- 报错信息
- 直接看报错内容
W0930 01:29:17.811947 792 device_context.cc:404] Please NOTE: device: 0, GPU Compute Capability: 6.1, Driver API Version: 10.2, Runtime API Version: 10.1
W0930 01:29:17.815150 792 dynamic_loader.cc:238] Note: [Recommend] copy cudnn into CUDA installation directory.
For instance, download cudnn-10.0-windows10-x64-v7.6.5.32.zip from NVIDIA’s official website,
then, unzip it and copy it into C:Program FilesNVIDIA GPU Computing ToolkitCUDAv10.0
You should do this according to your CUDA installation directory and CUDNN version.
Traceback (most recent call last):
File “D:/zd_code/学习/测试.py”, line 2, in
paddle.utils.run_check()
File “D:python36libsite-packagespaddleutilsinstall_check.py”, line 196, in run_check
_run_static_single(use_cuda)
File “D:python36libsite-packagespaddleutilsinstall_check.py”, line 124, in _run_static_single
exe.run(startup_prog)
File “D:python36libsite-packagespaddlefluidexecutor.py”, line 1110, in run
six.reraise(*sys.exc_info())
File “D:python36libsite-packagessix.py”, line 703, in reraise
raise value
File “D:python36libsite-packagespaddlefluidexecutor.py”, line 1108, in run
return_merged=return_merged)
File “D:python36libsite-packagespaddlefluidexecutor.py”, line 1239, in _run_impl
use_program_cache=use_program_cache)
File “D:python36libsite-packagespaddlefluidexecutor.py”, line 1329, in _run_program
[fetch_var_name])
RuntimeError: (PreconditionNotMet) The third-party dynamic library (cudnn64_7.dll) that Paddle depends on is not configured correctly. (error code is 126)
Suggestions:
- Check if the third-party dynamic library (e.g. CUDA, CUDNN) is installed correctly and its version is matched with paddlepaddle you installed.
- 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:homeworkspacePaddle_releasepaddlefluidplatformdynloaddynamic_loader.cc:265)
你好！ 这是你第一次使用 Markdown编辑器 所展示的欢迎页。如果你想学习如何使用Markdown编辑器, 可以仔细阅读这篇文章，了解一下Markdown的基本语法知识。
1.已经安装cuda版本10.2，paddlepaddle安装的对应cuda10.2版本，（看到这里没错啊！）画风一转，提示：”你需要安装对应版本的cuda和cudann“。看到这里我就懵了，我已经安装了显卡驱动版本对应的cuda了，如下图：
找到了cuda官网 https://developer.nvidia.com/cuda-toolkit-archive 下载了10.2，然后又到cudnn下载地址 https://developer.nvidia.com/rdp/cudnn-archive 下载了与cuda10.2对应的cudann了.
为什么还会报错呢？
然后就继续往下看，发现了报错建议之前的信息：
“RuntimeError: (PreconditionNotMet) The third-party dynamic library (cudnn64_7.dll) that Paddle depends on is not configured correctly. (error code is 126)”
找不到(cudnn64_7.dll）这个文件，在网上搜索才了解这是cudnn里面的文件，那么问题又回到了安装cudnn那里，在cudann那个下载界面存在着许多的与cuda10.2对应的cudnn的版本，本来想着版本更新时间越靠后功能越全，结果想错了：需要下载与paddlepaddle对应的cudnn（吐槽一句，在网上搜索的教程里都没有写），后来根据(cudnn64_7.dll）下载了7.8的版本，解压之后找到了(cudnn64_7.dll文件。再次运行pathon输入import paddle ，再输入 paddle.utils.run_check()，显示"PaddlePaddle is installed successfully!"可用。
引用：
[1]: https://www.cnblogs.com/lqylucky/p/12808844.html
[2]:https://www.paddlepaddle.org.cn/install/quick?docurl=/documentation/docs/zh/install/pip/windows-pip.html
[3]:https://blog.csdn.net/kieven2008/article/details/120410650
[4]:https://blog.csdn.net/qq_29007291/article/details/110878764
