fragile 0.0.29

Framework for developing FractalAI based algorithms.

Fragile

Fragile is a framework for developing algorithms inspired by the Fractal AI theory and testing them at scale.

Features

• Provides classes and an API for easily developing planning algorithms
• Provides an classes and an API for function optimization
• Build in visuzalizations of the sampling process
• Fully documented and tested
• Support for parallelization and distributed search processes

About FractalAI

FractalAI is based on the framework of non-equilibrium thermodynamics, and It allows to derive new mathematical tools for efficiently exploring state spaces.

The principles of our work are accessible online:

• Arxiv manuscript describing the fundamental principles of our work.
• Blog that describes our early research process.
• Youtube channel with videos showing how different prototypes work.
• GitHub repository containing a prototype that solves most Atari games.

Getting started

Check out the Getting started with Atari games section of the docs, or check out the examples folder.

Running in docker

The docker container will execute a Jupyter notebook accessible on port 8080 with password: fragile

   make docker build
   docker run -d -p 8080:8080 -v PATH_TO_REPO/fragile fragile 

You can also run the tests inside the docker container

    make docker-test

Installation

This framework has been tested in Ubuntu 18.04 and supports Python 3.6, 3.7 and 3.8. If you find any problems running it in a different OS or Python version please open an issue.

It can be install with pip install fragile.

Building from source

Please take a look at the Dockerfile to find out about all the dependencies, and the detailed installation process.

   git clone https://github.com/FragileTech/fragile.git
   cd fragile
   pip3 install -r requirements.txt
   pip3 install -r requirements-viz.txt
   pip3 install -e .

Documentation

You can access the documentation on GitHub Pages.

• Building the documentation:

  cd fragile/docs
  make html

• Accessing the documentation locally:
  • Launch an http server:

    cd build/html # assuming you are inside fragile/docs
    python3 -m http.server      
  

  • Visit http://0.0.0.0:8000 to display the documentation.

Roadmap

Upcoming features: (Not necessary in order)

• Add support for saving visualizations.
• Fix documentation and add examples for the distributed module
• Upload Montezuma solver
• Add new algorithms to sample different state spaces.
• Add a module to generate data for training deep learning models
• Add a benchmarking module
• Add deep learning API

Contributing

Contribution are welcome. Please take a look at contributining and respect the code of conduct.

Cite us

If you use this framework in your research please cite us as:

@misc{1803.05049,
    Author = {Sergio Hernández Cerezo and Guillem Duran Ballester},
    Title = {Fractal AI: A fragile theory of intelligence},
    Year = {2018},
    Eprint = {arXiv:1803.05049},
  }

License

This project is currently licensed under AGPLv3.0.

However, if you are considering using it for applications that require a more permissive license, please let me know in this Issue