PeterH0323 / Smart_Construction

Base on YOLOv5 Head Person Helmet Detection on Construction Sites,基于目标检测工地安全帽和禁入危险区域识别系统,🚀😆附 YOLOv5 训练自己的数据集超详细教程🚀😆2021.3新增可视化界面❗❗
GNU General Public License v3.0
2.19k stars 464 forks source link

这里给大家提几个小建议! #83

Open xs-web-lhdd opened 1 year ago

xs-web-lhdd commented 1 year ago

1、作者提供的网盘数据集中JPEGImages要检查所有图片后缀都是jpg,把JGP的改为jpg要不然会在数据转换时运行报错 2、在yolo.py里面

        for mi, s in zip(m.m, m.stride):  #  from
            b = mi.bias.view(m.na, -1)  # conv.bias(255) to (3,85)
            with torch.no_grad(): # 在原有代码基础上添加这行代码
                b[:, 4] += math.log(8 / (640 / s) ** 2)  # obj (8 objects per 640 image) # 缩进
                b[:, 5:] += math.log(0.6 / (m.nc - 0.99)) if cf is None else torch.log(cf / cf.sum())  # cls # 缩进
            mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True)

3、在output_to_target中进行修改,将以下行: return np.array(targets)修改为:return np.array(torch.tensor(targets).cpu())

以上代表我个人观点,大佬轻喷哈哈哈哈哈!

daqiudi commented 4 months ago

1、作者提供的网盘数据集中JPEGImages要检查所有图片后缀都是jpg,把JGP的改为jpg要不然会在数据转换时运行报错 2、在yolo.py里面

        for mi, s in zip(m.m, m.stride):  #  from
            b = mi.bias.view(m.na, -1)  # conv.bias(255) to (3,85)
            with torch.no_grad(): # 在原有代码基础上添加这行代码
                b[:, 4] += math.log(8 / (640 / s) ** 2)  # obj (8 objects per 640 image) # 缩进
                b[:, 5:] += math.log(0.6 / (m.nc - 0.99)) if cf is None else torch.log(cf / cf.sum())  # cls # 缩进
            mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True)

3、在output_to_target中进行修改,将以下行: return np.array(targets)修改为:return np.array(torch.tensor(targets).cpu())

以上代表我个人观点,大佬轻喷哈哈哈哈哈!

牛的,完美预判!