Skip to content

🔢 CausalIQ Data Project

The CausalIQ Data project provides the data-related capabilities that causal discovery requires. These are:

  • data import and caching - data can be imported from standard tabular formats (comma-separated variables) and cached for high performance
  • graph scoring - provide graph score derived from the data which is the objective function used by score-based structure learning algorithms. This is based upon how likely the data is to be seen for a given graph, typically modified by a penalty for complex graphs (e.g. BIC score), or modified by a prior belief about the graph strcuture (e.g. BDeu score)
  • independence tests - used to determine conditional independence tests which are intrinsic to the operataion of constraint-based structure learning algorithms.

Quick Links:

Upcoming Key Innovations

🧩 Plugin Architecture

  • use by third-party software - ability to use these data capabilities in third party structure learning algorithms so that comparisons are based on a common scoring or conditional independence framework, and performance optimisations speed up third-party algorithms.

🏛️ Stability Integration

  • Stable scores - stable resolution of equal-score situations for unstable algorithms e.g. Tabu

🧠 LLM-assisted Causal Discovery

  • Data values - Data values and variable names may provide part of the context for LLM-assisted causal discovery
  • Knowledge integration - incorporation of LLM and human expertise in scores and priors via the CausalIQ Knowledge package.
  • Relationship explanations: Natural language descriptions of relationships in data

⚡Optimised Performance

  • GPU Data provider - support for optimised data handling on GPU hardware
  • Intelligent data scanning - reduce number of full-row data scans

🎲 Enhanced Distribution Support

  • Mixed Types: scores and independence tests that support mixtures of continuous and categorical variables

Integration with CausalIQ Ecosystem

  • 🔍 CausalIQ Discovery makes use of this package to provide objective functions and conditional independence tests for structure learning algorithms.
  • 🧪 CausalIQ Analysis uses score functions as part of the evaluation of learnt graphs.
  • 💎 CausalIQ Core makes use of the BNFit interface to estimate parameters based on data.
  • 🤖 CausalIQ Workflow uses the in-memory randomisation of this package for stability experiments.


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.