What's In My Fridge is a personalized web-application that recommends delicious recipes based on what you have in your fridge. Our goal is to maximize yumminess and convenience for the user with minimal pain when deciding what to eat. The website can be found at https://www.whatsinmyfridgeco.com/.

How to build/run our program from the command line

In order to build our program, run mvn package. To run it, use ./run The gui will be hosted at http://localhost:4567/. Along the way we used arguments in order to isolate certain functionality. For example ./run --init does a bunch of our one time set up like creating the database.

Functionality

Recommending based on ingredients :

Recommending from user ingredients using recipe vector-sum algo and filtering on dietary restrictions.

Pantry :

Users can maintain a list of ingredients with their account to avoid reentering.

Favorites :

Users can record and review recipes they like easily in their account.

Google Sign In :

Customizes user experience, lets them use the same account across devices

Recommending based on favorites:

Recommending from user favorite recipes using the second approach

Databases

We use heroku postgres to store our SQL tables: recipe, guser, favorite, pantry, exclude. Recipes in recipe table were scraped from BBC Good Foods with the help of this script -> https://github.com/dspray95/open-recipe.

How to run our tests

In order to run our unit tests (without recompiling), run mvn test. To recompile and test, run mvn package. In order to run our system tests, run the cs32-test executable with any system test located in the ./tests directory. Many of the system tests will require 60 seconds, so use the -t 60 flag.

System Tests

Made sure our gui supported all our commands and functionality. We created a list of commands that our gui should be able to perform and then filmed all of them to ensure that they worked and documented it. Below are the two videos we took: https://drive.google.com/file/d/1wfpJoXDLN48OlmcRmGEZlP6FVdvGc-W5/view?usp=drivesdk https://drive.google.com/file/d/1zJc7uVdnF0xxCmb_DitFdSrEOvUUOSEL/view?usp=drivesdk https://docs.google.com/document/d/14hCH3-k2Tm8l9KTy_AbMS0pbzlgn_PhZnhA7_VvH0KQ/edit?usp=sharing

Partner division of labor

Nate

Nearest Neighbors KDTree Alg, Word Embeddings, recommended recipes

Eyal

Ingredient-based algorithm, Word Embeddings, Restriction food groups

Shalan

Initial recipe sqlite DB + tokenizing. Front end Gui design and functionality.

Ben

Front end caching and functionality, databse management, word suggestion alg and tokenizing, Google Sign In.