WhenCanIPlay AKA GoofyBehaviour

• GIF

Inspiration

We were inspired by the lobby of a DBD game, and were excited to practice machine learning practices learned in class.

What it does

WhenCanIPlay.tech asks you your rank, your server, your party size and your role for you to get a top-notch estimated wait time until your next game. We have three models available, one where more killers are online, one where there are more survivors, and one where we adapt it with the time of the week.

How we built it

We created a flask app and vite. In the flask app, we query the models we previously trained by creating random data and tuning it with our own perception, since we did not have available dataset. We also use Axios to make API calls between our frontend and backend.

Challenges we ran into

We did not have a working dataset. Therefore we spent several hours trying to come up with a script that generated the required amount of data (50000 players). Having people working on macs and others on windows, we ran into the problem of compatibility with flask_cors.

Accomplishments that we're proud of

We could generate 150'000 lines of data and train it completely (thanks M1), and we created a good looking website that is able to query our models while keeping things simple

What we learned

What's next for WhenCanIPlay AKA GoofyBehaviour

Built With

• flask
• machine-learning
• python
• react
• scikit-learn
• vite