Nvidia setup for DL-RIG

tutorial:
1. https://medium.com/@vivek.yadav/deep-learning-setup-for-ubuntu-16-04-tensorflow-1-2-keras-opencv3-python3-cuda8-and-cudnn5-1-324438dd46f0
2. https://towardsdatascience.com/build-and-setup-your-own-deep-learning-server-from-scratch-e771dacaa252 [for opencv only]

** installing tensorflow-gpu follow above tutorial
** installing pytorch-gpu
# install torch (cuda 9)
$ conda install pytorch torchvision cuda90 -c pytorch
# test gpu install
python -c 'import torch; print(torch.rand(2,3).cuda())'

important note:
# Some Deeplearning frameworks are not yet ready for CUDA 9.1. Only install CUDA 9.0
1. Download CUDA Toolkit 9.0 (link: https://developer.nvidia.com/cuda-90-download-archive)
2. Download cuDNN v7.1.3 (April 17, 2018), for CUDA 9.0 [download cudnn-9.0-linux-x64-v7.1.tgz file and follow the tutorial]

https://launchpad.net/~graphics-drivers/+archive/ubuntu/ppa
current official release: nvidia-390

3. why use conda ? : https://stackoverflow.com/questions/20994716/what-is-the-difference-between-pip-and-conda?utm_medium=organic&utm_source=google_rich_qa&utm_campaign=google_rich_qa
4. conda commands: https://conda.io/docs/commands/conda-install.html