This project aims to provide a framework to execute structural equation model searches as unit tests.
It gives access to a wide range of causal relations by allowing the creation of functional equations models from its dataset generator.
The environment is self contained by enjoying the power of docker. All dependencies will be installed in an anaconda virtual environment inside a docker container.
Run make start to install and launch the container. It will take a few minutes only the first time to install all packages.
Once installed run make run to launch causal explorations on a few datasets to validate everything was properly installed.
Datasets and algorithms executed by the runcommand are defined in the src/ui/config.py configuration file.
A list of datasets are available in the folder src/datasets/ grouped by type:
- unit: as unitary datasets
- causality: as a range of causal relation types
- integration: as simulated real case datasets
To execute a jupyter notebook from the container run make jupyter notebook
Check the makefile for a list of all available commands.
This is a master degree final project, and it is not meant to be production ready.
It was only tested on a linux environment, and some commands might not work on windows.