Nengo 2.8.0 released


The Nengo team is elated to announce the release of Nengo 2.8.0!

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 2.8.0 contains a few bugfixes and changes to learning rules.

Learning rules that can received inputs (usually error signals) can now be slices, like other Nengo objects you can connect to.

Learning rules that used to take in tau values denoting the decay time constant of a lowpass filter have been changed to instead take in a Synapse instance. The tau parameters still exist for backwards compatibility, but are deprecated and will be removed in Nengo 3.0.

Also deprecated is the nengo.ipynb IPython notebook extension, which was previously required for rich progress bars in the notebook environment. We now use the IPython rich display system to render progress bars, which will work automatically without the nengo.ipynb extension.

To see the full list of changes in Nengo 2.8.0, 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.