Inspiration
Our love for food
What it does
SnapDish is an AI and API powered web application, that allows a user to enter an ingredient, or take a picture of an ingredient to suggest recipes (using AI image detection).
How we built it
Used TheMealDB API and Ultralytics YOLO pre trained AI for image detection.
Tech Stack Front end: HTML, CSS, Jinja Back end: Python Flask, MySQL
Challenges we ran into
We initially ran into issues related to collaborating with Git and GitHub, such as merge conflicts, but we eventually sorted it out. Working with AI was another challenge, as it was all of our first time. Training the AI model to detect food objects took about 7 hours of just letting it run.
Accomplishments that we're proud of
Worked with AI for the first time and completed our first hackathon. Within 24 hours, we were able to produce a project that we're all very proud of.
What we learned
How to integrate and train AI into a web application, expanded on Python Flask knowledge, and collaboration using GitHub and Git.
What's next for SnapDish
Proper image detection for food (further training), User Fridge functionality
Log in or sign up for Devpost to join the conversation.