【小土堆】 AttributeError: Can‘t get attribute ‘SPPF‘ on ＜module ‘models.common‘ from 报错

ERROR 1：AttributeError: Can’t get attribute ‘SPPF’ on <module ‘models.common’ from '地址

在 common.py 文件中添加SPPF类

class SPPF(nn.Module):
    # Spatial Pyramid Pooling - Fast (SPPF) layer for YOLOv5 by Glenn Jocher
    def __init__(self, c1, c2, k=5):  # equivalent to SPP(k=(5, 9, 13))
        super().__init__()
        c_ = c1 // 2  # hidden channels
        self.cv1 = Conv(c1, c_, 1, 1)
        self.cv2 = Conv(c_ * 4, c2, 1, 1)
        self.m = nn.MaxPool2d(kernel_size=k, stride=1, padding=k // 2)

    def forward(self, x):
        x = self.cv1(x)
        with warnings.catch_warnings():
            warnings.simplefilter('ignore')  # suppress torch 1.9.0 max_pool2d() warning
            y1 = self.m(x)
            y2 = self.m(y1)
            return self.cv2(torch.cat([x, y1, y2, self.m(y2)], 1))

warrings会标红，记得 import warrings

ERROR2：RuntimeError: The size of tensor a (80) must match the size of tensor b (56) at non-singleton dimension 3

这个我猜可能是预训练权重yolov5s.pt]的问题，因为小土堆是5.0的工程，默认下载的是最新的模型，所以不匹配
所以我又去下载了一个，然后把 yolov5s.pt 给覆盖掉了，这个错误就消失了

附上下载链接：https://github.com/ultralytics/yolov5/releases/download/v5.0/yolov5s.pt
————————————————
参考 王十二er
原文链接：https://blog.csdn.net/qq_41885018/article/details/132697325