Closed hopeforus closed 1 year ago
看看版本啥的有没有装对,用nvidia-smi或者echo $CUDA_HOME可以看看现在用的cuda版本是啥,感觉类似的问题网上解决方法很多,可以参考一下
谢谢,我重新安装了 cuda 12,cudnn 前面不报错了,新问题来了,在找原因,呵呵
(vicuna) hope@hope-08:~/work/Chinese-Vicuna/scripts$ bash generate.sh
Traceback (most recent call last):
File "/home/hope/work/Chinese-Vicuna/generate.py", line 3, in
没有的库用pip或者conda装不就行了?
(vicuna) hope@hope-08:~/work/Chinese-Vicuna$ python generate.py Traceback (most recent call last): File "generate.py", line 2, in
import torch
File "/home/hope/miniconda3/envs/vicuna/lib/python3.8/site-packages/torch/init.py", line 217, in
_load_global_deps()
File "/home/hope/miniconda3/envs/vicuna/lib/python3.8/site-packages/torch/init.py", line 177, in _load_global_deps
raise err
File "/home/hope/miniconda3/envs/vicuna/lib/python3.8/site-packages/torch/init.py", line 172, in _load_global_deps
ctypes.CDLL(lib_path, mode=ctypes.RTLD_GLOBAL)
File "/home/hope/miniconda3/envs/vicuna/lib/python3.8/ctypes/init.py", line 373, in init
self._handle = _dlopen(self._name, mode)
OSError: libcudnn.so.8: cannot open shared object file: No such file or directory