tensorflow_2.3.1_cuda_10.1_installation_on_Ubuntu_18.04

# Deeplearning Environment Setup for tensorflow_2.3.1 with CUDA 10.1 and cuDNN 7.6.0

### If you have previous installation remove it first. 
sudo apt-get purge nvidia*
sudo apt remove nvidia-*
sudo apt remove --autoremove nvidia-cuda-toolkit
sudo rm /etc/apt/sources.list.d/cuda*
sudo apt-get autoremove && sudo apt-get autoclean
sudo rm -rf /usr/local/cuda*


# Get latest nvidia driver
apt-cache search nvidia-driver
sudo apt install nvidia-driver-470

# check nvidia driver version
nvidia-smi


# install cuda 10.1 method-1
# https://developer.nvidia.com/cuda-10.1-download-archive-base?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1804&target_type=runfilelocal
wget https://developer.nvidia.com/compute/cuda/10.1/Prod/local_installers/cuda_10.1.105_418.39_linux.run
# run below command and follow the command-line prompts
sudo sh cuda_10.1.105_418.39_linux.run 

# install cuda 10.1 method-2
# https://developer.nvidia.com/cuda-10.1-download-archive-base?target_os=Linux&target_arch=x86_64&target_distro=Ubuntu&target_version=1804&target_type=deblocal
wget https://developer.nvidia.com/compute/cuda/10.1/Prod/local_installers/cuda-repo-ubuntu1804-10-1-local-10.1.105-418.39_1.0-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu1804-10-1-local-10.1.105-418.39_1.0-1_amd64.deb
sudo apt-key add /var/cuda-repo-10-1-local-10.1.105-418.39/7fa2af80.pub
sudo apt-get update
sudo apt-get install cuda


# install cuDNN(CUDA® Deep Neural Network library) 7.6.0 version (CUDA 10.1 is compatable with cuDNN 7.5.1 - 7.6.2)
# source https://anaconda.org/anaconda/cudnn/files
wget https://anaconda.org/anaconda/cudnn/7.6.0/download/linux-64/cudnn-7.6.0-cuda10.1_0.tar.bz2
tar -xvf cudnn-7.6.0-cuda10.1_0.tar.bz2 -C cuda
sudo cp cuda/include/cudnn.h /usr/local/cuda/include
sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*
echo 'export PATH=/usr/local/cuda-10.1/bin${PATH:+:${PATH}}' >> ~/.bashrc
echo 'export LD_LIBRARY_PATH=/usr/local/cuda-10.1/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}' >> ~/.bashrc
echo 'export CUDA_HOME=/usr/local/cuda' >> ~/.bashrc
source ~/.bashrc

# test cuDNN
cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2
#Should see something like below:
#define CUDNN_MAJOR      6
#define CUDNN_MINOR      0
#define CUDNN_PATCHLEVEL 21
--
#define CUDNN_VERSION    (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)
#include "driver_types.h"

# check driver with cuda
lspci | grep -i nvidia

# install tensorflow 2.3.1 and pytorch 1.7.1
pip install tensorflow==2.3.1
pip install torch==1.7.1+cu101 torchvision==0.8.2+cu101 torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch_stable.html

# Test tensorflow_2
import tensorflow as tf
print(tf.config.list_physical_devices('GPU'))

# install nvidia-docker
# first install docker-ce
sudo apt-get update
sudo apt-get install -y nvidia-docker2
sudo pkill -SIGHUP dockerd

# Test environment and to make sure everything is installed correctly
sudo docker run --runtime=nvidia --rm nvidia/cuda:10.1-base nvidia-smi