Hotdog Sales Forecast

Inspiration

We want to help hotdog sellers to not waste food, helping our environment and their economic business!

What it does

Using a random forest model, we were able to predict the number of hotdogs on a given day as well as use feature importance analysis to determine the most important features of the dataset.

How we built it

We used pandas for data preprocessing. Then, we used scikit-learn and numpy for our random forest model.

Challenges we ran into

It was a small dataset and there were a limited number of features.

Accomplishments that we're proud of

We added more important features (like days of the week) that helped our model, and chose a model that was able to get a good RMSE error.

What we learned

We learned how to manipulate datasets, add more relevant features, and train new models.

What's next for Hotdog Sales Forecast

We would try to get more specific features like weather. Then we would make a tool to let the hotdog store owners know how many hotdogs they should make to minimize waste!