Mraksu / Winning-Tickets-in-Smaller-Neural-Networks

Experiments about Lottery Ticket Hypothesis

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Implementation of the Master Thesis: Identifying Winning Lottery Tickets in Small Neural Network Architectures by Murat Aksu

Experiments notebook can be used to reproduce the results (using the same random seed).

Also you can modify parameters and make your own experiments as well.

In figures notebook you can find the code the produce the figures in the paper.

Requirements

  • Python 3.7
  • Tensorflow 2.1
  • Keras 2.3
  • Tensorflow Model Optimization API

Acknowledgement

I wish to thank to Arjun Majumdar, his work was inspration for me.

The Lottery Ticket Hypothesis by Jonathan Frankle, Michael Carbin: Implemented by Arjun Majumdar

GitHub: https://github.com/arjun-majumdar/Lottery_Ticket_Hypothesis-TensorFlow_2

About

Experiments about Lottery Ticket Hypothesis


Languages

Language:Jupyter Notebook 97.5%Language:Python 2.5%