Bevy

Have you ever been in a grocery store and you want to purchase a beverage for dinner, but you don't want to buy a subpar beverage? Introducing Bevy, your personal beverage specialist.

Comment

Inspiration

Our inspiration for this project was to help users figure out what beverage would work the best with their preferences. We could not figure out what we would like to drink with our dinner and we set out to solve this issue. A simple start, but a well developed app.

What it does

Bevy is your personal beverage specialist and uses computer vision and machine learning to help you select the best beverages according to your preferences.

How we built it

We built Bevy using Python, Computer Vision, OpenAI, Figma, JSON, and Dash.

Challenges we ran into

Being able read the barcodes on items was a struggle at first, however, once we found the appropriate API, we managed to read the back of them with relative ease. Another challenge we ran into was creating a formula that would accurately predict whether or not the user would like the selected beverage based on their preferences. After thinking for a bit, we came up with a great formula.

Accomplishments that we're proud of

Getting our Computer Vision model to be able to only read the barcodes of beverages and not other products was a great accomplishment for us. Additionally, being able to get our app to complete the recommendations was great for us.

What we learned

During this project we exposed ourselves to computer vision and machine learning models as well as dash, technologies that we had previously not used much in our development.

What's next for Bevy

Built With

Share this project:

Updates