What it does
Uses a neural network to figure out how safe food is to eat. Helps users compare foods and fosters a community of the colorblind. Provides information on disability inequality and lifestyle effects.
How I built it
We built the website with HTML, CSS and JS. We wrote the iOS app in Swift and used Firebase Auth and Firestore to store user data. The model was trained with Microsoft Azure Custom Vision and integrates with CoreML on-device. We used an open-source design container controller.
Challenges I ran into
Training a respectably accurate model in time. Getting the classifier to work in real-time.
Accomplishments that I'm proud of
Getting a product working on two different platforms in less than a day, having comprehensive features, and training a neural net.
What's next for Dejuice
Continuing to expand and improve on the model.