报错整理1.24

1. 报错1：
   1. 报错：AssertionError: Torch not compiled with CUDA enabled
   2. 解决：先使用pip将当前具有的torch卸载，然后去pytorch官网复制下载指令
2. 报错2：
   1. 报错：ValueError: too many values to unpack (expected 2)
   2. 原因：函数返回值的个数和接收值的变量个数不对应
3. 报错3：
   1. 报错：RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument mat1 in method wrapper_addmm)：   
   2. 解决：使用tensor.is_cuda查看是否所有变量都在cuda上，并且要记得把model.to('cuda')
4. 报错4：
   1. 报错：在windows下的git bash中使用wget报错，找不到指令
   2. 解决：
      1. 下载wget.exe（下载地址：https://eternallybored.org/misc/）
      2. 使用where git 查看bin目录位置，将wget.exe放置到bin目录下再使用即可（后续linux中专有指令需要在git bash中使用，感觉都可以使用此方法）
5. 报错5：
   1. 报错：ImportError: Library “GLU“ not found.问题
   2. 解决：apt-get install python-opengl
6. 报错6：
   1. 报错：ValueError('Failed to initialize Pyglet window with an OpenGL >= 3+ context. If you\'re logged in via SSH, ensure that you\'re running your script with vglrun (i.e. VirtualGL). The internal error message was "Cannot connect to "None""',)
   2. 解决：加上import os; os.environ['PYOPENGL_PLATFORM'] = 'egl'即可解决
7. 报错7：
   1. 报错：ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
      pyrender 0.1.43 requires PyOpenGL==3.1.0, but you have pyopengl 3.1.6 which is incompatible.
   2. 通用的解决：当包的版本出现冲突时，把所有相关的包都uninstall ，然后再根据逐次进行安装
8. 报错8：
   1. 报错：mesa configure: error: Package requirements (expat21) were not met:  No package 'expat21' found  Con
   2. 解决：apt-get install libexpat1-dev
9. 报错9：
   1. 报错：Import "tensorflow.keras.utils" could not be resolved
   2. 解决：更新tensorflow版本 pip install --upgrade tensorflow
10. 报错10：
    1. 报错：torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 23.69 GiB total capacity; 20.91 GiB already allocated; 14.31 MiB free; 22.15 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
    2. 解决：改小batch_size（有用！！！）
11. 报错11：
    1. 报错：RuntimeError: DataLoader worker (pid 26533) is killed by signal: Bus error. It is possible that dataloader's workers are out of shared memory. Please try to raise your shared memory limit.
    2. 解决：减少num_workers数量 ，可解决
12. 报错12：
    1. 报错：Descriptors cannot not be created directly. If this call came from a _pb2.py file, your generated co 或者TypeError: Descriptors cannot not be created directly. If this call came from a _pb2.py file, your g
    2. 解决：pip uninstall protobuf
       pip install protobuf==3.20.1
13. 报错13：
    1. 报错：The command $ docker-compose could not be found in this WSL 1 distro. We recommend to convert this distro to WSL 2 and activate the WSL integration in Docker Desktop settings.
    2. 解决：打开docker desktop 设置->resource->wsl 将Ubuntu选项激活->apply && restart
14. 报错14：
    1. 报错：Cannot connect to the Docker daemon at unix:///var/run/docker.sock. Is the docker daemon running?
    2. 解决：以管理员身份运行wsl即可解决
15. 报错15：
    1. 首次运行anaconda环境时，报错：CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
       To initialize your shell, run
    2. 解决：先根据提示初始化shell,比如conda init bash，然后vi ~/.bashrc 增加anaconda3的环境变量，然后source ~/.bashrc 即可看到进入base虚拟环境了
16. 报错16：
    1. gzip: stdin: unexpected end of file tar: Unexpected EOF in archive tar: Unexpected EOF in archive ta
    2. 原因：有可能是数据还没有下载完成，即看到的压缩包还是和windows上没下载完全的一样，是个临时文件。
    3. 解决：删除，重新下载
17. 报错17：
    1. 报错 missing: line 81: makeinfo: command not found
    2. 解决：apt-get install texinfo
18. 报错18：
    1. 报错：PackagesNotFoundError: The following packages are not available from current channels:
    2. 解决：访问官网查询下载不了的包，看是在哪个channel上，然后将添加那个channel：conda config --append channels channel名字，然后再安装即可成功