Nengo 2.3 released

The Nengo team is elated to announce the release of Nengo 2.3!

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 2.3 features several improvements under the hood which should speed up certain operations, such as accessing probe data via sim.data.

One nice new feature is the ability to copy and pickle Nengo objects. Nengo objects like Ensemble now have a copy method that will clone that ensemble in the current network context. If you created a model in a console session, for example, you can pickle that model to a file using pickle.dump(model, file), then load it back up anywhere else with model = pickle.load(file).

We have also added a progress bar for models that take a long time to build. Previously, we showed a progress bar when simulating large models, but not when building them. Now you will get a progress bar for each step.

The Nengo team also grew this release. Welcome to the team, Ben Morcos!

To see the full list of changes in Nengo 2.3, 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?

Please come to the Nengo forum! We welcome any questions and suggestions you might have, and invite you to share your Nengo creations there.