我下载了 2.1.1 版本飞浆，然后下载了 11.2.0 的 CUDA 和 8.1.1 的 cudnn
但是系统始终报错，对于cuda的版本有什么要求？
RuntimeError: (PreconditionNotMet) The third-party dynamic library (cublas64_112.dll;cublas64_11.dll) that Paddle depends on is not configured correctly. (error code is 126) Suggestions: 1. Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed. 2. Configure third-party dynamic library environment variables as follows: - Linux: set LD_LIBRARY_PATH by `export LD_LIBRARY_PATH=...` - Windows: set PATH by `set PATH=XXX; (at C:\Paddle_release4\paddle\fluid\platform\dynload\dynamic_loader.cc:265)
我下载了 2.1.1 版本飞浆，然后下载了 11.2.0 的 CUDA 和 8.1.1 的 cudnn 但是系统始终报错，对于cuda的版本有什么要求？ RuntimeError: (PreconditionNotMet) The third-party dynamic library (cublas64_112.dll;cublas64_11.dll) that Paddle depends on is not configured correctly. (error code is 126) Suggestions: 1. Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed. 2. Configure third-party dynamic library environment variables as follows: - Linux: set LD_LIBRARY_PATH by `export LD_LIBRARY_PATH=...` - Windows: set PATH by `set PATH=XXX; (at C:\Paddle_release4\paddle\fluid\platform\dynload\dynamic_loader.cc:265)
看到paddle已经更新支持了昆仑多卡，建议更新文档
顶
顶一下
我在远程服务器上同时布置了两个虚拟环境，一个是paddle2.1 另外一个是Pytorch1.9
后者可以使用CUDA，paddle2.1跑不起来， run_check就报错了， no kernel image is available for execution on the device
感觉还是不好用啊
我下载了 2.1.1 版本飞浆，然后下载了 11.2.0 的 CUDA 和 8.1.1 的 cudnn
但是系统始终报错，对于cuda的版本有什么要求？
RuntimeError: (PreconditionNotMet) The third-party dynamic library (cublas64_112.dll;cublas64_11.dll) that Paddle depends on is not configured correctly. (error code is 126)
Suggestions:
1. Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed.
2. Configure third-party dynamic library environment variables as follows:
- Linux: set LD_LIBRARY_PATH by `export LD_LIBRARY_PATH=...`
- Windows: set PATH by `set PATH=XXX; (at C:\Paddle_release4\paddle\fluid\platform\dynload\dynamic_loader.cc:265)
Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed.
检查以下CUDA的版本号
顶
通过官方python -m pip install paddlepaddle-gpu==0.0.0.post112 -f https://www.paddlepaddle.org.cn/whl/windows/gpu/develop.html安装，普通默认安装匹配的cuda是10.2