Nengo 3.1.0 released

General Discussion Announcements
1

The Nengo team is happy to announce the release of Nengo 3.1.0!

What is Nengo?

Nengo is a Python library for building and simulating large-scale neural models for AI and robotics. It can be thought of as a neural compiler, transforming a functional description of a neural model to a network of spiking or non-spiking neurons that can run on multiple backends including GPUs and neuromorphic hardware.

What’s new?

Nengo 3.1.0 is a minor release that is compatible with Python 3.5 - 3.8.

We have added several new objects that are useful in many types of models.

  • nengo.RegularSpiking, nengo.StochasticSpiking, and nengo.PoissonSpiking neuron types use the firing rates from a non-spiking neuron and emit spikes at regular intervals, stochastically rounded intervals, or with Poisson statistics, respectively.
  • The nengo.RLS learning rule is a drop-in replacement for the nengo.PES error-modulated learning rule that gives better performance at the cost of biological plausibility.
  • The ScatteredHypersphere distribution is a drop-in replacement for the UniformHypersphere, but tiles the n-dimensional space better while still having some randomness.
  • The NoTransform transform transmits signals between two objects without any transformation; previously, it was common (and the default) to multiply signals by 1 unnecessary.

We use these new objects as defaults in several places. Ensemble eval_points and encoders are sampled from the ScatteredHypersphere, and the default connection transform is NoTransform.

Along with adding the new spiking neuron types, we also made some change to the overall interface of NeuronType. While these changes are generally backwards compatible, we recommend that anyone who has made a custom NeuronType look at the new interface and update their classes accordingly.

In total, Nengo 3.1.0 contains 10 bugfixes and over 20 new and changed features. To see the full list of changes, head to the Github release page.

How do I get it?

To get the new version of Nengo, use pip.

pip install --upgrade nengo

Where can I learn more?

Where can I get help?

You’re already there! If you have an issue upgrading or have any other questions, please post them in this forum.

5 Likes