Skip to content

Latest commit

 

History

History
29 lines (19 loc) · 1.76 KB

README.md

File metadata and controls

29 lines (19 loc) · 1.76 KB

Neuro-FLOPER

A program that allows you to translate neural networks created with Keras to fuzzy logic programs, in order to tune these networks from a given dataset. I created this tool during the development of my Final Year Project, and it wants to solve structural deficiencies created during the training process of a multilayer perceptron. This project has been developed with DEC-Tau research team.

Prerequisites

In order to execute Neuro-FLOPER in your machine, you must have installed:

  • Python 2.7
  • Keras 2.1.4
  • Tensorflow 1.3.0
  • SWI-Prolog 6.0.2 for x86_64-linux
  • PyQt4

DOCUMENTATION

The full document of this project appears also in this repository. It corresponds to the final memory for the FYP "Developing Fuzzy Neural Networks With The FLOPER Environment". Also, there is additional information in the paper published on SISTEDES for PROLE 2019.

A video of the basic walkthrough in Neuro-FLOPER is available on YouTube:

About this work

Please, cite Neuro-FLOPER as:

  • Moreno G., Pérez J., Riaza J.A. (2019) Fuzzy Logic Programming for Tuning Neural Networks. In: Fodor P., Montali M., Calvanese D., Roman D. (eds) Rules and Reasoning. RuleML+RR 2019. Lecture Notes in Computer Science, vol 11784. Springer, Cham