Rev It Up

This is the first screen you'll see! You enter your song here
Here is a recommended song
Here is part of the python code in juypter

Inspiration

Lets face it, finding a good song can be really hard. We love music and we wanted to generate songs similar to our unique music taste!

What it does

This web app takes a song that the user inputs, and, when you click the new song button, it will provide a recommendation for a song that is similar. If you don't like that song you can press "new song" again, and be given another similar song!

How we built it

The Back End

The back end was programmed in python using Jupyter Lab. We first imported a data set from Kaggle that cataloged songs. Then we converted all of the data into floats so we could compare numerical data. We created a weighting system to emphasize certain characteristics of the music, such as the genre, that are most important to providing an accurate song recommendation. We also used Machine Learning through the nearest neighbors system to further locate and confirm recommendations.

The front end

The front end was designed using html and css. We created a visually appealing web app where you enter a song title and artist name, and then when the new song button is clicked, a second page pops up showing the recommended song and some of its statistics.

Combining the two

We combined the front and back end using javascript and flask so that the python could be run in a web app.

Challenges we ran into

We really struggled with getting the front and back end fully integrated however, after multiple tickets we continued to see improvement in the apps ability to fetch songs. It also took a considerable amount of time to get the machine learning and weighting system to take the genre into account as it wasn't a numerical value.

Accomplishments that we're proud of

We are really proud of python code because it took a long time to get successful, and it was the first code that Kiersten had truly came up with. We are also proud of our formatting of the front end, and we think it looks visually appealing.