FuelPrediction

Inspiration

Chevron's presence at the competition inspired us to focus on something regarding fueling and/or vehicles. We wanted to create a project that could affect peoples' everyday lives in a small or simple way.

What it does

Determines what grade of fuel a car should pump, based on an image of the vehicle

How we built it

We used tflearn in Python to create a convolutional neural network that created an algorithm that determines what car a fuel takes.

Challenges we ran into

-The lack of available data took much more of our time than originally expected. -The lack of time to refine and present the project.

Accomplishments that we're proud of

Being able to gather up enough data to come up with a reasonable model.

What we learned

-The difficulties of finding specific data sets, as well as specific data, in general. -The amount of time to train a model took too long in the context of a timed competition, which didn't allow us much time to tweak and refine our weights within the model