Skip to content

🔢 CausalIQ Analysis Project

The CausalIQ Analysis project provides Tools for analysing and visualising learned causal graphs, including structural metrics, stability assessment, significance tests, and publication-ready tables and charts.

  • Foundation metrics: CausalIQ and Bayesys structural graph metrics and KL metric.
  • Legacy trace: Support for reading and writing structure learning traces in legacy pickle format (this will be superseded by a more open format).
  • Graph Averaging: Graph averaging to produce arc probabilities.

Quick Links:

Upcoming Key Innovations

🧠 LLM-assisted Graph Averaging

  • Uncertain or conflicting edges - resolved using LLM queries

📊 Publication-ready chart generation

  • Seaborn charts - flexible, but standardised publication-ready chart generation

▦ Publication-ready table generation

  • LaTeX tables - converts tabular analysis data into publication-ready LaTeX tables

Integration with CausalIQ Ecosystem

  • 🔍 CausalIQ Discovery generates causal graphs which this package evaluates and visualises.
  • 🤖 CausalIQ Workflow can access all features of this package (through the Action interface) so that analysis and visualisation are incorporated into CausalIQ workflows.
  • 🧪 CausalIQ Papers uses the analysis, table and chart features of this package to generate published paper assets.


CausalIQ Data represents the foundational data processing layer that enables robust, high-performance causal discovery through optimized scoring functions, conditional independence testing, and seamless integration across the entire CausalIQ ecosystem.