检查py310\Lib\site-packages目录下,是否有这个文件夹torch-1.13.1+cu117.dist-info,如果没有,往往可能是装了torch-2.0.x,将两个torch目录删除,并把exe所在目录的cache.json删除,然后运行runner,让它自己重新安装依赖 10.输出乱码 请更新显卡驱动 11.Torch not compiled with CUDA enabled 和上面第9点一样 最后,一个标准的离线环境目录结构是这样的
Open 我尝试了我所有设备包括 v100/a100/L40S 的设备 ,都无法正常跑通 RWKV-v5 /demo-training-prepare.sh (可能是设备比较老旧) 最接近的一次出现了如下错误: RWKV_MY_TESTING x060 Using /root/.cache/torch_extensions/py310_cu116 as PyTorch extensions root... Detected CUDA files, patching ldflags ...
W0327 01:20:49.642678 3016 init.cc:182] Compiled with WITH_GPU, but no GPU found in runtime. /opt/conda/envs/python35-paddle120-env/lib/python3.9/site-packages/paddle/fluid/framework.py:634: UserWarning: You are using GPU version Paddle, but your CUDA device is not set properly. CPU...
Cuda has to be 12.1 for now because deepspeed is currently compiled by CUDA 12.1. Python environments: accelerate==0.27.2 aiohttp==3.9.3 aiosignal==1.3.1 annotated-types==0.6.0 async-timeout==4.0.3 attrs==21.4.0 cbor2==5.6.2 certifi==2024.2.2 charset-normalizer==3.3.2 deepspeed==0.14...
10.输出乱码 请更新显卡驱动 11.Torch not compiled with CUDA enabled 和上面第9点一样 最后,一个标准的离线环境目录结构是这样的