🌈🌈😄😄

歡迎來(lái)到茶色島獨(dú)家島嶼，本期將為大家揭曉YOLOV5報(bào)錯(cuò)解決辦法，做好準(zhǔn)備了么，那么開(kāi)始吧。

🌲🌲🐴🐴

1.在pycharm終端使用pip install requirements.txt報(bào)錯(cuò)時(shí)：

2.AssertionError: Image Not Found D:\PycharmProjects\yolov5-hat\VOCdevkit\images\train\000000.jpg問(wèn)題

3.?File "F:\All_fail\Procedure\Pytorch-Volo5\utils\datasets.py", line 312, in updatetime.sleep(2 / self.fps) # wait timeZeroDivisionError: float division by zero

4.安裝requirements.txt時(shí)出現(xiàn)錯(cuò)誤subprocess.CalledProcessError: Command 'pip install 'pycocotools>=2.0'' returned non-zero exit status 1.，這要怎么解決?

5.ModuleNotFoundError: No module named 'tensorboard'

6.RuntimeError: result type Float can't be cast to the desired output type __int64報(bào)錯(cuò)

?1.在pycharm終端使用pip install requirements.txt報(bào)錯(cuò)時(shí)：

給大家避個(gè)坑，我搞了2天才發(fā)現(xiàn)

安裝requirement后運(yùn)行train或者detect的時(shí)候仍出現(xiàn)import報(bào)錯(cuò)比如yaml的

去anaconda里面 conda activate pytorch

然后 pip install -r （路徑requirements.txt） -i https://pypi.tuna.tsinghua.edu.cn/simple

其中requirement.txt建議單獨(dú)放到桌面，可以把括號(hào)替換成Desktop\requirement.txt

?2.AssertionError: Image Not Found D:\PycharmProjects\yolov5-hat\VOCdevkit\images\train\000000.jpg問(wèn)題

刪除緩存的方法：刪除yolov5-hat\VOCdevkit\labels里tran和val cache

3.?File "F:\All_fail\Procedure\Pytorch-Volo5\utils\datasets.py", line 312, in update
time.sleep(2 / self.fps) # wait time
ZeroDivisionError: float division by zero

鐵子們，我報(bào)了這個(gè)錯(cuò)誤，記錄一下，有用自取：我把括號(hào)里的(2 / self.fps)改成0解決了

4.安裝requirements.txt時(shí)出現(xiàn)錯(cuò)誤subprocess.CalledProcessError: Command 'pip install 'pycocotools>=2.0'' returned non-zero exit status 1.，這要怎么解決?

在pycharm終端pip install pycocotools

5.ModuleNotFoundError: No module named 'tensorboard'

?直接在終端pip install?tensorboard

6.RuntimeError: result type Float can't be cast to the desired output type __int64報(bào)錯(cuò)

將loss.py中211行改為
indices.append((b, a, gj.clamp_(0, int(gain[3]) - 1), gi.clamp_(0, int(gain[2]) - 1)))

以上是我遇到的一些問(wèn)題，若有其他問(wèn)題無(wú)法解決，可留言。?