micheleciciolla / tensorflow-binarized-nn

Tensorflow application of a binarized neural network project for Neural Networks course 2018-19

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

A Binarized Neural Networks implementation

Swagger Validator

This is an implementation of binarized neural networks using Tensorflow. Based on https://arxiv.org/abs/1602.02830

Final Project for Neural Network course 2018-2019, UniversitĂ  La Sapienza Roma

Project Status

Completed

Starting the training

It's possible to start training networks on Cifar or MNIST datasets. At the end a plot about performance is printed. One can edit parameters inside each script.

How to do it

python MNISTtrain.py or python cifarTrain.py

Built With

  • [Visual Studio Code]
  • [Tensorflow]
  • [Python]

Authors

  • Michele Ciciolla

  • Flavio Lorenzi

Documentation

You can read our technical documentation about this project here PDF DOCUMENTATION

License

This project is under MIT license - LICENSE

About

Tensorflow application of a binarized neural network project for Neural Networks course 2018-19


Languages

Language:Python 100.0%