Nengo Loihi 0.7.0 released

The Nengo team at ABR is psyched to announce the release of Nengo Loihi 0.7.

What is Nengo Loihi?

Nengo Loihi is a backend for the Nengo neural simulator allowing standard spiking Nengo models to run on Intel’s new Loihi neuromorphic chip. Nengo Loihi includes an emulator so that you can develop spiking neuron models for Loihi without having access to hardware.

How do I use it?

Once installed, Nengo Loihi simulates Nengo networks. After defining your network, instead of simulating it with




and that’s it!

What’s new?

Version 0.7.0 is the first version of Nengo Loihi to support multi-chip communication. Currently, the only way to split up a model onto multiple chips is with the RoundRobin allocator, which assigns each ensemble to a different chip in a round-robin format. To use it, pass hardware_options={"allocator": RoundRobin(n_chips=8)} to nengo_loihi.Simulator.

One small improvement that will immediately help anyone using the set_defaults function is that we now use a better intercept range, going from -1 to 0.5 instead of -0.5 to 0.5.

Under the hood, we have made several improvements, including the use of Nengo Bones, a significant splitter refactoring, and obfuscation of low-level parts of NxSDK.

Finally, several less common use cases are now either possible or had a bug which has been fixed. This includes connecting from Neurons to Ensemble objects, making connections and probes with object slices, and connecting to on-chip neurons. To see the full list of changes in Nengo Loihi 0.7.0, head to the Github release page.

How do I get it?

If you are using the emulator, then you can simply use pip.

pip install nengo-loihi

If you are setting up Nengo Loihi to use Loihi hardware for the first time, see the Nengo Loihi installation instructions.

Where can I learn more?

Where can I get help?

You’re already there! If you have any questions about Nengo Loihi, please ask in the Loihi category. And if you run in to any bugs or have suggestions for new features, file an issue through Github.