This repository contains the code for the webapp that accompanies the Children's Social Care Demand Model.
The existing live application runs on pyodide and requires the user to upload the 903 files. This tool is expected to read the data from a remote source (an S3 bucket or similar) and create similar analysis as before, with the following beneficts:
- A new user interface with a focus on user experience
- It uses the new generic demand model which, besides predicting the number of children, also includes a confidence interval.
- Since it's a standard webapp, it now has a database, which makes it possible to save user preferences and scenarios, as well as handle authentication and permissions.
You need to have python and poetry.
- Install dependencies with
poetry install; - Launch a poetry shell so the virtual environment is active:
poetry shell; - Run
python manage.py migrateto create the databases tables; - Add the pre-commit hook by running
pre-commit install. This will ensure your code is formatted before you commit something;
To load some dummy data, run:
python manage.py load-fixtures
This will load some users:
Regular user
- email: [email protected]
- password: test
Admin user
- email: [email protected]
- password: test
To manage data transfer between views, users are assigned a session scenario. These scenarios should be deleted weekly through a recurrent job wherever the app is deployed. To run this command:
python manage.py erase_session_scenarios
Run python manage.py runserver to access the regional demand model platform. it will be live at http://localhost:8000.