Nengo extras

Extra utilities and add-ons for Nengo.

This repository contains utilities that occupy a liminal space not quite generic enough for inclusion in Nengo, but useful enough that they should be publicly accessible.

Some of these utilities may eventually migrate to Nengo core, and others may be split off into their own separate repositories. If something that you were using has been moved, please consult the release history below to see where it has been moved.

Installation

To install Nengo extras, we recommend using pip.

pip install nengo-extras

• Loihi training
  • Neuron output noise models
  • NengoDL builders
• Deep learning
  • MNIST classifier with Keras and Nengo
  • CIFAR-10 classifier with a spiking CNN
  • ImageNet ILSVRC-2012 classifier with a spiking CNN
  • Datasets
  • Keras
  • Networks
  • Processes
• Distributions
• Learning rules
• Networks
  • Utilities
• Neuron types
  • Utilities
• Vision models
  • Encoding for image recognition
  • Camera input
• Visualization tools
  • Visualizing Nengo networks with Gephi
  • Creating spike raster plots
  • GraphViz diagrams
  • Image manipulation
  • Matplotlib plots
  • TkInter GUIs
  • Gephi visualization

• Index

Release history

0.5.0 (January 12, 2022)

Added

• Added NengoDL builders for NengoLoihi neuron types. These will automatically be used by the NengoLoihi repository if it is installed. (#94)

Fixed

• Updated to work with more recent versions of Nengo. (#94)

0.4.0 (November 15, 2019)

Added

• Added nengo_extras.simulators.RealTimeSimulator, which will ensure that simulations don’t run faster than real time. (#85, #151)

• Added nengo_extras.neurons.NumbaLIF, which is a numba-accelerated drop-in replacement for the nengo.LIF neuron model (requires numba to be installed). (#86)

Fixed

• Fixed some Nengo 3.0.0 compatibility issues. (#90)

0.3.0 (June 4, 2018)

Changed

• Submodules are no longer automatically imported into the nengo_extras namespace, as it can be difficult to install requirements for the various tools in Nengo Extras. (#77, #78)

0.2.0 (May 31, 2018)

Added

• Added the association matrix learning rule (AML) to learn associations from cue vectors to target vectors in a one-shot fashion without catastrophic forgetting. (#72)

• Added classes to convert Nengo models to GEXF for visualization with Gephi. (#54)

• Added a Camera process to stream images from a camera to Nengo. (#61)

0.1.0 (March 14, 2018)

Initial release of Nengo Extras! Tested with Nengo 2.7.0, but should work with earlier versions. If you run into any issues, please file a bug report.