qidao123.com ToB IT社区-企服评测·应用市场»论坛 › 软件与程序人生 › DevOps与敏捷开发 › apex安装
apex安装 [复制链接]
发表于 2025-1-8 11:59:18 | 显示全部楼层 |阅读模式

马上注册，结交更多好友，享用更多功能，让你轻松玩转社区。

您需要 登录 才可以下载或查看，没有账号？立即注册

×
 安装过程复杂曲折，网上说的很多办法，貌似乐成了，实际还是没起作用。
先说乐成过程，执行下面下令，安装乐成（当然，条件是你要先设置好编译情况）：
（我的情况：windows10 、embeded python3.11.9 、pytorch version: 2.5.1+cu124、RTX2080ti）
  1. git clone https://github.com/NVIDIA/apex.git
  2.  
  3. cd apex
  5. curl -L "https://github.com/NVIDIA/apex/files/14844602/windows_support.patch" | git apply
  7. pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings "--build-option=--cpp_ext" --config-settings "--build-option=--cuda_ext" .
复制代码
这个信息，不知道是否说明我的显卡是2系（TORCH_CUDA_ARCH_LIST=7.5）？


乐成安装的apex，在卸载apex，你会发现安装的apex含有下面部件： 

---------------下面只是曲折过程，如无须要就不要浪费时间往下看了--------------
1、在comfyUI中，我安装了54个节点，只在安装了节点ComfyUI-PuLID-Flux-Enhanced后，在没有安装apex时，会有如下提示：

于是便想着要安装APEX，看看会不会速率快点，颠末了漫长而曲折的过程，终于安装乐成，但效果好像没什么变革。

从apex上述介绍中，是否表示，大多时候在Pytorch中已含相关代码？所以，究竟是否值得再花大力气去安装apex呢？尚有待进一步探究。
我的情况是python3.11.9 、pytorch version: 2.5.1+cu124
2、起首是报如下错误： 
  1.   building 'amp_C' extension
  2.   H:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\utils\cpp_extension.py:1964: UserWarning: TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation.
  3.   If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST'].
  4.     warnings.warn(
  5.   Emitting ninja build file H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\build.ninja...
  6.   Compiling objects...
  7.   Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
  8.   [1/7] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\bin\nvcc --generate-dependencies-with-compile --dependency-output H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\csrc/multi_tensor_scale_kernel.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\torch\csrc\api\include -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\TH -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\include" -IH:\ComfyUI_cp3.11.9\python_embeded\include -IH:\ComfyUI_cp3.11.9\python_embeded\Include "-IC:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Tools\MSVC\14.40.33807\include" "-IC:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\cppwinrt" -c H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_scale_kernel.cu -o H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\csrc/multi_tensor_scale_kernel.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -lineinfo -O3 --use_fast_math -DVERSION_GE_1_1 -DVERSION_GE_1_3 -DVERSION_GE_1_5 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=amp_C -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_75,code=compute_75 -gencode=arch=compute_75,code=sm_75
  9.   FAILED: H:/ComfyUI_cp3.11.9/apex/build/temp.win-amd64-cpython-311/Release/csrc/multi_tensor_scale_kernel.obj
  10.   C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\bin\nvcc --generate-dependencies-with-compile --dependency-output H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\csrc/multi_tensor_scale_kernel.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\torch\csrc\api\include -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\TH -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\include" -IH:\ComfyUI_cp3.11.9\python_embeded\include -IH:\ComfyUI_cp3.11.9\python_embeded\Include "-IC:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Tools\MSVC\14.40.33807\include" "-IC:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\cppwinrt" -c H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_scale_kernel.cu -o H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\csrc/multi_tensor_scale_kernel.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -lineinfo -O3 --use_fast_math -DVERSION_GE_1_1 -DVERSION_GE_1_3 -DVERSION_GE_1_5 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=amp_C -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_75,code=compute_75 -gencode=arch=compute_75,code=sm_75
  11.   C:/Program Files (x86)/Microsoft Visual Studio/2022/BuildTools/VC/Tools/MSVC/14.40.33807/include\type_traits(1167): error: static assertion failed with "You've instantiated std::aligned_storage<Len, Align> with an extended alignment (in other words, Align > alignof(max_align_t)). Before VS 2017 15.8, the member "type" would non-conformingly have an alignment of only alignof(max_align_t). VS 2017 15.8 was fixed to handle this correctly, but the fix inherently changes layout and breaks binary compatibility (*only* for uses of aligned_storage with extended alignments). To suppress this error, please define either (1) _ENABLE_EXTENDED_ALIGNED_STORAGE to confirm that you want a type with an extended alignment, or (2) _DISABLE_EXTENDED_ALIGNED_STORAGE to get the old non-conforming behavior."
  12.         static_assert(_Always_false<_Aligned>,
  13.         ^
  14.             detected during:
  15.               instantiation of class "std::_Aligned<_Len, _Align, double, false> [with _Len=16ULL, _Align=16ULL]" at line 1185
  16.               instantiation of class "std::_Aligned<_Len, _Align, int, false> [with _Len=16ULL, _Align=16ULL]" at line 1192
  17.               instantiation of class "std::_Aligned<_Len, _Align, short, false> [with _Len=16ULL, _Align=16ULL]" at line 1199
  18.               instantiation of class "std::_Aligned<_Len, _Align, char, false> [with _Len=16ULL, _Align=16ULL]" at line 1207
  19.               instantiation of class "std::_Aligned_storage<_Len, _Align> [with _Len=16ULL, _Align=16ULL]" at line 1211
  20.               instantiation of type "std::_Aligned_storage_t<16ULL, 16ULL>" at line 1215
  21.               instantiation of class "std::aligned_storage<_Len, _Align> [with _Len=16ULL, _Align=16ULL]" at line 25 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_scale_kernel.cu
  22.               instantiation of "void load_store(T *, T *, int, int) [with T=float]" at line 64 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_scale_kernel.cu
  23.               instantiation of "void ScaleFunctor<in_t, out_t>::operator()(int, volatile int *, TensorListMetadata<2> &, float) [with in_t=float, out_t=float]" at line 38 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_apply.cuh
  24.               instantiation of "void multi_tensor_apply_kernel(int64_t, volatile int *, T, U, ArgTypes...) [with T=TensorListMetadata<2>, U=ScaleFunctor<float, float>, ArgTypes=<float>]" at line 109 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_apply.cuh
  25.               instantiation of "void multi_tensor_apply<depth,T,ArgTypes...>(int64_t, int64_t, const at::Tensor &, const std::vector<std::vector<at::Tensor, std::allocator<at::Tensor>>, std::allocator<std::vector<at::Tensor, std::allocator<at::Tensor>>>> &, T, ArgTypes...) [with depth=2, T=ScaleFunctor<float, float>, ArgTypes=<float>]" at line 124 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_scale_kernel.cu
  29.   1 error detected in the compilation of "H:/ComfyUI_cp3.11.9/apex/csrc/multi_tensor_scale_kernel.cu".
  31.   multi_tensor_scale_kernel.cu
复制代码
 针对出现的第一个错误：
  1. H:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\utils\cpp_extension.py:1964: UserWarning: TORCH_CUDA_ARCH_LIST is not set, all archs for visible cards are included for compilation.
  2.   If this is not desired, please set os.environ['TORCH_CUDA_ARCH_LIST'].
复制代码
参考文章《如何设置 TORCH_CUDA_ARCH_LIST 情况变量以优化 PyTorch 性能》，我的显卡是RTX2080Ti，所以设置：
  1. set TORCH_CUDA_ARCH_LIST=7.5
复制代码
再次编译，错误信息如下：
  1.   已完成代码的生成
  2.   building 'amp_C' extension
  3.   Emitting ninja build file H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\build.ninja...
  4.   Compiling objects...
  5.   Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
  6.   [1/14] C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\bin\nvcc --generate-dependencies-with-compile --dependency-output H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\csrc/multi_tensor_l2norm_scale_kernel.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\torch\csrc\api\include -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\TH -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\include" -IH:\ComfyUI_cp3.11.9\python_embeded\include -IH:\ComfyUI_cp3.11.9\python_embeded\Include "-IC:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Tools\MSVC\14.40.33807\include" "-IC:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\cppwinrt" -c H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_l2norm_scale_kernel.cu -o H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\csrc/multi_tensor_l2norm_scale_kernel.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -lineinfo -O3 --use_fast_math -DVERSION_GE_1_1 -DVERSION_GE_1_3 -DVERSION_GE_1_5 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=amp_C -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_75,code=sm_75
  7.   FAILED: H:/ComfyUI_cp3.11.9/apex/build/temp.win-amd64-cpython-311/Release/csrc/multi_tensor_l2norm_scale_kernel.obj
  8.   C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\bin\nvcc --generate-dependencies-with-compile --dependency-output H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\csrc/multi_tensor_l2norm_scale_kernel.obj.d -std=c++17 --use-local-env -Xcompiler /MD -Xcompiler /wd4819 -Xcompiler /wd4251 -Xcompiler /wd4244 -Xcompiler /wd4267 -Xcompiler /wd4275 -Xcompiler /wd4018 -Xcompiler /wd4190 -Xcompiler /wd4624 -Xcompiler /wd4067 -Xcompiler /wd4068 -Xcompiler /EHsc -Xcudafe --diag_suppress=base_class_has_different_dll_interface -Xcudafe --diag_suppress=field_without_dll_interface -Xcudafe --diag_suppress=dll_interface_conflict_none_assumed -Xcudafe --diag_suppress=dll_interface_conflict_dllexport_assumed -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\torch\csrc\api\include -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\TH -IH:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\torch\include\THC "-IC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\include" -IH:\ComfyUI_cp3.11.9\python_embeded\include -IH:\ComfyUI_cp3.11.9\python_embeded\Include "-IC:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Tools\MSVC\14.40.33807\include" "-IC:\Program Files (x86)\Microsoft Visual Studio\2022\BuildTools\VC\Auxiliary\VS\include" "-IC:\Program Files (x86)\Windows Kits\10\include\10.0.22621.0\ucrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\um" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\shared" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\winrt" "-IC:\Program Files (x86)\Windows Kits\10\\include\10.0.22621.0\\cppwinrt" -c H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_l2norm_scale_kernel.cu -o H:\ComfyUI_cp3.11.9\apex\build\temp.win-amd64-cpython-311\Release\csrc/multi_tensor_l2norm_scale_kernel.obj -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -lineinfo -O3 --use_fast_math -DVERSION_GE_1_1 -DVERSION_GE_1_3 -DVERSION_GE_1_5 -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=amp_C -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_75,code=sm_75
  9.   C:/Program Files (x86)/Microsoft Visual Studio/2022/BuildTools/VC/Tools/MSVC/14.40.33807/include\type_traits(1167): error: static assertion failed with "You've instantiated std::aligned_storage<Len, Align> with an extended alignment (in other words, Align > alignof(max_align_t)). Before VS 2017 15.8, the member "type" would non-conformingly have an alignment of only alignof(max_align_t). VS 2017 15.8 was fixed to handle this correctly, but the fix inherently changes layout and breaks binary compatibility (*only* for uses of aligned_storage with extended alignments). To suppress this error, please define either (1) _ENABLE_EXTENDED_ALIGNED_STORAGE to confirm that you want a type with an extended alignment, or (2) _DISABLE_EXTENDED_ALIGNED_STORAGE to get the old non-conforming behavior."
  10.         static_assert(_Always_false<_Aligned>,
  11.         ^
  12.             detected during:
  13.               instantiation of class "std::_Aligned<_Len, _Align, double, false> [with _Len=16ULL, _Align=16ULL]" at line 1185
  14.               instantiation of class "std::_Aligned<_Len, _Align, int, false> [with _Len=16ULL, _Align=16ULL]" at line 1192
  15.               instantiation of class "std::_Aligned<_Len, _Align, short, false> [with _Len=16ULL, _Align=16ULL]" at line 1199
  16.               instantiation of class "std::_Aligned<_Len, _Align, char, false> [with _Len=16ULL, _Align=16ULL]" at line 1207
  17.               instantiation of class "std::_Aligned_storage<_Len, _Align> [with _Len=16ULL, _Align=16ULL]" at line 1211
  18.               instantiation of type "std::_Aligned_storage_t<16ULL, 16ULL>" at line 1215
  19.               instantiation of class "std::aligned_storage<_Len, _Align> [with _Len=16ULL, _Align=16ULL]" at line 24 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_l2norm_scale_kernel.cu
  20.               instantiation of "void load_store(T *, T *, int, int) [with T=float]" at line 75 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_l2norm_scale_kernel.cu
  21.               instantiation of "void L2NormScaleFunctor<in_t, out_t>::operator()(int, volatile int *, TensorListMetadata<2> &, float *, float *, float, __nv_bool, int) [with in_t=float, out_t=float]" at line 38 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_apply.cuh
  22.               instantiation of "void multi_tensor_apply_kernel(int64_t, volatile int *, T, U, ArgTypes...) [with T=TensorListMetadata<2>, U=L2NormScaleFunctor<float, float>, ArgTypes=<float *, float *, float, __nv_bool, int>]" at line 109 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_apply.cuh
  23.               instantiation of "void multi_tensor_apply<depth,T,ArgTypes...>(int64_t, int64_t, const at::Tensor &, const std::vector<std::vector<at::Tensor, std::allocator<at::Tensor>>, std::allocator<std::vector<at::Tensor, std::allocator<at::Tensor>>>> &, T, ArgTypes...) [with depth=2, T=L2NormScaleFunctor<float, float>, ArgTypes=<float *, float *, float, __nv_bool, int>]" at line 292 of H:\ComfyUI_cp3.11.9\apex\csrc\multi_tensor_l2norm_scale_kernel.cu
  27.   1 error detected in the compilation of "H:/ComfyUI_cp3.11.9/apex/csrc/multi_tensor_l2norm_scale_kernel.cu".
  29.   multi_tensor_l2norm_scale_kernel.cu
复制代码
3、无奈之下，想着试试 torch.__version__  = 2.5.1+cpu时，错误信息提示如下：
  1. H:\ComfyUI_cp3.11.9\apex>..\python_embeded\python -m pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings "--build-option=--cpp_ext" --config-settings "--build-option=--cuda_ext" .
  2. Using pip 24.3.1 from H:\ComfyUI_cp3.11.9\python_embeded\Lib\site-packages\pip (python 3.11)
  3. Processing h:\comfyui_cp3.11.9\apex
  4.   Running command Preparing metadata (pyproject.toml)
  6.   Warning: Torch did not find available GPUs on this system.
  7.    If your intention is to cross-compile, this is not an error.
  8.   By default, Apex will cross-compile for Pascal (compute capabilities 6.0, 6.1, 6.2),
  9.   Volta (compute capability 7.0), Turing (compute capability 7.5),
  10.   and, if the CUDA version is >= 11.0, Ampere (compute capability 8.0).
  11.   If you wish to cross-compile for a single specific architecture,
  12.   export TORCH_CUDA_ARCH_LIST="compute capability" before running setup.py.
  14.   torch.__version__  = 2.5.1+cpu
复制代码
4、文章《NVIDIA APEX安装完全指南及Megatron-LM/Pytorch运行题目解决》 
切换到差别的分支：从 GitHub - NVIDIA/apex: A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch 的标签页中，可以知道发布的版本，不至于在众多的分支中无所适从。


 切换到分支：24.04.01-devel
不带参数，表现如今所选分支，git clone的初始分支为master：

  1. git checkout 24.04.01-devel
复制代码

依然报错。
5、参考文章《利用 Visual Studio 2022、CUDA 12.1 和 PyTorch 2.2.2 举行 Windows 编译的最新补丁 #1792》
以下只是想完全照抄原文，利用捏造情况编译，终极因为我是利用的embeddable python，需要搭建太多东西，放弃，以下只是记录尝试过程中的一些知识。
-------------------------------------------------------------------------------------------------------
文章里的路径，与我的情况并不一样，所以利用dir /x 查看短路径，并且只能一级一级查看：

终极，我的路径如下： 
  1. cmd /c 'C:\PROGRA~2\MICROS~3\2022\BUILDT~1\VC\AUXILI~1\Build\VCVARS~1.BAT amd64 & busybox64u sh -ilX'
复制代码
将情况中文件复制到windows文件夹，文件夹需要双 \\ ：


创建的捏造情况，以及git clone 的路径，在“C:\Users\你的用户名” 路径下。
----------------------------------------------------------------------------------------
受文章启发，直接在我的情况中再次安装（执行下面下令生成的whl要比文章头的下令较小，不知道是否因为设置了TORCH_CUDA_ARCH_LIST=7.5只针对我的RTX2080ti）：
  1. git clone https://github.com/NVIDIA/apex.git
  3. cd apex
  5. curl -L "https://github.com/NVIDIA/apex/files/14844602/windows_support.patch" | git apply
  7. set DISTUTILS_USE_SDK=1
  9. set TORCH_CUDA_ARCH_LIST=7.5
  11. pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings "--build-option=--cpp_ext" --config-settings "--build-option=--cuda_ext" .
复制代码
颠末十几分钟，安装乐成（如欲将编译乐成的whl文件生存下来，需要在编译乐成，要举行安装时，按下暂停键Pause，从缓存路径将文件复制，否则，装完后，文件会被删除）。 
乐成后，再次启动comfyUI，没有了开始提到的信息。
编译乐成后，缓存里面有个文件origin.json，有如下内容，不知道与文章中提到的url有关联？：
   {"dir_info": {}, "url": "file:///H:/ComfyUI_cp3.11.9/apex"}
  


免责声明：如果侵犯了您的权益，请联系站长，我们会及时删除侵权内容，谢谢合作！更多信息从访问主页：qidao123.com:ToB企服之家，中国第一个企服评测及商务社交产业平台。
回复

使用道具 举报

返回列表
不到断气不罢休
+ 我要发帖

登录后关闭弹窗

登录参与点评抽奖  加入IT实名职场社区
去登录
快速回复 返回顶部 返回列表