/tensorflow-formula

ML stack environment on docker-compose which includes: Tensorflow + Python and other useful tools

Primary LanguageShell

Build Status Docker Stars Docker Pulls Docker Automated Docker Build Gitter chat

Tensorflow-Formula

ML stack environment on docker-compose which includes: Tensorflow + Python 3.7 and other useful tools

Stack includes

  • Python 3.7
  • TensorFlow
  • Jupyter Notebook
  • Portainer
  • Nginx as sidecar

Extra python libs are included.

Note

Before installing this project, please, make sure you have installed docker and docker-compose

To install docker execute:

$ curl -fsSL https://get.docker.com -o get-docker.sh
$ sh get-docker.sh
$ pip install docker-compose

Installation

Clone this project into your work directory:

$ git clone "https://github.com/trydirect/tensorflow-formula.git"

Then build it with the following command:

$ cd tensorflow-formula
$ ./setup.sh

Once the docker images are built/pulled, the stack will be deployed and a Jupyter token will be printed out.

$ ./setup.sh
Creating network "user_default" with the default driver
Creating volume "user_jupyter-notebooks" with local driver
Creating volume "user_portainer-data" with local driver
Creating user_jupyter-tensorflow_1 ... done
Creating user_portainer_1          ... done
Jupyter token: e7c7bb2956c899e7cce6fbd5587108ef701c98ca5ab0ac84

Default ports:

  • Jypter Notebook: 8888
  • Portainer: 9000
  • Applications are also accessible using the following endpoints:
    • Jupyter Notebook - ip_address/
    • Portainer - ip_address/portainer/

Features

  • Full Docker integration
  • Docker Compose integration and optimization for local development

The final project structure will look like this:

.
├── README.md
├── cleanup.sh
├── setup.sh
├── start.sh
├── stop.sh
└── v01
    └── dockerfiles
        ├── build
        │   └── app
        │       └── Dockerfile
        ├── configs
        │   └── nginx
        │       └── default.conf
        └── docker-compose.yml

6 directories, 8 files

              Name                            Command               State           Ports         
--------------------------------------------------------------------------------------------------
dockerfiles_jupyter-tensorflow_1   jupyter notebook --port=88 ...   Up      0.0.0.0:8888->8888/tcp
dockerfiles_nginx_1                nginx -g daemon off;             Up      0.0.0.0:80->80/tcp    
dockerfiles_portainer_1            /portainer                       Up      0.0.0.0:9000->9000/tcp

Start, Stop and Clean-Up

For stopping a running stack without deleting its resources - use:

$ ./stop.sh

For starting an existing stack - use:

$ ./start.sh

For removing all the containers and volumes - use:

$ ./cleanup.sh

Quick deployment to cloud

Amazon AWS, Digital Ocean, Hetzner and others

[] TODO

Contributing

  1. Fork it (https://github.com/trydirect/tensorflow-formula/fork)
  2. Create your feature branch (git checkout -b feature/fooBar)
  3. Commit your changes (git commit -am 'Add some fooBar')
  4. Push to the branch (git push origin feature/fooBar)
  5. Create a new Pull Request

Support Development

Donate