This vehicle detection machine learning algorithm takes as input images or videos of vehicles and outputs the same images/videos and draws bounding boxes around them with a label and probability of that label being true.
You can find the dataset used on kaggle here: https://www.kaggle.com/datasets/alkanerturan/vehicledetection
The code is based off of alkan erturan's code which you can find here: https://www.kaggle.com/code/alkanerturan/objectdetection-yolov8
My main goal was to see how running a jupyter notebook locally worked, and it turned out to be very educational.
This is how the algo performed on a video I filmed: https://youtu.be/lBezAjISCm8
I hope you have fun tweaking and running this notebook on your computer.