Skip to content

HBPNeurorobotics/auryn

 
 

Repository files navigation

Auryn

Auryn is Simulator for recurrent spiking neural networks with synaptic plasticity. It comes with the GPLv3 (please see COPYING).

This is a fork of Auryn implementing event-driven random backpropagation for spike-based deep learning.

Quick start

Note, Auryn needs a C++ compiler, the boost libraries (www.boost.org) with MPI support installed. To download and compile the examples under Ubuntu Linux try:

sudo apt-get install cmake git build-essential libboost-all-dev
git clone https://ids-git.fzi.de/friedric/erbp.git
cd erbp
pip install -r requirements.txt
pip install -e .
cd build/release
./bootstrap.sh
make -j8

Run eRBP on event-based dataset

If you use this code in your paper, please add the following citation:

@article{kaiser2020embodied,
  title={Embodied Neuromorphic Vision with Event-Driven Random Backpropagation},
  author={Kaiser, Jacques and Friedrich, Alexander and Tieck, J and Reichard, Daniel and Roennau, Arne and Neftci, Emre and Dillmann, R{\"u}diger},
  year={2020}
}

Download Data Sets:

cd build/release/experiments/

# run with covert attention model (re-indexing events with respect to median in time window):
python2 run_classification_gesture_dual_vis_attention.py

# run with resizing the whole frame:
python2 run_classification_gesture_dual_vis.py

Deprecated Data Sets:

cd build/release/experiments/
python2 run_classification_mnist_saccade.py
cd build/release/experiments/
python2 run_classification_mnist_flash.py

Most important files

  • Read aedat files: build/release/experiments/utils/jaer_data_handler.py
  • Generate ras: build/release/experiments/utils/generate_ras.py
  • Run experiment: build/release/experiments/run_classification_*.py
  • Evaluate experiment: build/release/experiments/experimentLib.py
  • Plotting: build/release/experiments/utils/erbp_plotter.py
  • Experiment network: experiments/exp_rbp_flash.cpp

License & Copyright (eRBP DVS)

Copyright 2017-2020 FZI

eRBP for DVS processing scripts under Auryn is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

eRBP DVS is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with Auryn. If not, see http://www.gnu.org/licenses/.

License & Copyright (eRBP)

Copyright 2016-2017 Emre Neftci

eRBP scripts under Auryn is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

eRBP is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with Auryn. If not, see http://www.gnu.org/licenses/.

License & Copyright (Auryn)

Copyright 2014-2017 Friedemann Zenke

Auryn is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

Auryn is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with Auryn. If not, see http://www.gnu.org/licenses/.

About

Auryn Plastic Recurrent Network Simulator

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • C++ 67.1%
  • Python 29.7%
  • Makefile 1.9%
  • CMake 0.8%
  • Shell 0.4%
  • Gnuplot 0.1%