The Nengo team at ABR is excited to announce the release of Nengo Loihi 0.5.
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
nengo.Simulator(model)
do
nengo_loihi.Simulator(model)
and that’s it!
What’s new?
Nengo Loihi is still changing rapidly, but in this release we have made several changes under the hood that should improve model stability and solidify the Nengo Loihi API.
With this release, we now support NxSDK 0.8.0 while maintaining support for 0.7.0 and 0.7.5.
We have fixed several bugs that impacted the performance of several common types of models. Current and voltage is now integrated more accurately, which significantly improves performance of recurrently connected networks like integrators. We now use more neurons with more diverse properties to communicate values between ensembles, which improves performance significantly for networks with multiple ensembles connected to the same ensemble (i.e., when one ensemble sums input from multiple ensembles).
Error-driven learning with the PES learning rule is now more user friendly because PES learning in the emulator more closely matches PES learning on hardware.
We now use population spikes in certain situations, which allow for more efficient transmission of information to the chip. Population spikes have been used to classify CIFAR-10 images with a convolutional network.
Finally, we have made several other bug fixes and improvements under the hood, including several fixes for different types of probes. To see the full list of changes in Nengo Loihi 0.5.0, 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?
- Nengo Loihi documentation
- Instructions for installing Nengo and Nengo Loihi
- An introduction to Nengo
- General Nengo documentation
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.