Running the Demo

This project is written in Python, and tested with pytest.

The assignment

After building your CLSLab:Light demo and testing it per the tutorial instructions*, edit your main.py file and replace these lines:

    my_id = hexlify(unique_id()).decode()
    my_encrypted_id = encrypt_id(my_id, verbose=True)

with:

    my_id = "<your_id_here>"
    my_encrypted_id = my_id

where <your_id_here> is the GitHub Classroom course identifier you chose. Then, edit course_identifier.py and replace the string with your course identifier. For example, if your course identifier is funky-zebra, then you would replace:

course_id = "<your_id_here>"

with

course_id="funky-zebra"

If your CLSLab:Light demo is up-and-running, the tests will pass and a full optimization campaign will run on your device.

*NOTE: Recently, HiveMQ Cloud changed such that hivemq-com-chain.der (a Certificate Authority (CA) file) is not transferrable across different broker instances. The latest hivemq-com-chain.der file from self-driving-lab-demo will be hard-coded to the self-driving-lab-demo public test credentials (i.e., what is used in Module 1 - Running the Demo). However, this assignment requires you to create your own HiveMQ Cloud broker instance, so you will need to generate a hivemq-com-chain.der file specific to your instance and upload it to your microcontroller.

Optional

You can also view the full workflow interactively. First, navigate to 4.2-paho-mqtt-colab-sdl-demo-test.ipynb and replace the string value for PICO_ID with the one corresponding to your device. For example, if you PICO_ID is a1b2c3, then you would replace PICO_ID = "test" with PICO_ID = "a1b2c3". Then, run each cell in the notebook to see the full optimization campaign, including visual comparisons between grid search, random search, and Bayesian optimization.

Setup command

pip install -r requirements.txt

Run command

pytest

Notes

For context about how or when to edit main.py, see step 14 of the build instructions