Installing cuDNN on Lenovo-Y50 Ubuntu 16.04 - felt this is kind of tricky, it might help some

instalcudnn.txt

# I had so many troubles getting CUDA 8 on my Lenovo-Y50 Ubuntu 16.04, so here you go, hope this helps a lot of 

sudo apt-get purge libvdpau-va-gl1 bumblebee* nvidia*
# Driver
sudo apt-get install nvidia-384 nvidia-settings nvidia-prime

# Do a reboot here
sudo reboot

# Downloads the respective driver
cd ~/Downloads
wget https://developer.nvidia.com/compute/cuda/8.0/Prod2/local_installers/cuda_8.0.61_375.26_linux-run
chmod +x cuda_8.0.61_375.26_linux.run

# Run the driver install
cd ~/Downloads
sudo bash cuda_8.0.61_375.26_linux.run --no-opengl-libs
# Control+C skip eula to end 
# accept
# no for the accelerated driver
# yes for toolkit, symbolic link. and yes for samples

# Download cuDNN v6 from the website
# https://developer.nvidia.com/cudnn
# Extract cudnn-8.0-linux-x64-v5.1.tgz you will get a folder "cuda"
# Make sure to download the right
cd ~/Downloads/cuda
sudo cp cuda/include/cudnn.h /usr/local/cuda/include
sudo cp cuda/lib64/* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/lib64/libcudnn*

# Install tensorflow by activating your virtual environment and installing in gpu-enabled mode

# Check tensorflow installation as given on tensorflow installation page
# If import tensorflow throws an error for which indicates a libcudnn.so.7 or something like that, try to downgrade or upgrade cuDNN

For those who struggles with verifying cuDNN v7 64-bit installation:

• If you receive a message like: ./mnistCUDNN: error while loading shared libraries: libcudart.so.9.0: cannot open shared object file: No such file or directory
• Check what your CUDA installation folder is in /usr/local directory, it can be cuda or cuda-9.0. You will set the environmental variables accordingly.
• Check your PATH variable ($ echo $PATH) and LD_LIBRARY_PATH variable ($ echo $LD_LIBRARY_PATH) and make sure both of it contains /usr/local/cuda/lib64 or /usr/local/cuda-9.0/lib64
• Check the directories and make sure the files copied from the .tgz file are in there.

In case the you are missing the required PATHs use these command below in terminal to add them:

• $ export PATH=/usr/local/cuda/bin${PATH:+:${PATH}} or $ export PATH=/usr/local/cuda-9.0/bin${PATH:+:${PATH}}

• $ export LD_LIBRARY_PATH=/usr/local/cuda/lib64\${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}} or $ export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64\ ${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}