I am currently pretty new to machine learning in devices. I will be implementing this project of eye drowsiness detection in Arduino Uno Q. I have started with some few data inputs of both eyes awake and drowsy eyes. I am using my camera’s 2K Resolution Webcam
My first attempt didn’t work, I used the object classification, and attempted the visual anomaly detection and F1 score in classification is always 0%. I would like to ask for a guide in starting this out and implementing this project. Thank you!
1 Like
Hello @rellebruhh first of all welcome to the Edge Impulse community!
Interesting project! Looking forward to see this working!
Could you please share more details on how did you train your object classification model? How many classes do you have? How many images on each class?
Please share more details! Thanks 
Hello, during the past few days I failed to realize that I should’ve used Transfer Learning instead and it’s showing good results recently based on the dataset I provided. The split between training and testing is now also balanced 80/20, I appreciate the response and I will be updating the forum if any issues occur during this process. Thank you!