子模组未对路径注册

报错：

git clone https://github.com/muskie82/MonoGS.git --recursive正克隆到 'MonoGS'...
remote: Enumerating objects: 246, done.
remote: Counting objects: 100% (134/134), done.
remote: Compressing objects: 100% (64/64), done.
remote: Total 246 (delta 94), reused 70 (delta 70), pack-reused 112
接收对象中: 100% (246/246), 25.25 MiB | 3.49 MiB/s, 完成.
处理 delta 中: 100% (120/120), 完成.
子模组 'submodules/diff-gaussian-rasterization'（https://github.com/rmurai0610/diff-gaussian-rasterization-w-pose.git）未对路径 'submodules/diff-gaussian-rasterization' 注册
子模组 'submodules/simple-knn'（https://gitlab.inria.fr/bkerbl/simple-knn.git）未对路径 'submodules/simple-knn' 注册
正克隆到 '/home/xx/MonoGS/submodules/diff-gaussian-rasterization'...
remote: Enumerating objects: 240, done.        
remote: Counting objects: 100% (240/240), done.        
remote: Compressing objects: 100% (72/72), done.        
remote: Total 240 (delta 155), reused 240 (delta 155), pack-reused 0        
接收对象中: 100% (240/240), 83.75 KiB | 539.00 KiB/s, 完成.
处理 delta 中: 100% (155/155), 完成.
正克隆到 '/home/xx/MonoGS/submodules/simple-knn'...
fatal: unable to access 'https://gitlab.inria.fr/bkerbl/simple-knn.git/': Failed to connect to gitlab.inria.fr port 443: 连接超时
fatal: 无法克隆 'https://gitlab.inria.fr/bkerbl/simple-knn.git' 到子模组路径 '/home/nuaa/MonoGS/submodules/simple-knn'
克隆 'submodules/simple-knn' 失败。按计划重试
正克隆到 '/home/xx/MonoGS/submodules/simple-knn'...
fatal: unable to access 'https://gitlab.inria.fr/bkerbl/simple-knn.git/': Failed to connect to gitlab.inria.fr port 443: 连接超时
fatal: 无法克隆 'https://gitlab.inria.fr/bkerbl/simple-knn.git' 到子模组路径 '/home/xx/MonoGS/submodules/simple-knn'
第二次尝试克隆 'submodules/simple-knn' 失败，退出

错误显示克隆simple-knn这个模块失败了，不知道是网络原因还是其他

执行：（初始化并更新所以子模块）

git submodule update --init --recursive

显示：

正克隆到 '/home/nuaa/MonoGS/submodules/simple-knn'...
子模组路径 'submodules/simple-knn'：检出 '44f764299fa305faf6ec5ebd99939e0508331503'

但是这又显示子模块安装上了，不清楚什么原因.......

ps:Git 已经成功克隆了子模块并检出了指定的提交（44f764299fa305faf6ec5ebd99939e0508331503 这个特定的提交哈希）这些信息表示你的子模块已经正确地被初始化和更新了，并且没有错误码。

总结：可能是网络问题，多试几次git submodule update --init --recursive，说不定就好了

-----------------------------------------------------------------------------------------------------------分割线

我这个项目submodule一共就俩，git clone项目的时候只给我报simple-knn没装上，后来多试几次就装上了，结果conda create的时候，又说那个子模块没装上。

报错：

CondaHTTPError: HTTP 000 CONNECTION FAILED for url <https://conda.anaconda.org/conda-forge/linux-64/cudatoolkit-11.3.1-hb98b00a_13.conda>                       
Elapsed: -                                                                      
                                                                                
An HTTP error occurred when trying to retrieve this URL.                        
HTTP errors are often intermittent, and a simple retry will get you on your way.
                                                                                
CondaHTTPError: HTTP 000 CONNECTION FAILED for url <https://conda.anaconda.org/conda-forge/linux-64/cudatoolkit-11.3.1-hb98b00a_13.conda>                       
Elapsed: -

An HTTP error occurred when trying to retrieve this URL.
HTTP errors are often intermittent, and a simple retry will get you on your way.

Installing pip dependencies: / Ran pip subprocess with arguments:
['/home/xx/anaconda3/envs/MonoGS/bin/python', '-m', 'pip', 'install', '-U', '-r', '/home/nuaa/MonoGS/condaenv.0qyohhwt.requirements.txt', '--exists-action=b']
Pip subprocess output:

Pip subprocess error:
ERROR: Directory 'submodules/diff-gaussian-rasterization' is not installable. Neither 'setup.py' nor 'pyproject.toml' found.

failed

CondaEnvException: Pip failed

解决方案：

去GitHub项目界面，进入子模块，在Ubuntu里对应项目对应子文件夹下，直接git clone，就能安装下来了

又报一些错：

Did you run some cuda functions before calling NumCudaDevices() that might have already set an error? Error 803: system has unsupported display driver / cuda driver combination (Triggered internally at  /opt/conda/conda-bld/pytorch_1659484809535/work/c10/cuda/CUDAFunctions.cpp:109.)
        return torch._C._cuda_getDeviceCount() > 0
      No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda-11.6'
      running bdist_wheel

 好像我nvidia的驱动掉了，输nvidia-smi没反应

输入nvcc --version

1.添加NVIDIA软件包仓库：

sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt-get update

sudo apt-get update是更新软件包索引，sudo apt-get upgrade是升级系统中已安装的软件包

2.安装推荐的驱动程序：

sudo ubuntu-drivers autoinstall

 安装驱动见另一篇我的另一篇博客