From 8f7b15bdf9bd5ad112cea0c6796d85dd84a6780e Mon Sep 17 00:00:00 2001
From: Jammy2211 <JNightingale2211@gmail.com>
Date: Sat, 11 Jul 2026 10:34:44 +0100
Subject: [PATCH] fix: regenerate navigator catalogue for
 expectation_propagation cross-ref

The EP #85 merge added scripts/features/expectation_propagation.py and a
cross-ref from graphical_models.py, but llms-full.txt / workspace_index.json
were never regenerated, turning the Navigator Check 'Catalogue staleness' job
RED on main. Regenerated with regenerate_navigator.py autofit (catalogue only;
idempotent; check_navigator green).

Closes #86

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---
 llms-full.txt        |  2 ++
 workspace_index.json | 27 ++++++++++++++++++++++++++-
 2 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/llms-full.txt b/llms-full.txt
index 9dd68c27..bbfc1da0 100644
--- a/llms-full.txt
+++ b/llms-full.txt
@@ -32,6 +32,8 @@ AUTO-GENERATED by PyAutoBuild — do not edit by hand; regenerate with generate.
 
 ## features
 
+- [Feature: Expectation Propagation (Step-by-Step)](scripts/features/expectation_propagation.py): The `graphical_models.py` feature example composed a graphical model from many datasets and fitted it with a single non-linear search over the joint parameter space. That approach hits a ceiling: with tens of datasets the joint parameter space becomes too high dimensional to sample.
+  - Contents: Example Source Code (`af.ex`), Dataset, Model, Factor Graph, Mean Field, Cavity, Tilted Fit, Damped Update, Convergence, Full EP Fit, Wrap Up
 - [Feature: Graphical Models](scripts/features/graphical_models.py): The examples so far have focused on fitting one model to one dataset, for example fitting 1D profiles composed of Gaussians to noisy 1D data. When multiple datasets were available each is fitted individually and their results interpreted one-by-one.
   - Contents: Example Source Code (`af.ex`), Dataset, Analysis, Model, Analysis Factors, Factor Graph, Search, Hierarchical Models, Expectation Propagation
 - [Feature: Interpolate](scripts/features/interpolate.py): It is common to fit a model to many similar datasets, where it is anticipated that one or more model parameters vary smoothly across the datasets.
diff --git a/workspace_index.json b/workspace_index.json
index a0c9349a..6dc5047d 100644
--- a/workspace_index.json
+++ b/workspace_index.json
@@ -237,6 +237,29 @@
     "summary": "This cookbook provides an overview of the non-linear searches available in **PyAutoFit**, and how to use them.",
     "title": "Cookbook: Searches"
   },
+  {
+    "contents": [
+      "Example Source Code (`af.ex`)",
+      "Dataset",
+      "Model",
+      "Factor Graph",
+      "Mean Field",
+      "Cavity",
+      "Tilted Fit",
+      "Damped Update",
+      "Convergence",
+      "Full EP Fit",
+      "Wrap Up"
+    ],
+    "cross_refs": [
+      "features/graphical_models.py",
+      "graphical_models.py"
+    ],
+    "notebook": null,
+    "path": "scripts/features/expectation_propagation.py",
+    "summary": "The `graphical_models.py` feature example composed a graphical model from many datasets and fitted it with a single non-linear search over the joint parameter space. That approach hits a ceiling: with tens of datasets the joint parameter space becomes too high dimensional to sample.",
+    "title": "Feature: Expectation Propagation (Step-by-Step)"
+  },
   {
     "contents": [
       "Example Source Code (`af.ex`)",
@@ -249,7 +272,9 @@
       "Hierarchical Models",
       "Expectation Propagation"
     ],
-    "cross_refs": [],
+    "cross_refs": [
+      "features/expectation_propagation.py"
+    ],
     "notebook": "notebooks/features/graphical_models.ipynb",
     "path": "scripts/features/graphical_models.py",
     "summary": "The examples so far have focused on fitting one model to one dataset, for example fitting 1D profiles composed of Gaussians to noisy 1D data. When multiple datasets were available each is fitted individually and their results interpreted one-by-one.",