We were inspired by the stories of food waste in restaurants across America.

What it does

Provides restaurants with a tool to know exactly how much inventory to buy every week/month, and helps them reduce food waste and ensure that they always have enough food.

How we built it

We used the python library sklearn for the machine learning aspect. Specifically we used the linear regression and multi-layer perceptron regressor algorithms. Our data include the information for amount of food bought, thrown away, sales, and season, for different kinds of food. We split our data into different categories for each kind of food, and applied both regression models to each category to determine which type was the best. We then used Flask to make a web application to handle the user inputs and display the results of the regression.

Challenges we ran into

We couldn't find a dataset, so we had to create our own. We also had to learn how to use MLP regression. Working with Flask was also a challenge, since it was our first time working with it.

Accomplishments that we're proud of

We are proud of making our own dataset and getting the regression algorithm code to work.

What we learned

We learned how data processing works and how important it is. We also learned how to use Flask and how to apply the regression models to data.

What's next for Food Inventory

We would like to add more user functionality to the website, such as the ability to filter between certain food groups and cuisines.

Share this project: