YOLOV5引入SE注意力机制以及精度提升问题

1. 如何增加SE注意力机制

• 在model/common.py中添加SE结构

class SE(nn.Module):def __init__(self, c1, c2, r=16):super(SE, self).__init__()self.avgpool = nn.AdaptiveAvgPool2d(1)self.l1 = nn.Linear(c1, c1 // r, bias=False)self.relu = nn.ReLU(inplace=True)self.l2 = nn.Linear(c1 // r, c1, bias=False)self.sig = nn.Sigmoid()def forward(self, x):print(x.size())b, c, _, _ = x.size()y = self.avgpool(x).view(b, c)y = self.l1(y)y = self.relu(y)y = self.l2(y)y = self.sig(y)y = y.view(b, c, 1, 1)return x * y.expand_as(x)

• 以直接修改yolov5s.yaml 为例讲两种思路

  思路1：直接放在backbone末尾

  backbone:# [from, number, module, args][[-1, 1, Conv, [64, 6, 2, 2]],  # 0-P1/2[-1, 1, Conv, [128, 3, 2]],  # 1-P2/4[-1, 3, C3, [128]],[-1, 1, Conv, [256, 3, 2]],  # 3-P3/8[-1, 6, C3, [256]],[-1, 1, Conv, [512, 3, 2]],  # 5-P4/16[-1, 9, C3, [512]], # 6[-1, 1, Conv, [1024, 3, 2]],  # 7-P5/32[-1, 3, C3, [1024]],[-1, 1, SPPF, [1024, 5]],  # 9[-1, 1, SE, [1024，2]], ]
  

  思路2：放在SPPF前

  backbone:# [from, number, module, args][[-1, 1, Conv, [64, 6, 2, 2]],  # 0-P1/2[-1, 1, Conv, [128, 3, 2]],  # 1-P2/4[-1, 3, C3, [128]],[-1, 1, Conv, [256, 3, 2]],  # 3-P3/8[-1, 6, C3, [256]],[-1, 1, Conv, [512, 3, 2]],  # 5-P4/16[-1, 9, C3, [512]], # 6[-1, 1, Conv, [1024, 3, 2]],  # 7-P5/32[-1, 3, C3, [1024]],[-1, 1, SE, [1024，2]], [-1, 1, SPPF, [1024, 5]],  # 10]
  

  (以上思路2选1)

  重要！：添加完SE之后，相应的head层（超过10的）都需要将层数+1

  head修改为：

  head:[[-1, 1, Conv, [512, 1, 1]],[-1, 1, nn.Upsample, [None, 2, 'nearest']],[[-1, 6], 1, Concat, [1]],  # cat backbone P4[-1, 3, C3, [512, False]],  # 13[-1, 1, Conv, [256, 1, 1]],[-1, 1, nn.Upsample, [None, 2, 'nearest']],[[-1, 4], 1, Concat, [1]],  # cat backbone P3[-1, 3, C3, [256, False]],  # 17 (P3/8-small)[-1, 1, Conv, [256, 3, 2]],[[-1, 15], 1, Concat, [1]],  # cat head P4  +![-1, 3, C3, [512, False]],  # 20 (P4/16-medium)[-1, 1, Conv, [512, 3, 2]],[[-1, 11], 1, Concat, [1]],  # cat head P5    +1[-1, 3, C3, [1024, False]],  # 23 (P5/32-large)[[18, 21, 24], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5)   +1]
  

• 修改yolo.py ，在def parse_model(d, ch): 下添加SE模块判断语句：

elif m is SE:c1 = ch[f]c2 = args[0]if c2 !=no:c2 = make_divisible(c2 * gw, 8)args = [c1, args[1]]

即完整yolo.py:

# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
"""
YOLO-specific modulesUsage:$ python models/yolo.py --cfg yolov5s.yaml
"""import argparse
import contextlib
import os
import platform
import sys
from copy import deepcopy
from pathlib import PathFILE = 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
if platform.system() != 'Windows':ROOT = Path(os.path.relpath(ROOT, 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, profile, scale_img, select_device,time_sync)try:import thop  # for FLOPs computation
except ImportError:thop = Noneclass Detect(nn.Module):# YOLOv5 Detect head for detection modelsstride = None  # strides computed during builddynamic = False  # force grid reconstructionexport = False  # export modedef __init__(self, nc=80, anchors=(), ch=(), inplace=True):  # detection layersuper().__init__()self.nc = nc  # number of classesself.no = nc + 5  # number of outputs per anchorself.nl = len(anchors)  # number of detection layersself.na = len(anchors[0]) // 2  # number of anchorsself.grid = [torch.empty(0) for _ in range(self.nl)]  # init gridself.anchor_grid = [torch.empty(0) for _ in range(self.nl)]  # init anchor gridself.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2))  # shape(nl,na,2)self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch)  # output convself.inplace = inplace  # use inplace ops (e.g. slice assignment)def forward(self, x):z = []  # inference outputfor i in range(self.nl):x[i] = self.m[i](x[i])  # convbs, _, 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()if not self.training:  # inferenceif self.dynamic or self.grid[i].shape[2:4] != x[i].shape[2:4]:self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i)if isinstance(self, Segment):  # (boxes + masks)xy, wh, conf, mask = x[i].split((2, 2, self.nc + 1, self.no - self.nc - 5), 4)xy = (xy.sigmoid() * 2 + self.grid[i]) * self.stride[i]  # xywh = (wh.sigmoid() * 2) ** 2 * self.anchor_grid[i]  # why = torch.cat((xy, wh, conf.sigmoid(), mask), 4)else:  # Detect (boxes only)xy, wh, conf = x[i].sigmoid().split((2, 2, self.nc + 1), 4)xy = (xy * 2 + self.grid[i]) * self.stride[i]  # xywh = (wh * 2) ** 2 * self.anchor_grid[i]  # why = torch.cat((xy, wh, conf), 4)z.append(y.view(bs, self.na * nx * ny, self.no))return x if self.training else (torch.cat(z, 1),) if self.export else (torch.cat(z, 1), x)def _make_grid(self, nx=20, ny=20, i=0, torch_1_10=check_version(torch.__version__, '1.10.0')):d = self.anchors[i].devicet = self.anchors[i].dtypeshape = 1, self.na, ny, nx, 2  # grid shapey, x = torch.arange(ny, device=d, dtype=t), torch.arange(nx, device=d, dtype=t)yv, xv = torch.meshgrid(y, x, indexing='ij') if torch_1_10 else torch.meshgrid(y, x)  # torch>=0.7 compatibilitygrid = torch.stack((xv, yv), 2).expand(shape) - 0.5  # add grid offset, i.e. y = 2.0 * x - 0.5anchor_grid = (self.anchors[i] * self.stride[i]).view((1, self.na, 1, 1, 2)).expand(shape)return grid, anchor_gridclass Segment(Detect):# YOLOv5 Segment head for segmentation modelsdef __init__(self, nc=80, anchors=(), nm=32, npr=256, ch=(), inplace=True):super().__init__(nc, anchors, ch, inplace)self.nm = nm  # number of masksself.npr = npr  # number of protosself.no = 5 + nc + self.nm  # number of outputs per anchorself.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch)  # output convself.proto = Proto(ch[0], self.npr, self.nm)  # protosself.detect = Detect.forwarddef forward(self, x):p = self.proto(x[0])x = self.detect(self, x)return (x, p) if self.training else (x[0], p) if self.export else (x[0], p, x[1])class BaseModel(nn.Module):# YOLOv5 base modeldef forward(self, x, profile=False, visualize=False):return self._forward_once(x, profile, visualize)  # single-scale inference, traindef _forward_once(self, x, profile=False, visualize=False):y, dt = [], []  # outputsfor m in self.model:if m.f != -1:  # if not from previous layerx = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f]  # from earlier layersif profile:self._profile_one_layer(m, x, dt)x = m(x)  # runy.append(x if m.i in self.save else None)  # save outputif visualize:feature_visualization(x, m.type, m.i, save_dir=visualize)return xdef _profile_one_layer(self, m, x, dt):c = m == self.model[-1]  # 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 fuse(self):  # fuse model Conv2d() + BatchNorm2d() layersLOGGER.info('Fusing layers... ')for m in self.model.modules():if isinstance(m, (Conv, DWConv)) and hasattr(m, 'bn'):m.conv = fuse_conv_and_bn(m.conv, m.bn)  # update convdelattr(m, 'bn')  # remove batchnormm.forward = m.forward_fuse  # update forwardself.info()return selfdef info(self, verbose=False, img_size=640):  # print model informationmodel_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, Segment)):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 selfclass DetectionModel(BaseModel):# YOLOv5 detection modeldef __init__(self, cfg='yolov5s.yaml', ch=3, nc=None, anchors=None):  # model, input channels, number of classessuper().__init__()if isinstance(cfg, dict):self.yaml = cfg  # model dictelse:  # is *.yamlimport yaml  # for torch hubself.yaml_file = Path(cfg).namewith open(cfg, encoding='ascii', errors='ignore') as f:self.yaml = yaml.safe_load(f)  # model dict# Define modelch = self.yaml['ch'] = self.yaml.get('ch', ch)  # input channelsif 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 valueif anchors:LOGGER.info(f'Overriding model.yaml anchors with anchors={anchors}')self.yaml['anchors'] = round(anchors)  # override yaml valueself.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch])  # model, savelistself.names = [str(i) for i in range(self.yaml['nc'])]  # default namesself.inplace = self.yaml.get('inplace', True)# Build strides, anchorsm = self.model[-1]  # Detect()if isinstance(m, (Detect, Segment)):s = 256  # 2x min stridem.inplace = self.inplaceforward = lambda x: self.forward(x)[0] if isinstance(m, Segment) else self.forward(x)m.stride = torch.tensor([s / x.shape[-2] for x in forward(torch.zeros(1, ch, s, s))])  # forwardcheck_anchor_order(m)m.anchors /= m.stride.view(-1, 1, 1)self.stride = m.strideself._initialize_biases()  # only run once# Init weights, biasesinitialize_weights(self)self.info()LOGGER.info('')def forward(self, x, augment=False, profile=False, visualize=False):if augment:return self._forward_augment(x)  # augmented inference, Nonereturn 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):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])  # saveyi = 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 _descale_pred(self, p, flips, scale, img_size):# de-scale predictions following augmented inference (inverse operation)if self.inplace:p[..., :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 _initialize_biases(self, cf=None):  # initialize biases into Detect(), cf is class frequency# https://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[:, 5:5 + m.nc] += math.log(0.6 / (m.nc - 0.99999)) if cf is None else torch.log(cf / cf.sum())  # clsmi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True)Model = DetectionModel  # retain YOLOv5 'Model' class for backwards compatibilityclass SegmentationModel(DetectionModel):# YOLOv5 segmentation modeldef __init__(self, cfg='yolov5s-seg.yaml', ch=3, nc=None, anchors=None):super().__init__(cfg, ch, nc, anchors)class ClassificationModel(BaseModel):# YOLOv5 classification modeldef __init__(self, cfg=None, model=None, nc=1000, cutoff=10):  # yaml, model, number of classes, cutoff indexsuper().__init__()self._from_detection_model(model, nc, cutoff) if model is not None else self._from_yaml(cfg)def _from_detection_model(self, model, nc=1000, cutoff=10):# Create a YOLOv5 classification model from a YOLOv5 detection modelif isinstance(model, DetectMultiBackend):model = model.model  # unwrap DetectMultiBackendmodel.model = model.model[:cutoff]  # backbonem = model.model[-1]  # last layerch = m.conv.in_channels if hasattr(m, 'conv') else m.cv1.conv.in_channels  # ch into modulec = Classify(ch, nc)  # Classify()c.i, c.f, c.type = m.i, m.f, 'models.common.Classify'  # index, from, typemodel.model[-1] = c  # replaceself.model = model.modelself.stride = model.strideself.save = []self.nc = ncdef _from_yaml(self, cfg):# Create a YOLOv5 classification model from a *.yaml fileself.model = Nonedef parse_model(d, ch):  # model_dict, input_channels(3)# Parse a YOLOv5 model.yaml dictionaryLOGGER.info(f"\\n{'':>3}{'from':>18}{'n':>3}{'params':>10}  {'module':<40}{'arguments':<30}")anchors, nc, gd, gw, act = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'], d.get('activation')if act:Conv.default_act = eval(act)  # redefine default activation, i.e. Conv.default_act = nn.SiLU()LOGGER.info(f"{colorstr('activation:')} {act}")  # printna = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors  # number of anchorsno = na * (nc + 5)  # number of outputs = anchors * (classes + 5)layers, save, c2 = [], [], ch[-1]  # layers, savelist, ch outfor i, (f, n, m, args) in enumerate(d['backbone'] + d['head']):  # from, number, module, argsm = eval(m) if isinstance(m, str) else m  # eval stringsfor j, a in enumerate(args):with contextlib.suppress(NameError):args[j] = eval(a) if isinstance(a, str) else a  # eval stringsn = 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, nn.ConvTranspose2d, DWConvTranspose2d, C3x}:c1, c2 = ch[f], args[0]if c2 != no:  # if not outputc2 = make_divisible(c2 * gw, 8)args = [c1, c2, *args[1:]]if m in {BottleneckCSP, C3, C3TR, C3Ghost, C3x}:args.insert(2, n)  # number of repeatsn = 1elif m is nn.BatchNorm2d:args = [ch[f]]elif m is Concat:c2 = sum(ch[x] for x in f)elif m is SE:c1 = ch[f]c2 = args[0]if c2 !=no:c2 = make_divisible(c2 * gw, 8)args = [c1, args[1]]# TODO: channel, gw, gdelif m in {Detect, Segment}:args.append([ch[x] for x in f])if isinstance(args[1], int):  # number of anchorsargs[1] = [list(range(args[1] * 2))] * len(f)if m is Segment:args[3] = make_divisible(args[3] * gw, 8)elif m is Contract:c2 = ch[f] * args[0] ** 2elif m is Expand:c2 = ch[f] // args[0] ** 2else:c2 = ch[f]m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args)  # modulet = 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}')  # printsave.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1)  # append to savelistlayers.append(m_)if i == 0:ch = []ch.append(c2)return nn.Sequential(*layers), sorted(save)if __name__ == '__main__':parser = argparse.ArgumentParser()parser.add_argument('--cfg', type=str, default='yolov5s.yaml', help='model.yaml')parser.add_argument('--batch-size', type=int, default=1, help='total batch size for all GPUs')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('--line-profile', action='store_true', help='profile model speed layer by layer')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(vars(opt))device = select_device(opt.device)# Create modelim = torch.rand(opt.batch_size, 3, 640, 640).to(device)model = Model(opt.cfg).to(device)# Optionsif opt.line_profile:  # profile layer by layermodel(im, profile=True)elif opt.profile:  # profile forward-backwardresults = profile(input=im, ops=[model], n=3)elif opt.test:  # test all modelsfor cfg in Path(ROOT / 'models').rglob('yolo*.yaml'):try:_ = Model(cfg)except Exception as e:print(f'Error in {cfg}: {e}')else:  # report fused model summarymodel.fuse()

2. 训练

在colab中，以coco128为数据集，官方的yolov5s.pt作为预训练模型，训练400个epochs。

!python train.py --img 640 --batch 16 --epochs 400 --data coco128.yaml --weights yolov5s.pt --cache

3. 精度

• 当SE添加在backbone最后一层时
  

mAP50: 87.4

mAP50-95: 65.9

• 当SE添加在backbone倒数第二层时
  

mAP50: 80.1

mAP50-95: 60.7

整体上看，SE的添加可以显著提升mAP，而添加在backbone最后一层时最棒。