What is ulab?

ULab is a library that has some of what the python Numpy library offers, but in the CircuitPython or Micropython context. Ulab is definitely worth getting to know on the Microcontroller Python variants.

It can run on small devices, like the Raspberry Pi Pico so you can manipulate arrays of numbers on a robot (or gadget) a bit quicker than simply looping over them. It also makes it convenient to express a formulae like take a list, add some other lists values to every item, then convert that to radians:

from ulab import numpy as np
angles = np.array(get_angles_from_some_sensor(), dtype=np.float)
noise = get_an_np_array_of_noise(angles.shape[0])
angles += noise
angles = np.radians(angles)

So the get_angles_from_some_sensor and get_an_np_array_of_noise are fictitious, and may need to be considered for your context, but assume one reads a number angles from some sensor, and the other provides n points of random noise.

The following line lets us add them, and the final line converts them all to radians. Nice and convenient.

Multi-dimensional arrays

Things get interesting when you are talking about coordinates. You can make numpy arrays that are multi-dimensional:

coordinates = np.array([random_x(), random_y()] for n in range(400), dtype=np.float)

We can also add things to this array, and manipulate slices of it along either axis. For example coordinates[:,2] allows us to reference, manipulate or set every y-coordinate in that list.

Debugging ulab or numpy arrays

When debugging them, the most useful tip is to print the shapes of the arrays. Any time things seem a bit off, print the shapes:

print(coordinates.shape)

This can also help in the full computer Bumpy context, although there you might be able to use a debugger (that can be tricky if you are using interactive matplotlib or have asynchronous things going on).

On the Raspberry Pi Pico (or other CircuitPython/micropython devices), step debugging python is much less of an option. Printing out the whole arrays can be too big and then unhelpful. So - print the shapes!

Interesting gotcha with for loops

It is fair to say that when using numpy, you are probably going to try to avoid loops at all costs. However, if you have them, beware that a problem with a range will show the last line of the loop, and not the first line. This can be quite confusing behaviour. I’ve seen this by leaving the .shape[0] off when looping over multiple arrays (ie one persistent array plus a temporary one so using an index).

Robotics at Home with Raspberry Pi Pico

This post was based on research I did for the book Robotics at Home with Raspberry Pi Pico which is available now.