Inspiration

After receiving a lot of data, Azfar got the idea of making an app that instructed students when and where to work according to temperature, wifi strength, humidity, and location density.

What it does

Our custom made machine learning model, using the sklearn library, determines if a day is good to study at school. To achieve a 97% accuracy our model uses the small dataset given to run a Logistic Regression model.

Using a machine learning model requires less computing than a deep learning model and supports the low amount of data. We can allow multiple students simultaneously to use the website and a fast speed with a machine learning model. Students are able to avoid highly populated areas with this and find places to affectively study. The signal map provides students with the places with the top signal so that they can study affectively.

How we built it

On this link you'll find a detailed explanation to the machine learning model we made. https://erindale.tk/works

Challenges we ran into

At first we were going to use an express.js server written with node.js, but we ran into the issue of language integration. Since we didn't have much time we decided to go with a flask server to keep everything limited to python.

Accomplishments that we're proud of

  • We were able to successfully use the Google Maps Api and tie that in with the server-side rendering of the information.
  • We are also proud of successfully being able to live update the information. Meaning, if the day changes the information will automatically change too ## What we learned
  • We learned how to use a small amount of data effectively to help for a good cause
  • We also learned how to make a simplified UI so students could use the app without trouble

What's next for Grind Time

  • With more data, we hope to expand Grind Time into something bigger with more information to all types of schools around the world!

Built With

Share this project:
×

Updates