eliminate schema_fit: it's clustering coefficient
schema_fit was algebraically identical to clustering_coefficient (both compute 2E/(d*(d-1)) = fraction of connected neighbor pairs). Remove the redundant function, field, and metrics column. - Delete schema_fit() and schema_fit_all() from graph.rs - Remove schema_fit field from Node struct - Remove avg_schema_fit from MetricsSnapshot (duplicated avg_cc) - Replace all callers with graph.clustering_coefficient() - Rename ReplayItem.schema_fit to .cc - Query: "cc" and "schema_fit" both resolve from graph CC - Low-CC count folded into health report CC line Co-Authored-By: Kent Overstreet <kent.overstreet@linux.dev>
This commit is contained in:
ProofOfConcept
parent
fb7aa46e03
commit
ec8b4b2ed2
5 changed files with 33 additions and 83 deletions
|
|
@ -145,8 +145,6 @@ pub struct Node {
|
|||
#[serde(default)]
|
||||
pub clustering_coefficient: Option<f32>,
|
||||
#[serde(default)]
|
||||
pub schema_fit: Option<f32>,
|
||||
#[serde(default)]
|
||||
pub degree: Option<u32>,
|
||||
}
|
||||
|
||||
|
|
@ -925,7 +923,6 @@ impl Store {
|
|||
position: 0,
|
||||
community_id: None,
|
||||
clustering_coefficient: None,
|
||||
schema_fit: None,
|
||||
degree: None,
|
||||
}
|
||||
}
|
||||
|
|
@ -1505,13 +1502,11 @@ impl Store {
|
|||
pub fn update_graph_metrics(&mut self) {
|
||||
let g = self.build_graph();
|
||||
let communities = g.communities();
|
||||
let fits = graph::schema_fit_all(&g);
|
||||
|
||||
for (key, node) in &mut self.nodes {
|
||||
node.community_id = communities.get(key).copied();
|
||||
node.clustering_coefficient = Some(g.clustering_coefficient(key));
|
||||
node.degree = Some(g.degree(key) as u32);
|
||||
node.schema_fit = fits.get(key).copied();
|
||||
}
|
||||
}
|
||||
|
||||
|
|
@ -1874,7 +1869,6 @@ fn read_content_node(r: memory_capnp::content_node::Reader) -> Result<Node, Stri
|
|||
position: r.get_position(),
|
||||
community_id: None,
|
||||
clustering_coefficient: None,
|
||||
schema_fit: None,
|
||||
degree: None,
|
||||
})
|
||||
}
|
||||
|
|
|
|||
39
src/graph.rs
39
src/graph.rs
|
|
@ -500,22 +500,6 @@ fn label_propagation(
|
|||
labels
|
||||
}
|
||||
|
||||
/// Schema fit for a node: how well-integrated into its local neighborhood.
|
||||
///
|
||||
/// Algebraically identical to clustering_coefficient — both measure the
|
||||
/// fraction of neighbor pairs that are connected. Kept as an alias for
|
||||
/// semantic clarity: CC is a graph metric, schema fit is a cognitive one
|
||||
/// (how well does this node fit the surrounding schema?).
|
||||
pub fn schema_fit(graph: &Graph, key: &str) -> f32 {
|
||||
graph.clustering_coefficient(key)
|
||||
}
|
||||
|
||||
/// Compute schema fit for all nodes
|
||||
pub fn schema_fit_all(graph: &Graph) -> HashMap<String, f32> {
|
||||
graph.nodes().iter()
|
||||
.map(|key| (key.clone(), graph.clustering_coefficient(key)))
|
||||
.collect()
|
||||
}
|
||||
|
||||
/// A snapshot of graph topology metrics, for tracking evolution over time
|
||||
#[derive(Clone, Debug, Serialize, Deserialize)]
|
||||
|
|
@ -530,7 +514,8 @@ pub struct MetricsSnapshot {
|
|||
pub gini: f32,
|
||||
pub avg_cc: f32,
|
||||
pub avg_path_length: f32,
|
||||
pub avg_schema_fit: f32,
|
||||
// Removed: avg_schema_fit was identical to avg_cc.
|
||||
// Old snapshots with the field still deserialize (serde ignores unknown fields by default).
|
||||
}
|
||||
|
||||
fn metrics_log_path() -> std::path::PathBuf {
|
||||
|
|
@ -595,12 +580,10 @@ pub fn health_report(graph: &Graph, store: &Store) -> String {
|
|||
let alpha = graph.degree_power_law_exponent();
|
||||
let gini = graph.degree_gini();
|
||||
|
||||
// Schema fit distribution
|
||||
let fits = schema_fit_all(graph);
|
||||
let avg_fit = if fits.is_empty() { 0.0 } else {
|
||||
fits.values().sum::<f32>() / fits.len() as f32
|
||||
};
|
||||
let low_fit = fits.values().filter(|&&f| f < 0.1).count();
|
||||
// Low-CC nodes: poorly integrated
|
||||
let low_cc = graph.nodes().iter()
|
||||
.filter(|k| graph.clustering_coefficient(k) < 0.1)
|
||||
.count();
|
||||
|
||||
// Category breakdown
|
||||
let cats = store.category_counts();
|
||||
|
|
@ -614,7 +597,6 @@ pub fn health_report(graph: &Graph, store: &Store) -> String {
|
|||
nodes: n, edges: e, communities,
|
||||
sigma, alpha, gini, avg_cc,
|
||||
avg_path_length: avg_pl,
|
||||
avg_schema_fit: avg_fit,
|
||||
};
|
||||
save_metrics_snapshot(&snap);
|
||||
|
||||
|
|
@ -641,7 +623,6 @@ pub fn health_report(graph: &Graph, store: &Store) -> String {
|
|||
let alpha_d = delta(alpha, prev.map(|p| p.alpha));
|
||||
let gini_d = delta(gini, prev.map(|p| p.gini));
|
||||
let cc_d = delta(avg_cc, prev.map(|p| p.avg_cc));
|
||||
let fit_d = delta(avg_fit, prev.map(|p| p.avg_schema_fit));
|
||||
|
||||
let mut report = format!(
|
||||
"Memory Health Report
|
||||
|
|
@ -649,15 +630,13 @@ pub fn health_report(graph: &Graph, store: &Store) -> String {
|
|||
Nodes: {n} Relations: {e} Communities: {communities}
|
||||
|
||||
Degree: max={max_deg} median={median_deg} avg={avg_deg:.1}
|
||||
Clustering coefficient (avg): {avg_cc:.4}{cc_d}
|
||||
Clustering coefficient (avg): {avg_cc:.4}{cc_d} low-CC (<0.1): {low_cc} nodes
|
||||
Average path length: {avg_pl:.2}
|
||||
Small-world σ: {sigma:.3}{sigma_d} (>1 = small-world)
|
||||
Power-law α: {alpha:.2}{alpha_d} (2=hub-dominated, 3=healthy, >3=egalitarian)
|
||||
Degree Gini: {gini:.3}{gini_d} (0=equal, 1=one-hub)
|
||||
|
||||
Community sizes (top 5): {top5}
|
||||
Schema fit: avg={avg_fit:.3}{fit_d} low-fit (<0.1): {low_fit} nodes
|
||||
|
||||
Categories: core={core} tech={tech} gen={gen} obs={obs} task={task}",
|
||||
top5 = sizes.iter().take(5)
|
||||
.map(|s| s.to_string())
|
||||
|
|
@ -674,8 +653,8 @@ Categories: core={core} tech={tech} gen={gen} obs={obs} task={task}",
|
|||
if history.len() >= 3 {
|
||||
report.push_str("\n\nMetrics history (last 5):\n");
|
||||
for snap in &history[history.len().saturating_sub(5)..] {
|
||||
report.push_str(&format!(" {} — σ={:.1} α={:.2} gini={:.3} cc={:.4} fit={:.3}\n",
|
||||
snap.date, snap.sigma, snap.alpha, snap.gini, snap.avg_cc, snap.avg_schema_fit));
|
||||
report.push_str(&format!(" {} — σ={:.1} α={:.2} gini={:.3} cc={:.4}\n",
|
||||
snap.date, snap.sigma, snap.alpha, snap.gini, snap.avg_cc));
|
||||
}
|
||||
}
|
||||
|
||||
|
|
|
|||
|
|
@ -221,7 +221,6 @@ pub fn migrate() -> Result<(), String> {
|
|||
position: 0,
|
||||
community_id: None,
|
||||
clustering_coefficient: None,
|
||||
schema_fit: None,
|
||||
degree: None,
|
||||
};
|
||||
all_nodes.push(node);
|
||||
|
|
@ -261,7 +260,6 @@ pub fn migrate() -> Result<(), String> {
|
|||
position: 0,
|
||||
community_id: None,
|
||||
clustering_coefficient: None,
|
||||
schema_fit: None,
|
||||
degree: None,
|
||||
};
|
||||
all_nodes.push(node);
|
||||
|
|
|
|||
64
src/neuro.rs
64
src/neuro.rs
|
|
@ -19,12 +19,12 @@ const SECS_PER_DAY: f64 = 86400.0;
|
|||
/// With spectral data:
|
||||
/// priority = spectral_displacement × overdue × emotion
|
||||
/// Without:
|
||||
/// priority = (1 - schema_fit) × overdue × emotion
|
||||
/// priority = (1 - cc) × overdue × emotion
|
||||
///
|
||||
/// Spectral displacement is the outlier_score clamped and normalized —
|
||||
/// it measures how far a node sits from its community center in the
|
||||
/// eigenspace. This is a global signal (considers all graph structure)
|
||||
/// vs schema_fit which is local (only immediate neighbors).
|
||||
/// vs CC which is local (only immediate neighbors).
|
||||
pub fn consolidation_priority(
|
||||
store: &Store,
|
||||
key: &str,
|
||||
|
|
@ -44,8 +44,8 @@ pub fn consolidation_priority(
|
|||
// outlier=5 and outlier=10 matters less than 1 vs 2.
|
||||
(outlier / 3.0).min(3.0)
|
||||
} else {
|
||||
let fit = graph::schema_fit(graph, key) as f64;
|
||||
1.0 - fit
|
||||
let cc = graph.clustering_coefficient(key) as f64;
|
||||
1.0 - cc
|
||||
};
|
||||
|
||||
// Spaced repetition: how overdue is this node for replay?
|
||||
|
|
@ -69,7 +69,7 @@ pub struct ReplayItem {
|
|||
pub priority: f64,
|
||||
pub interval_days: u32,
|
||||
pub emotion: f32,
|
||||
pub schema_fit: f32,
|
||||
pub cc: f32,
|
||||
/// Spectral classification: "bridge", "outlier", "core", "peripheral"
|
||||
pub classification: &'static str,
|
||||
/// Raw spectral outlier score (distance / median)
|
||||
|
|
@ -91,8 +91,6 @@ pub fn replay_queue_with_graph(
|
|||
graph: &Graph,
|
||||
emb: Option<&SpectralEmbedding>,
|
||||
) -> Vec<ReplayItem> {
|
||||
let fits = graph::schema_fit_all(graph);
|
||||
|
||||
// Build spectral position map if embedding is available
|
||||
let positions: HashMap<String, SpectralPosition> = if let Some(emb) = emb {
|
||||
let communities = graph.communities().clone();
|
||||
|
|
@ -116,14 +114,12 @@ pub fn replay_queue_with_graph(
|
|||
store, key, graph,
|
||||
pos.map(|p| p.outlier_score),
|
||||
);
|
||||
let fit = fits.get(key).copied().unwrap_or(0.0);
|
||||
|
||||
ReplayItem {
|
||||
key: key.clone(),
|
||||
priority,
|
||||
interval_days: node.spaced_repetition_interval,
|
||||
emotion: node.emotion,
|
||||
schema_fit: fit,
|
||||
cc: graph.clustering_coefficient(key),
|
||||
classification,
|
||||
outlier_score,
|
||||
}
|
||||
|
|
@ -178,7 +174,7 @@ pub fn detect_interference(
|
|||
/// Low fit (<0.2): deep examination needed — new schema seed, bridge, or noise?
|
||||
pub fn schema_assimilation(store: &Store, key: &str) -> (f32, &'static str) {
|
||||
let graph = store.build_graph();
|
||||
let fit = graph::schema_fit(&graph, key);
|
||||
let fit = graph.clustering_coefficient(key);
|
||||
|
||||
let recommendation = if fit > 0.5 {
|
||||
"auto-integrate"
|
||||
|
|
@ -267,8 +263,8 @@ fn format_nodes_section(store: &Store, items: &[ReplayItem], graph: &Graph) -> S
|
|||
};
|
||||
|
||||
out.push_str(&format!("## {} \n", item.key));
|
||||
out.push_str(&format!("Priority: {:.3} Schema fit: {:.3} Emotion: {:.1} ",
|
||||
item.priority, item.schema_fit, item.emotion));
|
||||
out.push_str(&format!("Priority: {:.3} CC: {:.3} Emotion: {:.1} ",
|
||||
item.priority, item.cc, item.emotion));
|
||||
out.push_str(&format!("Category: {} Interval: {}d\n",
|
||||
node.category.label(), node.spaced_repetition_interval));
|
||||
if item.outlier_score > 0.0 {
|
||||
|
|
@ -485,8 +481,8 @@ pub fn consolidation_batch(store: &Store, count: usize, auto: bool) -> Result<()
|
|||
let node_type = store.nodes.get(&item.key)
|
||||
.map(|n| if n.key.contains("journal") { "episodic" } else { "semantic" })
|
||||
.unwrap_or("?");
|
||||
println!(" [{:.3}] {} (fit={:.3}, interval={}d, type={})",
|
||||
item.priority, item.key, item.schema_fit, item.interval_days, node_type);
|
||||
println!(" [{:.3}] {} (cc={:.3}, interval={}d, type={})",
|
||||
item.priority, item.key, item.cc, item.interval_days, node_type);
|
||||
}
|
||||
|
||||
// Also show interference pairs
|
||||
|
|
@ -556,13 +552,12 @@ pub fn agent_prompt(store: &Store, agent: &str, count: usize) -> Result<String,
|
|||
let items: Vec<ReplayItem> = episode_keys.iter()
|
||||
.filter_map(|k| {
|
||||
let node = store.nodes.get(k)?;
|
||||
let fit = graph::schema_fit(&graph, k);
|
||||
Some(ReplayItem {
|
||||
key: k.clone(),
|
||||
priority: consolidation_priority(store, k, &graph, None),
|
||||
interval_days: node.spaced_repetition_interval,
|
||||
emotion: node.emotion,
|
||||
schema_fit: fit,
|
||||
cc: graph.clustering_coefficient(k),
|
||||
classification: "unknown",
|
||||
outlier_score: 0.0,
|
||||
})
|
||||
|
|
@ -597,12 +592,7 @@ pub fn consolidation_plan(store: &Store) -> ConsolidationPlan {
|
|||
let graph = store.build_graph();
|
||||
let alpha = graph.degree_power_law_exponent();
|
||||
let gini = graph.degree_gini();
|
||||
let avg_fit = {
|
||||
let fits = graph::schema_fit_all(&graph);
|
||||
if fits.is_empty() { 0.0 } else {
|
||||
fits.values().sum::<f32>() / fits.len() as f32
|
||||
}
|
||||
};
|
||||
let avg_cc = graph.avg_clustering_coefficient();
|
||||
let interference_pairs = detect_interference(store, &graph, 0.5);
|
||||
let interference_count = interference_pairs.len();
|
||||
|
||||
|
|
@ -651,17 +641,17 @@ pub fn consolidation_plan(store: &Store) -> ConsolidationPlan {
|
|||
gini));
|
||||
}
|
||||
|
||||
// Target: avg schema fit ≥ 0.2
|
||||
if avg_fit < 0.1 {
|
||||
// Target: avg CC ≥ 0.2
|
||||
if avg_cc < 0.1 {
|
||||
plan.replay_count += 5;
|
||||
plan.rationale.push(format!(
|
||||
"Schema fit={:.3} (target ≥0.2): very poor integration → +5 replay",
|
||||
avg_fit));
|
||||
} else if avg_fit < 0.2 {
|
||||
"CC={:.3} (target ≥0.2): very poor integration → +5 replay",
|
||||
avg_cc));
|
||||
} else if avg_cc < 0.2 {
|
||||
plan.replay_count += 2;
|
||||
plan.rationale.push(format!(
|
||||
"Schema fit={:.3} (target ≥0.2): low integration → +2 replay",
|
||||
avg_fit));
|
||||
"CC={:.3} (target ≥0.2): low integration → +2 replay",
|
||||
avg_cc));
|
||||
}
|
||||
|
||||
// Interference: >100 pairs is a lot, <10 is clean
|
||||
|
|
@ -748,21 +738,14 @@ pub fn daily_check(store: &Store) -> String {
|
|||
let gini = graph.degree_gini();
|
||||
let sigma = graph.small_world_sigma();
|
||||
let avg_cc = graph.avg_clustering_coefficient();
|
||||
let avg_fit = {
|
||||
let fits = graph::schema_fit_all(&graph);
|
||||
if fits.is_empty() { 0.0 } else {
|
||||
fits.values().sum::<f32>() / fits.len() as f32
|
||||
}
|
||||
};
|
||||
|
||||
let history = graph::load_metrics_history();
|
||||
let prev = history.last();
|
||||
|
||||
let mut out = String::from("Memory daily check\n");
|
||||
|
||||
// Current state
|
||||
out.push_str(&format!(" σ={:.1} α={:.2} gini={:.3} cc={:.4} fit={:.3}\n",
|
||||
sigma, alpha, gini, avg_cc, avg_fit));
|
||||
out.push_str(&format!(" σ={:.1} α={:.2} gini={:.3} cc={:.4}\n",
|
||||
sigma, alpha, gini, avg_cc));
|
||||
|
||||
// Trend
|
||||
if let Some(p) = prev {
|
||||
|
|
@ -777,7 +760,7 @@ pub fn daily_check(store: &Store) -> String {
|
|||
let mut issues = Vec::new();
|
||||
if alpha < 2.0 { issues.push("hub dominance critical"); }
|
||||
if gini > 0.5 { issues.push("high inequality"); }
|
||||
if avg_fit < 0.1 { issues.push("poor integration"); }
|
||||
if avg_cc < 0.1 { issues.push("poor integration"); }
|
||||
if d_sigma < -5.0 { issues.push("σ declining"); }
|
||||
if d_alpha < -0.1 { issues.push("α declining"); }
|
||||
if d_gini > 0.02 { issues.push("inequality increasing"); }
|
||||
|
|
@ -802,7 +785,6 @@ pub fn daily_check(store: &Store) -> String {
|
|||
communities: graph.community_count(),
|
||||
sigma, alpha, gini, avg_cc,
|
||||
avg_path_length: graph.avg_path_length(),
|
||||
avg_schema_fit: avg_fit,
|
||||
});
|
||||
|
||||
out
|
||||
|
|
|
|||
|
|
@ -184,12 +184,9 @@ fn resolve_field(field: &str, key: &str, store: &Store, graph: &Graph) -> Option
|
|||
"community_id" => {
|
||||
graph.communities().get(key).map(|&c| Value::Num(c as f64))
|
||||
}
|
||||
"clustering_coefficient" => {
|
||||
"clustering_coefficient" | "schema_fit" | "cc" => {
|
||||
Some(Value::Num(graph.clustering_coefficient(key) as f64))
|
||||
}
|
||||
"schema_fit" => {
|
||||
node.schema_fit.map(|f| Value::Num(f as f64))
|
||||
}
|
||||
_ => None,
|
||||
}
|
||||
}
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue