[0] # forward# cv2.imwrite(f'img_{si}.jpg', 255 * xi[0].cpu().numpy().transpose((1, 2, 0))[:, :, ::-1]) # save# _descale_pred将推理结果恢复到相对原图图片尺寸yi = self._descale_pred(yi, fi, si, img_size)y.append(yi)y = self._clip_augmented(y) # clip augmented tailsreturn torch.cat(y, 1), None # augmented inference, traindef _forward_once(self, x, profile=False, visualize=False):""":params x: 输入图像:params profile: True 可以做一些性能评估:params feature_vis: True 可以做一些特征可视化:return train: 一个tensor list 存放三个元素 [bs, anchor_num, grid_w, grid_h, xywh+c+20classes]分别是 [1, 3, 80, 80, 25] [1, 3, 40, 40, 25] [1, 3, 20, 20, 25]inference: 0 [1, 19200+4800+1200, 25] = [bs, anchor_num*grid_w*grid_h, xywh+c+20classes]1 一个tensor list 存放三个元素 [bs, anchor_num, grid_w, grid_h, xywh+c+20classes][1, 3, 80, 80, 25] [1, 3, 40, 40, 25] [1, 3, 20, 20, 25]"""# y: 存放着self.save=True的每一层的输出，因为后面的层结构concat等操作要用到# dt: 在profile中做性能评估时使用y, dt = [], [] # outputsfor m in self.model:# 前向推理每一层结构 m.i=index m.f=from m.type=类名 m.np=number of params# if not from previous layer m.f=当前层的输入来自哪一层的输出 s的m.f都是-1if m.f != -1: # if not from previous layer# 这里需要做4个concat操作和1个Detect操作# concat操作如m.f=[-1,6] x就有两个元素,一个是上一层的输出,另一个是index=6的层的输出 再送到x=m(x)做concat操作# Detect操作m.f=[17, 20, 23] x有三个元素,分别存放第17层第20层第23层的输出 再送到x=m(x)做Detect的forwardx = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers# 打印日志信息 FLOPs time等# 打印日志信息 前向推理时间if profile:self._profile_one_layer(m, x, dt)x = m(x) # run正向推理 执行每一层的forward函数(除Concat和Detect操作)# print('层数',i,'特征图大小',x.shape)# 存放着self.save的每一层的输出，因为后面需要用来作concat等操作要用到 不在self.save层的输出就为Noney.append(x if m.i in self.save else None) # save output# 特征可视化 可以自己改动想要哪层的特征进行可视化if visualize:feature_visualization(x, m.type, m.i, save_dir=visualize)return xdef _descale_pred(self, p, flips, scale, img_size):"""用在上面的__init__函数上将推理结果恢复到原图图片尺寸 Test Time Augmentation(TTA)中用到de-scale predictions following augmented inference (inverse operation):params p: 推理结果:params flips::params scale::params img_size:"""# 不同的方式前向推理使用公式不同 具体可看Detect函数# de-scale predictions following augmented inference (inverse operation)if self.inplace: # 默认执行 不使用AWS Inferentiap[..., :4] /= scale # de-scaleif flips == 2:p[..., 1] = img_size[0] - p[..., 1] # de-flip udelif flips == 3:p[..., 0] = img_size[1] - p[..., 0] # de-flip lrelse:x, y, wh = p[..., 0:1] / scale, p[..., 1:2] / scale, p[..., 2:4] / scale # de-scaleif flips == 2:y = img_size[0] - y # de-flip udelif flips == 3:x = img_size[1] - x # de-flip lrp = torch.cat((x, y, wh, p[..., 4:]), -1)return pdef _clip_augmented(self, y):# Clip YOLOv5 augmented inference tailsnl = self.model[-1].nl # number of detection layers (P3-P5)g = sum(4 ** x for x in range(nl)) # grid pointse = 1 # exclude layer counti = (y[0].shape[1] // g) * sum(4 ** x for x in range(e)) # indicesy[0] = y[0][:, :-i] # largei = (y[-1].shape[1] // g) * sum(4 ** (nl - 1 - x) for x in range(e)) # indicesy[-1] = y[-1][:, i:] # smallreturn ydef _profile_one_layer(self, m, x, dt):c = isinstance(m, Detect) # is final layer, copy input as inplace fixo = thop.profile(m, inputs=(x.copy() if c else x,), verbose=False)[0] / 1E9 * 2 if thop else 0 # FLOPst = time_sync()for _ in range(10):m(x.copy() if c else x)dt.append((time_sync() - t) * 100)if m == self.model[0]:LOGGER.info(f"{'time (ms)':>10s} {'GFLOPs':>10s} {'params':>10s} {'module'}")LOGGER.info(f'{dt[-1]:10.2f} {o:10.2f} {m.np:10.0f} {m.type}')if c:LOGGER.info(f"{sum(dt):10.2f} {'-':>10s} {'-':>10s} Total")def _initialize_biases(self, cf=None): # initialize biases into Detect(), cf is class frequency"""用在上面的__init__函数上initialize biases into Detect(), cf is class frequencyhttps://arxiv.org/abs/1708.02002 section 3.3"""# cf = torch.bincount(torch.tensor(np.concatenate(dataset.labels, 0)[:, 0]).long(), minlength=nc) + 1.m = self.model[-1] # Detect() modulefor mi, s in zip(m.m, m.stride): # fromb = mi.bias.view(m.na, -1) # conv.bias(255) to (3,85)b.data[:, 4] += math.log(8 / (640 / s) ** 2) # obj (8 objects per 640 image)b.data[:, 5:] += math.log(0.6 / (m.nc - 0.999999)) if cf is None else torch.log(cf / cf.sum()) # clsmi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True)def _print_biases(self):"""打印模型中最后Detect层的偏置bias信息(也可以任选哪些层bias信息)"""m = self.model[-1] # Detect() modulefor mi in m.m: # fromb = mi.bias.detach().view(m.na, -1).T # conv.bias(255) to (3,85)LOGGER.info(('%6g Conv2d.bias:' + '%10.3g' * 6) % (mi.weight.shape[1], *b[:5].mean(1).tolist(), b[5:].mean()))# def _print_weights(self):# for m in self.model.modules():# if type(m) is Bottleneck:# LOGGER.info('%10.3g' % (m.w.detach().sigmoid() * 2)) # shortcut weightsdef fuse(self): # fuse model Conv2d() + BatchNorm2d() layers"""用在detect.py、val.pyfuse model Conv2d() + BatchNorm2d() layers调用torch_utils.py中的fuse_conv_and_bn函数和common.py中Conv模块的fuseforward函数"""LOGGER.info('Fusing layers... ') # 日志# 遍历每一层结构for m in self.model.modules():# 如果当前层是卷积层Conv且有bn结构, 那么就调用fuse_conv_and_bn函数讲conv和bn进行融合, 加速推理if isinstance(m, (Conv, DWConv)) and hasattr(m, 'bn'):m.conv = fuse_conv_and_bn(m.conv, m.bn) # 融合 update convdelattr(m, 'bn') # 移除bn remove batchnormm.forward = m.forward_fuse # 更新前向传播 update forward (反向传播不用管, 因为这种推理只用在推理阶段)self.info() # 打印conv+bn融合后的模型信息return selfdef info(self, verbose=False, img_size=640): # print model information"""用在上面的__init__函数上调用torch_utils.py下model_info函数打印模型信息"""model_info(self, verbose, img_size)def _apply(self, fn):# Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffersself = super()._apply(fn)m = self.model[-1] # Detect()if isinstance(m, Detect):m.stride = fn(m.stride)m.grid = list(map(fn, m.grid))if isinstance(m.anchor_grid, list):m.anchor_grid = list(map(fn, m.anchor_grid))return selfdef parse_model(d, ch): # model_dict, input_channels(3)"""主要功能：parse_model模块用来解析模型文件(从Model中传来的字典形式)，并搭建网络结构。在上面Model模块的__init__函数中调用这个函数其实主要做的就是: 更新当前层的args（参数）,计算c2（当前层的输出channel） =>使用当前层的参数搭建当前层 =>生成 layers + save:params d: model_dict 模型文件 字典形式 {dict:7} yolov5s.yaml中的6个元素 + ch:params ch: 记录模型每一层的输出channel 初始ch=[3] 后面会删除:return nn.Sequential(*layers): 网络的每一层的层结构:return sorted(save): 把所有层结构中from不是-1的值记下 并排序 [4, 6, 10, 14, 17, 20, 23]"""LOGGER.info(f"\n{'':>3}{'from':>18}{'n':>3}{'params':>10} {'module':<40}{'arguments':<30}")# 读取d字典中的anchors和parameters(nc、depth_multiple、width_multiple)# nc（number of classes）数据集类别个数；# depth_multiple，通过深度参数depth gain在搭建每一层的时候，实际深度 = 理论深度(每一层的参数n) * depth_multiple，起到一个动态调整模型深度的作用。# width_multiple，在模型中间层的每一层的实际输出channel = 理论channel(每一层的参数c2) * width_multiple，起到一个动态调整模型宽度的作用。anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple']# na: number of anchors 每一个predict head上的anchor数 = 3na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors# no: number of outputs 每一个predict head层的输出channel = anchors * (classes + 5) = 75(VOC)no = na * (nc + 5) #总共预测的anchors个数 number of outputs = anchors * (classes + 5)# 开始搭建网络# layers: 保存每一层的层结构# save: 记录下所有层结构中from中不是-1的层结构序号# c2: 保存当前层的输出channellayers, save, c2 = [], [], ch[-1] # layers, savelist, ch out# from(当前层输入来自哪些层), number(当前层次数 初定), module(当前层类别), args(当前层类参数 初定)for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # 遍历backbone和head的每一层 # from, number, module, args# eval(string) 得到当前层的真实类名# 例如: m= Focus -> <class 'models.common.Focus'>m = eval(m) if isinstance(m, str) else m # 将字符串处理成一个类名 或者 字符串，即实现名字向类的转换for j, a in enumerate(args): # 主要照顾 yolo.yaml文件中最后一列的, [nc, anchors]try:args[j] = eval(a) if isinstance(a, str) else a # eval strings，当他是一个字符串，就试图将它处理成一个变量名except NameError:pass# ------------------- 更新当前层的args（参数）,计算c2（当前层的输出channel） -------------------# depth gain 控制深度 如v5s: n*0.33 n: 当前模块的次数(间接控制深度)n = n_ = max(round(n * gd), 1) if n > 1 else n # depth gainif m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, DWConv, MixConv2d, Focus, CrossConv,BottleneckCSP, C3, C3TR, C3SPP, C3Ghost]:# c1: 当前层的输入的channel数# c2: 当前层的输出的channel数(初定)# ch: 记录着所有层的输出channel，f代表该ch中文最后一个，即对一下一层来说，这就是-1层的输入c1, c2 = ch[f], args[0] # args[0]为[-1, 1, Conv, [128, 3, 2]]这的128# if not output no=75 只有最后一层c2=no 最后一层不用控制宽度，输出channel必须是noif c2 != no: # if not output# width gain 控制宽度 如v5s: c2*width_multiple（yolo.yaml）# c2: 当前层的最终输出的channel数(间接控制宽度)c2 = make_divisible(c2 * gw, 8)# 在初始arg的基础上更新 加入当前层的输入channel并更新当前层# [in_channel, out_channel, *args[1:]]args = [c1, c2, *args[1:]] # [-1, 1, Conv, [128, 3, 2]] 变为 [-1, 1, Conv, [-1的值，128 × width_multiple , 3, 2]]# 如果当前层是BottleneckCSP/C3/C3TR, 则需要在args中加入bottleneck的个数# [in_channel, out_channel, Bottleneck的个数n, bool(True表示有shortcut 默认，反之无)]if m in [BottleneckCSP, C3, C3TR, C3Ghost]: # 因为这几个类的定义中，初始化中有n=1这个参数，整个过程就是在初始化卷积的参数罢了args.insert(2, n) # 在第二个位置插入bottleneck个数nn = 1 # 恢复默认值1elif m is nn.BatchNorm2d:# BN层只需要返回上一层的输出channelargs = [ch[f]]elif m is Concat:# Concat层则将f中所有的输出累加得到这层的输出channelc2 = sum(ch[x] for x in f) # 因为这个[[-1, 6], 1, Concat, [1]] 的第一个是个列表，所以需要遍历，然后将-1, 6层的输入加起来elif m is Detect: # Detect（YOLO Layer）层# 在args中加入三个Detect层的输出channelargs.append([ch[x] for x in f])if isinstance(args[1], int): # number of anchors 几乎不执行args[1] = [list(range(args[1] * 2))] * len(f)elif m is Contract:c2 = ch[f] * args[0] ** 2elif m is Expand:c2 = ch[f] // args[0] ** 2else:# Upsamplec2 = ch[f] # args不变# m_: 得到当前层module 如果n>1就创建多个m(当前层结构), 如果n=1就创建一个m# n只有在[BottleneckCSP, C3, C3TR]中才会用到m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args) # module# 打印当前层结构的一些基本信息t = str(m)[8:-2].replace('__main__.', '') # module typenp = sum(x.numel() for x in m_.parameters()) # number paramsm_.i, m_.f, m_.type, m_.np = i, f, t, np # attach index, 'from' index, type, number paramsLOGGER.info(f'{i:>3}{str(f):>18}{n_:>3}{np:10.0f} {t:<40}{str(args):<30}') # print# append to savelist 把所有层结构中from不是-1的值记下 [6, 4, 14, 10, 17, 20, 23]save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) # append to savelist# 将当前层结构module加入layers中layers.append(m_)if i == 0:ch = [] # 去除输入channel [3]# 把当前层的输出channel数加入chch.append(c2)return nn.Sequential(*layers), sorted(save) # nn.Sequential(*layers) 处理成一个模型if __name__ == '__main__':parser = argparse.ArgumentParser()parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='model.yaml')parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')parser.add_argument('--profile', action='store_true', help='profile model speed')parser.add_argument('--test', action='store_true', help='test all yolo*.yaml')opt = parser.parse_args()opt.cfg = check_yaml(opt.cfg) # check YAMLprint_args(FILE.stem, opt)device = select_device(opt.device)# Create modelmodel = Model(opt.cfg).to(device)model.train()# Profileif opt.profile:img = torch.rand(8 if torch.cuda.is_available() else 1, 3, 640, 640).to(device)y = model(img, profile=True)# Test all modelsif opt.test:for cfg in Path(ROOT / 'models').rglob('yolo*.yaml'):try:_ = Model(cfg)except Exception as e:print(f'Error in {cfg}: {e}')# Tensorboard (not working https://github.com/ultralytics/yolov5/issues/2898)# from torch.utils.tensorboard import SummaryWriter# tb_writer = SummaryWriter('.')# LOGGER.info("Run 'tensorboard --logdir=models' to view tensorboard at http://localhost:6006/")# tb_writer.add_graph(torch.jit.trace(model, img, strict=False), []) # add model graph
