I'm gonna be for real, I'm not 100% sure I understood the goal of this coursework but I hope I'm getting it right. I think I should replicate the results seen in the inventory analysis notebook which you sent us before. I did saw that you had variables for profit margins etc. in the original, so I decided to use the IPyWidgets library to create an interactive UI where you can change those variables, and see the results.

• analysis.ipynb — Analysis notebook
• environment.yml — Conda environment configuration
• database.sqlite — The database we'll be using

Initially, I did this with a PostgreSQL database and I included a Docker Compose with a migration script and environment variables, but then I realized that's really fucking stupid and overkill for this project, so I switched to SQLite3 instead.

You can see how I did the setup in the commit history (up to commit 19c674d) if you want.

I also created a view food_inventory_with_relative_expiry, which adds a column days_until_expiry that takes the difference between the latest expiry date and the row's expiry date:

CREATE VIEW food_inventory_with_relative_expiry AS
SELECT *,
    JULIANDAY((SELECT MAX(expiry_date) FROM food_inventory)) - JULIANDAY(expiry_date) AS days_until_expiry
FROM food_inventory;

I wanted this environment to be as replicable as possible, so the Python environment is running in Anaconda (download instructions) for replicability.

To create the Python environment, simply run:

$ conda env create -f environment.yml

When you run the notebook, you will be prompted to select a kernel. You will need to select the one that says coursework, as this is the Anaconda environment that we've just created.

I did use Claude Sonnet 4 to assist me with some more technical parts of this coursework, but for the most part it's actually me. You can see the Git commit history and see that it's mostly my own actual idiocy (AI).

Also, I needed to literally rewrite the Git history to remove some files because they were too large, like food_inventory.csv which I don't even use anymore, and database.sqlite. I then readded the database file and add it with Git LFS (Large File Storage).