More fun with Spiking Neural Networks and JAX

By -

I'm currently exploring Spiking Neural Networks.

SNNs (Spiking Neural Networks) try to model the brain more accurately than most of the Artificial Neural Networks used in Deep Learning.

There are some SNN implementations available in TensorFlow and PyTorch but I'm keen to explore them using pure Python. I find that Python code gives me confidence in my understanding.

But there's a problem.

SNNs need a lot of computing power. Even if I use numpy, large-scale simulations can run slowly.
Spike generation - code below

So I'm using JAX.

JAX code runs in a traditional Python environment. JAX has array processing modules that are closely based on numpy's syntax. It also has a JIT (Just-in-time) compiler that lets you transparently deploy your code to GPUs.

Jitting imposes some minor restrictions on your code but it leads to dramatic speed-ups.

JAX and robotics

As I mentioned in an earlier post, you can run JAX on NVIDIA's Jetson family. You get excellent performance on inexpensive hardware with a low power budget.

If, like me,  you're a robotics experimenter, this is very exciting!

A helpful and knowledgeable JAX community

To get the best from JAX you need to think in JAX. I've a head start in that I've been using APL for over five decades. APL, like JAX and numpy, encourages you to think and code functionally, applying your functions uniformly to tensors.

There a lot of JAX skills I still have to master, and yesterday I hit a problem I couldn't solve. I needed to create the outer product of two arrays using the mod function. I know how to do that using APL or numpy but I couldn't see how to do it using JAX.

I asked for help in JAX's GitHub discussion forum, and quickly got a simple, elegant solution. The solution showed me that I'd missed some fundamental aspects of the way that indexing and broadcasting work in numpy and JAX. 

The advice came from Jake Vanderplas - an Astronomer turned Google Researcher who has given some excellent talks about Python in Science, including tips on performance.

Generating input spikes with JAX

You may be interested to see the SNN problem I was trying to solve and the code I ended up with

I wanted to generate multiple regular spike trains, with each train having its own periodicity.

I've seen Python code to do that but it's been truly awful: hard to read and slow to execute.

I knew there had to be a simple way to do it.

In APL I'd just do a mod outer product of the periods and a vector of time indices. A zero in the result would indicate that the time was a multiple of the period, so that neuron should fire.

Here's the APL code:

      a ← 1 5 10 ⍝ firing period
      b ← 10 ⍝ time steps
      spikes ← 0 = (1+⍳b) ∘.| a

So far so good, but I could not see how to do the same thing in JAX.

Here's the code in JAX

I'm gradually building up a library of Spiking Neural Network code using Jax. If you're interested, let me know on twitter: @rareblog.



Comments

Post a Comment

Popular posts from this blog

Controlling a Raspberry Pi Pico remotely using PySerial

By -
Image
Update : I'm using thie code below in another project, and found that I had not correctly fixed the reported bug. The new version passes automated tests, and I am pretty sure it works OK. I have changed the name of the class to Talker  since it can both send and receive information. Apologies to all concerned for the bug! Introduction You can use a Raspberry Pi Pico as a powerful peripheral to a host - a Raspberry Pi, a Jetson Nano, a laptop or workstation. In this article you'll see how to interact with a Pico running MicroPython or CircuitPython by writing some Python code that runs on the host. The software is easy to use. It enables you to send a Python statement to the Pico and read the results. The statement can be any valid MicroPython code. Setting up the host and the Pico For this article I've used a Raspberry Pi as the host, but any computer running Windows, Linux or Mac OS will do so long as it has Python 3.5 or later installed. In particular, you can use this t
Read more

Five steps to connect Jetson Nano and Arduino

By -
Image
Yesterday's post showed how to link a micro:bit to the Jetson Nano. One of the members of the (unofficial) NVIDIA Jetson Nano group on Facebook asked about connecting an Arduino to the Jetson. Here's a simple recipe for getting data from the Arduino to the Jetson Nano. It should work on all the Jetson models, not just the Nano, but I only have Nanos to hand. On request, I've added a recipe at the end of this post which sends data from the Jetson Nano to the Arduino; it turns the default LED on the Arduino on or off. The recipe for sending data from the Arduino to the Jetson has just 5 stages: Program the Arduino. (I used the ASCIITable example). Connect the Arduino to the Jetson using a USB connector Install pyserial on the Jetson Download a three-line Python script Run the script.   Programming the Arduino I used an Arduino Uno, and checked it on a verteran Duemilanove (above), but any Arduino should work. You'll need to do this step usi
Read more

Raspberry Pi Pico project 2 - MCP3008

By -
Image
This post show you how to drive the MCP3008 8-channel ADC using a Raspberry Pi Pico. I'll start with a personally embarrassing and annoyingly relevant story. I'll also tell you about a minor 'gotcha' when using SPI on the Pico, and how you can avoid it. Finally, I've include the code and fritzing diagram I used to test the interface. First, the story. Back in 2012, when the Raspberry Pi was fresh and new, I was running a startup called Quick2Wire.  We made add-on boards for the Pi. Just before Christmas we sent out our first batch of boards to a small army of beta-testers. The beta boards didn't work. Somehow a single trace on the pcb had got deleted and one of the GPIO pins was isolated. The boards were still usable and it wasn't too hard to solder in a jumper wire to replace the missing trace, but it was very embarrassing. It wasn't just that we'd shipped boards with a defect. The really annoying thing was that I'd set up an  automated loop-bac
Read more