Durable experiment record · Local core code · curated results

Judo Calculus causal-discovery experiments

Sheaf/Kan causal-discovery experiments with DCDI and GES comparisons.

31 code surfaces0 frozen runs2 code packagescatagi-judo

Experimental contract

What this family tests

synthetic and Sachs benchmark studies

  • Synthetic j-stable DCDI comparison
  • Descriptive runtime measurement
  • Sachs pooled versus regime-stable study
  • GES, CGES, and TC-GES comparison
  • OECD PISA regime-stability study

Book-to-code crosswalk

The highlighted Experiment cards covered here

Implementation map

From entry point to source surface

Start here

These concrete files are selected from the complete registered source surface to orient the reader.

Author-declared entry points

  • artifacts/
  • kan_do_dcdi experiments

Resolved code

dependency · 6
  • kan-do-dcdi-archive:edge_scores.py
  • kan-do-dcdi-archive:edge_scores_fastpatch.py
  • kan-do-dcdi-archive:metrics.py
  • kan-do-dcdi-archive:mmd_utils.py
  • kan-do-dcdi-archive:rn_flow.py
  • kan-do-dcdi-archive:sheaf_metrics.py
runner · 25

Evidence record

Artifacts and frozen runs

Result artifacts

  • synthetic and Sachs result tables

Run records

No run-level packet is registered. The family retains code-level provenance, but public run reproduction still requires curation.

Evidence boundary

The mixed and negative Sachs results remain part of the evidence; runtime plots are descriptive implementation measurements.

This registry links the experiments reported by the three books to their executable or archived code surfaces. A code link establishes traceability, not independent reproduction, correctness, or support for a claim beyond the experiment's stated boundary.