Tensorflow docker

docker-compose.yml

# run tensorflow from dockers -  play-with-docker.com

# to run : http://play-with-docker.com/?stack=https://gist.githubusercontent.com/azizasm/2e8ba3c81a3ba1811f17d48275d6d2b8/raw/docker-compose.yml


# Use root/example as user/password credentials
version: '3.1'
services:
  tensorflow:
    restart: always
    image: tensorflow/tensorflow:nightly-py3
#    container_name : zys
    ports:
     - 8888:8888
#    environment:
#      MYSQL_ROOT_PASSWORD: root
volumes:
  logvolume01: {}
  sitevolume: {}

Run Tensorflow in play-with-docker (PWD)

1. open link below http://play-with-docker.com/?stack=https://gist.githubusercontent.com/azizasm/2e8ba3c81a3ba1811f17d48275d6d2b8/raw/docker-compose.yml

and wait until the container to load.

2. Click the 8888 in the PWD session. It will prompt for session pass that can be obtained from :

ssh ip172-18-0-28-xxx@direct.labs.play-with-docker.com

docker logs `docker container ls -q`

Output :

[C 08:22:53.083 NotebookApp]

    Copy/paste this URL into your browser when you connect for the first time,
    to login with a token:
        http://localhost:8888/?token=6bb1913fb7a04e95046f3599c9c395d58c8c876de6ea13d3

3. Open up the URL and you may start with python notebook

Note: if no 8888 Link appear in the session. You may enter manually :

http://{ssh-id}-8888.direct.labs.play-with-docker.com/tree?token={ssh}

example :
http://ip172-18-0-9-b9ql6sqr91k000fbs790-8888.direct.labs.play-with-docker.com/tree?token=c02e206041a43bdab6df846c93d2a4a047efb3891cd3cf07

Play with Sarasra/models

git clone https://github.com/Sarasra/models.git

docker cp models/. `docker container ls -q`:/notebooks
docker exec -t -i `docker container ls -q` /bin/bash

http://ip172-18-0-8-b9rb11ar91k000e68upg-8888.direct.labs.play-with-docker.com/notebooks/samples/outreach/blogs/housing_prices.ipynb
-8888.direct.labs.play-with-docker.com/tree/samples/outreach/blogs/housing_prices.ipynb