What is MicroPython? Do I Need to Know it as a Data Scientist?

When I saw MicroPython on the list of the Stack Overflow survey from this year, I wanted to know what I could use this language for. And I wondered if it could serve as a bridge between hardware and software. In this article, I break down what MicroPython is and what data scientists should know about it.

Table of Content 1 – What is MicroPython and why is it special? 2 – Why should I know MicroPython as a data scientist? 3 – What is the difference to Python and other programming languages? 4 – What does this look like in practice? (Only with web based simulator) 5 – Final Thoughts & Where to continue learning?

What is MicroPython and why is it special?

MicroPython is a simplified, compact version of Python 3 designed specifically for use on microcontrollers and other low-resource embedded systems. As we can read on the official website, the language offers a reduced standard library and special modules to interact directly with hardware components such as GPIO pins, sensors or LEDs.

Reference: Official MicroPython Website

Let’s break this definition down:

Simplified, compact Python: MicroPython is designed to use less memory and computing power than the standard Python version. The language is perfect for devices with just a few kilobytes of RAM.
Microcontrollers & embedded systems: Think of a microcontroller as a tiny computer on a chip. It can control devices such as IoT sensors, smart home devices and robots.
Low-resource systems: This means that these systems have little memory (often less than 1 MB) and limited computing power.
GPIO Pins: These are pins on a microcontroller that can be used for various input and output functions. For example, they can be used to control LEDs or read sensor data.

Why is MicroPython relevant?

If you know Python, you can program hardware with MicroPython – without learning a new complex language like C++ or assembly. Sure, you have more options with C++ and assembly and both are closer to machine languages. But if you want to create a prototype with relatively little effort, MicroPython offers you an ideal starting point.

Why should I know MicroPython as a data scientist?

Simply put: Because it is listed in the Stack Overflow survey and gaining traction in the developer community…

IoT and edge computing are playing an increasingly important role in AI and Data Science projects. Especially as we want to make our cities smarter (smart cities).

MicroPython can serve as a bridge between hardware and software here, as it makes it possible to collect sensor data and process it in data science pipelines or machine learning models. For example, a MicroPython sensor can measure air quality and send the data to a Machine Learning pipeline. MicroPython can also run simple AI models directly on devices (edge computing) – this makes it ideal for local computing without the device being dependent on the cloud.

So my conclusion: MicroPython makes hardware more accessible for data scientists. If you know Python, you can also use MicroPython and apply it in a smart home project.

What is the difference to Python and other programming languages?

While Python was developed for general software applications that run on powerful devices such as PCs or servers, MicroPython was developed for low-resource devices such as microcontrollers, which often only have a few kilobytes of memory and computing power.

As we all know, Python offers an extensive library for data analysis (pandas, numpy), machine learning (scikit-learn, tensorflow) or web development. MicroPython, on the other hand, only contains a reduced standard library and slimmed-down modules such as ‘math’ or ‘os’. Instead, it offers special hardware modules such as ‘utime’ for timers or ‘machine’ for controlling microcontroller pins.

While Python is better suited for data-intensive tasks, MicroPython enables direct access to hardware components and is therefore ideal for embedded systems (e.g. everyday electronics such as microwaves & smart TVs or medical devices such as blood pressure monitors) and IoT projects.

What does this look like in practice? Application areas and a quick simulator demo

In which areas is Micropython used?

Internet of Things (IoT): MicroPython can be used to control smart home devices or control sensor data for dashboards.
Edge computing: You can run machine learning models directly on edge devices (e.g. IoT sensors, smartphones, routers, intelligent cameras, smart home devices, etc.).
Prototyping: With relatively little effort, you can quickly set up a prototype for a hardware project – especially if you know Python.
Robotics: MicroPython can be used to control motors or sensors in robotics projects.

Flashing LED in the simulator as a practical example

Since as a data scientist or software specialist you probably don’t want to buy hardware just to try out MicroPython, I explored a MicroPython simulator available online. This is a simple and beginner-firendly way to get started with programming hardware concepts without the need for physical devices:

Open https://micropython.org/unicorn/
Import time, then define the function and call the function at the end. Type in each code snippet in the web terminal separately and then click ‘Enter’. You can use the following code for this:

#Provides functions to work with time
#(standard Python library instead of 'utime', as the code is used in the simulator)
import time

# Simulated LED by defining the function
def blink_led():
    for _ in range(5):
        print("LED ist jetzt: ON")
        time.sleep(0.5)  # Waits for 0.5 seconds
        print("LED ist jetzt: OFF")
        time.sleep(0.5)

# Start the blinking by calling the function
blink_led()

Now we see that the LED light (only in the console) switches back and forth between ON-OFF. In this simulator example, I only used the time library for the delay. To run the example with real hardware, you should use additional libraries such as ‘machine’ or ‘utime’.

In a web simulator, we can write a 'hello world' by writing a small script to output a flashing LED. — In a web simulator, we can write a ‘hello world’ by writing a small script to output a flashing LED.

Final Thoughts

MicroPython is certainly important for people working in hardware projects such as IoT and edge computing. But due to its easy accessibility and because anyone who knows Python can also use MicroPython, the language bridges a gap between data science, AI and hardware technology. It is certainly good to at least know the purpose of MicroPython and the differences to Python. If you are interested in trying out smart home devices or IoT for yourself, it is certainly a accessible entry point.