What is TensorFlow
TensorFlow is a foundation library based on Python that can be used for machine learning. In its turn, TensorFlow Stack is almost everything you need to start experimenting with deep learning models creation directly or by using wrapper libraries. It allows you to deploy the TensorFlow development environment to your favorite cloud server or VPS, selected from the supported cloud list. All the services are preconfigured and easy to tune.
What does TensorFlow stack include
A basic TensorFlow stack environment built using Docker Compose. It consists of the following:
- Python/TensorFlow
- Jupyter Notebook
Who is using TensorFlow
- Data scientists
- Software developers
What are the benefits of using dockerized TensorFlow stack
- The fastest way to start experimenting with machine learning on your server
- Out of the box solution with GPU support
- Facilitated upgrade to a new version provided by Docker containers
- Effortless app migration to another cloud provider
- Easily portable docker containers and effortless container management
- CI / CD processes and SSL simplified setup
- Fast startup, better performance of the app and lower system overhead
- Secure servers updates
Number of servers required: 1 VPS
Companion apps
Fail2ban scans log files and bans IPs that show the malicious signs. Highly recommended for simple VPS server setup
Docker Status Panel - is a minimalistic docker container management panel written in Python / Flask microframework.
Portainer is a lightweight management UI that allows you to easily manage your different Docker environments (Docker hosts or Swarm clusters)