RNN-GC

Introduction

This repo contains a python implementation of RNN-GC.
custom_lstm.py is a re-implementation of LSTM using keras.
rnn_gc.py is a implementation of NUE strategy

Requirements: software

• Python 2.7 or 3.6
• Keras
• theano or tensorflow

Testing Demo

run rnn_gc_demo.py

Results

Citing RNN-GC

If you find RNN-GC useful in your research, please consider citing: Yueming Wang, Kang Lin, Yu Qi, Qi Lian, Shaozhe Feng, Gang Pan, Zhaohui Wu, "Estimating Brain Connectivity with Varying Length Time Lags Using Recurrent Neural Network", IEEE Transactions on Biomedical Engineering, In press.