使用过程遇到下面错误，如何解决
Closed this issue · 2 comments
(base) matrix@matrix-MS-7C22:~/share/src/yolov5-label-xml$ python demo.py
5.jpg
Traceback (most recent call last):
File "demo.py", line 71, in
img0, xyxy_list,img_crop= yolo.detect(image) # img0检测后的图像，img_crop裁剪的图像
File "/home/matrix/share/src/yolov5-label-xml/utils/Detect_api.py", line 66, in detect
pred = self.model(img, augment=False)[0]
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/matrix/share/src/yolov5-label-xml/models/yolo.py", line 114, in forward
return self.forward_once(x, profile) # single-scale inference, train
File "/home/matrix/share/src/yolov5-label-xml/models/yolo.py", line 134, in forward_once
x = m(x) # run
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/matrix/share/src/yolov5-label-xml/models/common.py", line 92, in forward
return self.conv(torch.cat([x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]], 1))
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/matrix/share/src/yolov5-label-xml/models/common.py", line 27, in forward
return self.act(self.bn(self.conv(x)))
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/activation.py", line 438, in forward
return F.hardswish(input, self.inplace)
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 778, in getattr
raise ModuleAttributeError("'{}' object has no attribute '{}'".format(
torch.nn.modules.module.ModuleAttributeError: 'Hardswish' object has no attribute 'inplace'
我用的是yolov5s.pt文件，什么参数都没有修改测试的。
cuda:11.0, cuda:1.7.1
是哪边配置有问题?
(base) matrix@matrix-MS-7C22:~/share/src/yolov5-label-xml$ python demo.py
5.jpg
Traceback (most recent call last):
File "demo.py", line 71, in
img0, xyxy_list,img_crop= yolo.detect(image) # img0检测后的图像，img_crop裁剪的图像
File "/home/matrix/share/src/yolov5-label-xml/utils/Detect_api.py", line 66, in detect
pred = self.model(img, augment=False)[0]
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/matrix/share/src/yolov5-label-xml/models/yolo.py", line 114, in forward
return self.forward_once(x, profile) # single-scale inference, train
File "/home/matrix/share/src/yolov5-label-xml/models/yolo.py", line 134, in forward_once
x = m(x) # run
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/matrix/share/src/yolov5-label-xml/models/common.py", line 92, in forward
return self.conv(torch.cat([x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]], 1))
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/matrix/share/src/yolov5-label-xml/models/common.py", line 27, in forward
return self.act(self.bn(self.conv(x)))
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/activation.py", line 438, in forward
return F.hardswish(input, self.inplace)
File "/home/matrix/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 778, in getattr
raise ModuleAttributeError("'{}' object has no attribute '{}'".format(
torch.nn.modules.module.ModuleAttributeError: 'Hardswish' object has no attribute 'inplace'我用的是yolov5s.pt文件，什么参数都没有修改测试的。
cuda:11.0, cuda:1.7.1
是哪边配置有问题?
这个是yolov5-v3.0版本的，需要的torch环境是pytorch1.6，你的torch版本高了
谢谢~