forked from https://github.com/ray-project/ray

merged with https://github.com/coco66/RL/tree/tmp0408

To build docker image and RLlib container

Will build ray on top of tensorflow 1.14.0-gpu-jupyter image.

~/ray$ ./build-docker.sh

Image is now built and RLlib container is now running. You can now open a jupyter notebook in browser by copying link and appending to.

localhost:8888 ...

To use terminal

$ docker exec -it RLlib bash

Drops you into the rllib directory.

To close container

$ docker ps -a
$ docker container stop RLlib
$ docker rm ...