svn>> dfl>> 返回
项目作者: danilkolikov

项目描述 :
Differentiable Functional Language
高级语言: Haskell
项目地址: git://github.com/danilkolikov/dfl.git
创建时间: 2018-12-19T14:21:22Z
项目社区:https://github.com/danilkolikov/dfl

开源协议:MIT License

下载


GitHub
Build Status

DFL

A compiler of a Differentiable Functional Language.

I am providing code in the repository to you under an open source license.
Because this is my personal repository, the license you receive to my code is
from me and not my employer (Facebook).

What is it about?

The goal of this project is to explore relation between programs and
structures of neural networks.

There clearly is a correspondence between execution of a program, written in
functional paradigm, and flow of data in a neural network. I’ve already applied
this idea to design of sophisticated networks (see the FNN
language), which could learn non-trivial algorithms.

This repository is a next iteration of the research. I’ll try to implement a
compiler of a differentiable functional language with a haskell-like syntax.

How to use it?

Currently the project is under development, but one can check it out, build and
test:

  1. stack build # Build the project
  2. stack test # Run tests
  3. stack haddock # Build documentation
  4. stack run my_file.dfl # Run the compiler
  5. stack run my_file.dfl -- -d # Run the compiler and produce
  6. # debug output for every step of
  7. # compilation

License

This project is licensed under the MIT License - see the LICENSE file for details.