Inspiration
Most of the members of our group have some sort of dietary restriction, so we all knew how much of a problem it is for people to traverse stores and find products that they're able to eat.
What it does
Our project uses a Tensorflow model that we trained to compare an image you take to different kinds of foods, and tell you if a food violates some dietary restrictions which you selected on the main screen.
How we built it
We built our project on a group replit page, using HTML, CSS, and Javascript to host and create the website, as well as Python to create and test our Tensorflow model.
Challenges we ran into
This was the first time any of us used Tensorflow, or ML in general, and for many of us, the first time using Javascript, so we ran into many learning issues. As well, our project was originally going to be a Flask webpage in Python, but the version of Tensorflow model that we trained wasn't compatible with the version of Python replit was using, so we had to switch to pure web development without any Python halfway through the contest.
Accomplishments that we're proud of
Our team is very proud of the knowledge we've gained about working with ML and Web development, and the bonds we've formed together.
What we learned
We learned lots of things about ML, Web Dev, and even scheduling.
What's next for Tasty Snack or Heart Attack?
We plan to add more foods and dietary restrictions, to help make the program accessible to more people.
Built With
- css
- html
- javascript
- tensorflow
- tensorflow.js
- tensorflow.keras
Log in or sign up for Devpost to join the conversation.