问题描述

OSError: /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/nvidia/cublas/lib/libcublas.so.11: symbol cublasLtGetStatusString, version libcublasLt.so.11 not defined in file libcublasLt.so.11 with link time reference

问题原因

找了很多解决办法，发现是在安装torch是cuda版本和nvidia版本不匹配或者找不到🤔，好多建议重新 install 某个pytorch版本或者删除cuda，又得用，又懒得卸载…

解决办法

安装了，but 找不到？那就引导他让他能找到

于是，根据报错的具体信息，把lib路径加进来：

# 根据报错信息找到文件路径 
ldd /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/nvidia/cublas/lib/libcublas.so.11 

# 临时添加path
export LD_LIBRARY_PATH=/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/nvidia/cublas/lib/:$LD_LIBRARY_PATH

# 或者在vim ~/.bashrc中添加
export LD_LIBRARY_PATH=/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/nvidia/cublas/lib/:$LD_LIBRARY_PATH