Euridice is a framework designed to provide the tools for intuitive data quality improvement through the process of data cleansing. Built for a thesis project at Leiden University, the software was developed to clean data used in data science pipelines through interactive and intuitive workflows. It should be used in the context of a consultant assisting a user of the to-be-cleaned data with building workflows to clean data.
The framework consists of multiple core applications / services:
- Lens: React-based Workflow Editor
- Lake: Dask-wrapping DataFrame storage
- Flow: Workflow Templating & Rendering
(DAG -> BFS -> IPython) - Engine: GraphQL API mapping for JupyterLab Server
Euridice features a complete REST API as well as a GraphQL API for easily developing and integrating with other tools and systems.
Euridice runs as multiple APIs in the Django Python framework with a PostgreSQL database.