Cross-Cause Cost-Effectiveness Model

Overview

The project is contained in three repos:

1. ccm: The core model.

The Cross-Cause Cost-Effectiveness Model is a Python program that provides functions for computing distributions of outcome effectiveness under several models of intervention efficacy -- notably, existential risk and animal welfare.

2. ccm_api: The web server.

The project includes a web app for visualizing these distributions. The functions of ccm are served through an openapi framework built in ccm_api.

3. ccm_react: The front-end web app. This talks with the ccm_api server and displays results in html.

Setup

Requirements

You should have:

• Python 3.10
• Poetry
• node 18
• npm

Setup Virtual Environment

From the top-level project directory, run:

• poetry install
• cd ccm_react; npm install

You don't need to re-run these commands unless the project dependencies change.

Activate Virtual Environment

Running poetry shell will spawn a new shell within the virtual environment. You can run exit to exit the virtual environment. You can also configure your editor to use this virtual environment. You can find the relevant path by running poetry env info -p.

Running the API

• To start the API server, run uvicorn ccm_api.main:app --reload or npm run uv from the top-level directory.
• To see the API documentation, navigate to http://localhost:8000/docs in your browser.

Running the web UI

Note Web functionality requires the API being running locally.

• Install the dependencies by running npm install in the ccm_react directory.
• Run npm run dev to start the development server.
• Navigate to http://localhost:5173/.
• To see the UnoCSS inspector, navigate to http://localhost:5173/__unocss/ in your browser.

Regenerating Client

• The react app automatically configures some utilities to the structures returned by the webserver. After making changes to the API, you need to regenerate the client in ccm_react:
  • Start up the API. If it was already running, top it and restart it.
  • While the API is running, from ccm_react, run npm run generate-client.

Running tests

• To run all Python tests: pytest
  • Add -s argument to run all tests verbosely showing all stdout
  • Add -m "not slow" to skip slow tests
• You can also run individual test cases by providing the name: pytest tests/test_population.py::test_get_life_years_affected
• To run JavaScript tests: npm run test from the ccm_react directory
• To run Cypress E2E tests: npm run test-cypress from the ccm_react directory

Typechecking, formatting, and linting

• To validate Python code, run ruff check . and pyright.
• To format Python code, run black.
• To format JavaScript code, cd ccm_react and run npm run format.
• To validate JavaScript code, cd ccm_react and run npm run lint.