Problem:
error in tensorflow:
Loaded runtime CuDNN library: 5005 (compatibility version 5000) but source was compiled with 5105 (compatibility version 5100). If using a binary install, upgrade your CuDNN library to match.
olution:
you can fix it by updating your cudnn.
details come fromanother blog.
tar -zxvf cudnn.yourname.tgz
cd cuda
sudo cp lib64/* /usr/local/cuda/lib64/
sudo cp include/cudnn.h /usr/local/cuda/include/
设置环境变量,在/etc/profile中添加CUDA环境变量,
sudo gedit /etc/profile
在打开的文件中加入如下两句话
export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
保存后,使环境变量立即生效,
source /etc/profile
进入/usr/local/cuda/samples,执行下面的命令来build samples,
sudo make all -j4
全部编译完成后,进入 samples/bin/x86_64/linux/release,运行deviceQuery,
./deviceQuery
[Note] if you want to download cudnn and cannot find it in official website, you are free to use the following links.
cudnn-8.0-linux-x64-v5.1.tgz
链接: https://pan.baidu.com/s/1bQmvXs 密码: uj85
cudnn-7.5-linux-x64-v5.0-ga.tgz
链接: https://pan.baidu.com/s/1jHRYa6U 密码: n8dd
cudnn-8.0-windows10-x64-v6.0.zip
链接: https://pan.baidu.com/s/1c2dF4g4 密码: 76ci