Marigold | Devpost

Inspiration

Over the past year, COVID has made it difficult for communities to stay connected and keep in touch. While activities like party games, online social gatherings, and other virtual events have sprung up, there hasn’t been much to keep friends and communities updated on how everyone’s doing on a day-to-day basis. That’s where Marigold comes in.

What it does

Marigold is a web-based service that uses machine learning and a webcam to keep track of your emotions throughout the day and documents them in a calendar. This way, not only can you know yourself better, but you can also check in on friends and let others know how you’re doing throughout the month.

How we built it

This project was split into 3 parts. The first step was utilizing Figma to prototype and plan out the whole project and each what the finished product would look like. We then split the project into two parts, the machine learning algorithm and the web app development. The machine learning algorithm was created using python, while the web app was created using python alongside flask and firebase and html for styling.

Challenges we ran into

The biggest challenge we ran into was implementing our web app. We had several issues trying to create a website based in python that could run our machine learning algorithm. While we did have experience using javascript to make websites, coding in python made in much harder and was where we spent the most time.

Accomplishments that we're proud of

One accomplishment we are especially proud of is our machine learning algorithm that is able to classify a person's emotions. Previously we didn't have much experience working with any type of AI so this was a big challenge but one that we handed very well

What we learned

Through this project we learned a lot about coding with Firebase and Flask, two software tools that we didn't have much experience, if any, working with.