subprocess.CalledProcessError: 命令 '['ninja', '-v']' 返回非零退出状态 1

Question

帮助！

当我在 2080ti 中运行我的“semanticstylegan”代码时，一切正常。但是当我切换到 RTX 3090 时，出现了问题！！程序无法运行。

环境是：

RTX 3090，python：3.7 pytorch：1.7.1+cu110，ninja：1.10.0

当我使用命令：ninja --version 它输出 1.10.0 时，我认为 ninja 安装得很好。

2080ti Gpu 与 pytorch 1.12.1+cu102，ninja 版本 1.7.1，python 3.8

详情在这里：

python scripts/train.py --dataset_type=fashiondata   --exp_dir=/home/user/wz/jin/local_encoder_results --workers=2 --batch_size=2 --test_batch_size=1 --test_workers=2 --val_interval=5000 --save_interval=5000  --start_from_latent_avg
---
/home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/__init__.py
Traceback (most recent call last):
  File "/home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1539, in _run_ninja_build
    env=env)
  File "/home/user/anaconda3/envs/unleashing/lib/python3.7/subprocess.py", line 512, in run
    output=stdout, stderr=stderr)
subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.
...
...

File "/home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 997, in load
    keep_intermediates=keep_intermediates)
  File "/home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1202, in _jit_compile
    with_cuda=with_cuda)
  File "/home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1300, in _write_ninja_file_and_build_library
    error_prefix="Error building extension '{}'".format(name))
  File "/home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1555, in _run_ninja_build
    raise RuntimeError(message) from e
RuntimeError: Error building extension 'fused': ninja: warning: bad deps log signature or version; starting over

[1/3] /usr/bin/nvcc -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/TH -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/THC -isystem /home/user/anaconda3/envs/unleashing/include/python3.7m -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -std=c++14 -c /home/user/wz/jin/localEncoder/models/op/fused_bias_act_kernel.cu -o fused_bias_act_kernel.cuda.o
FAILED: fused_bias_act_kernel.cuda.o
/usr/bin/nvcc -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/TH -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/THC -isystem /home/user/anaconda3/envs/unleashing/include/python3.7m -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr -gencode=arch=compute_86,code=sm_86 --compiler-options '-fPIC' -std=c++14 -c /home/user/wz/jin/localEncoder/models/op/fused_bias_act_kernel.cu -o fused_bias_act_kernel.cuda.o
nvcc fatal   : Unsupported gpu architecture 'compute_86'


[2/3] c++ -MMD -MF fused_bias_act.o.d -DTORCH_EXTENSION_NAME=fused -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/TH -isystem /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/THC -isystem /home/user/anaconda3/envs/unleashing/include/python3.7m -D_GLIBCXX_USE_CXX11_ABI=0 -fPIC -std=c++14 -c /home/user/wz/jin/localEncoder/models/op/fused_bias_act.cpp -o fused_bias_act.o
In file included from /home/user/anaconda3/envs/unleashing/lib/python3.7/site-packages/torch/include/c10/core/DeviceType.h:8,

...

我尝试将 ninja 版本更改为 1.7.2，但它不起作用！