windows操作系统 在程序中import jittor后会出现如下报错
(pytorch) D:\dx\ANN\Final\PPLM-Jittor>python -u "c:\Users\LEGION\Desktop\xj\test.py"
[i 1208 23:14:28.991000 84 compiler.py:955] Jittor(1.3.6.2) src: e:\anaconda3\envs\pytorch\lib\site-packages\jittor
[i 1208 23:14:29.025000 84 compiler.py:956] cl at C:\Users\LEGION\.cache\jittor\msvc\VC\_\_\_\_\_\bin\cl.exe(19.29.30133)
[i 1208 23:14:29.026000 84 compiler.py:957] cache_path: C:\Users\LEGION\.cache\jittor\jt1.3.6\cl\py3.10.6\Windows-10-10.xd0\AMDRyzen75800Hxb1\default
[i 1208 23:14:29.028000 84 install_cuda.py:93] cuda_driver_version: [11, 6, 0]
[i 1208 23:14:29.057000 84 __init__.py:411] Found C:\Users\LEGION\.cache\jittor\jtcuda\cuda11.2_cudnn8_win\bin\nvcc.exe(11.2.67) at C:\Users\LEGION\.cache\jittor\jtcuda\cuda11.2_cudnn8_win\bin\nvcc.exe.
[i 1208 23:14:29.162000 84 __init__.py:411] Found gdb(7.8.1) at E:\Dev-Cpp\MinGW64\bin\gdb.EXE.
[i 1208 23:14:29.234000 84 __init__.py:411] Found addr2line(2.24.51) at E:\Dev-Cpp\MinGW64\bin\addr2line.EXE.
[i 1208 23:14:29.280000 84 compiler.py:1010] cuda key:cu11.2.67
[i 1208 23:14:29.282000 84 __init__.py:227] Total mem: 15.87GB, using 5 procs for compiling.
[i 1208 23:14:30.190000 84 jit_compiler.cc:28] Load cc_path: C:\Users\LEGION\.cache\jittor\msvc\VC\_\_\_\_\_\bin\cl.exe
[i 1208 23:14:30.191000 84 init.cc:62] Found cuda archs: [86,]
[i 1208 23:14:30.271000 84 compile_extern.py:519] mpicc not found, distribution disabled.
Traceback (most recent call last):
File "E:\Anaconda3\envs\pytorch\lib\site-packages\jittor\compile_extern.py", line 232, in setup_cuda_extern
setup_cuda_lib(lib_name, extra_flags=link_cuda_extern)
File "E:\Anaconda3\envs\pytorch\lib\site-packages\jittor\compile_extern.py", line 291, in setup_cuda_lib
ctypes.CDLL(ex_cudnn_path, dlopen_flags)
File "E:\Anaconda3\envs\pytorch\lib\ctypes\__init__.py", line 374, in __init__
self._handle = _dlopen(self._name, mode)
OSError: [WinError 127] 找不到指定的程序。
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "c:\Users\LEGION\Desktop\xj\test.py", line 2, in <module>
import jittor as jt
File "E:\Anaconda3\envs\pytorch\lib\site-packages\jittor\__init__.py", line 25, in <module>
from . import compile_extern
File "E:\Anaconda3\envs\pytorch\lib\site-packages\jittor\compile_extern.py", line 593, in <module>
setup_cuda_extern()
File "E:\Anaconda3\envs\pytorch\lib\site-packages\jittor\compile_extern.py", line 244, in setup_cuda_extern
LOG.f(msg)
File "E:\Anaconda3\envs\pytorch\lib\site-packages\jittor_utils\__init__.py", line 104, in f
def f(self, *msg): self._log('f', 0, *msg)
File "E:\Anaconda3\envs\pytorch\lib\site-packages\jittor_utils\__init__.py", line 89, in _log
cc.log(fileline, level, verbose, msg)
RuntimeError: [f 1208 23:14:31.479000 84 compile_extern.py:244] CUDA found but cudnn is not loaded:
Develop version of CUDNN not found,
please refer to CUDA offical tar file installation:
https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html#installlinux-tar
or you can let jittor install cuda and cudnn for you:
>>> python3.10 -m jittor_utils.install_cuda
运行底部指令会提示已经安装
[i 1208 23:21:09.100990 20 install_cuda.py:93] cuda_driver_version: [11, 6, 0]
[i 1208 23:21:09.101990 20 install_cuda.py:162] nvcc is installed at C:\Users\LEGION\.cache\jittor\jtcuda\cuda11.2_cudnn8_win\bin\nvcc.exe
或许是没有找到路径？但是没有找到在哪里可以设置