[Deep Learning] Loaded runtime CuDNN library

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