在部署yolov5自定义模型时引发错误

Question

Traceback (most recent call last):
  File "D:\Programming_Basic\AI_Basic\yolov5_AIBasic\yolov5\yo.py", line 4, in <module>
    model = torch.hub.load('ultralytics/yolov5', 'custom', path = '/yolov5/best.pt', force_reload=True)  # or yolov5m, yolov5l, yolov5x, custom
  File "C:\Users\Jayaram\AppData\Roaming\Python\Python39\site-packages\torch\hub.py", line 399, in load
    model = _load_local(repo_or_dir, model, *args, **kwargs)
  File "C:\Users\Jayaram\AppData\Roaming\Python\Python39\site-packages\torch\hub.py", line 428, in _load_local
    model = entry(*args, **kwargs)
  File "C:\Users\Jayaram/.cache\torch\hub\ultralytics_yolov5_master\hubconf.py", line 70, in custom
    return _create(path, autoshape=autoshape, verbose=verbose, device=device)
  File "C:\Users\Jayaram/.cache\torch\hub\ultralytics_yolov5_master\hubconf.py", line 65, in _create
    raise Exception(s) from e
Exception: path is on mount 'C:', start on mount 'D:'. Cache may be out of date, try `force_reload=True` or see https://github.com/ultralytics/yolov5/issues/36 for help.

Answer 1

我在使用yolov5烧瓶API时遇到了这个错误。

造成此错误的可能原因是os.relpath没有给出两个目录之间的相对路径。在Windows上，如果这两个目录位于不同的驱动器上(这正是错误消息所述)，则甚至不存在相对路径。

这个响应，这里，有助于更好地理解问题。

避免此问题的方法之一是将应用程序复制到C:驱动器，选择python解释器并尝试再次执行代码。Windows没有绝对根目录，相反，它对每个驱动器都有一个根目录。这意味着D:是相对于当前驱动器的路径，而不是绝对路径。

您可以在错误代码的前几行中看到c:和d:驱动器冲突。

Traceback (most recent call last):

File "D:\Programming_Basic\AI_Basic\yolov5_AIBasic\yolov5\yo.py", line 4, 
in <module> model = torch.hub.load('ultralytics/yolov5', 'custom', path = 
'/yolov5/best.pt', force_reload=True)  # or yolov5m, yolov5l, yolov5x, 
custom

File "C:\Users\Jayaram\AppData\Roaming\Python\Python39\site- 
packages\torch\hub.py", line 399, in load
model = _load_local(repo_or_dir, model, *args, **kwargs)

避免这种情况的最好方法是将代码粘贴到C驱动器中，否则您需要将路径变为绝对路径，而不是相对路径。