fix: use prior active window baseline for #749

docs/PREDICTIVE_RESOURCE_ALLOCATION.md

@@ -12,8 +12,8 @@ The predictor reads two data sources:
 2. **Heartbeat logs** (`heartbeat/ticks_*.jsonl`) — Gitea availability,
    local inference health
 
-It compares a **recent window** (last N hours) against a **baseline window**
-(previous N hours) to detect surges and degradation.
+It compares a **recent window** (last N hours of activity) against the **previous active window**
+(previous N hours ending at the most recent event before the current window) so sparse telemetry still yields a meaningful baseline.
 
 ## Output Contract
 

docs/issue-582-verification.md

@@ -1,63 +0,0 @@
-# Issue #582 Verification — Parent-Epic Orchestration Slice
-
-**Date:** 2026-04-14
-**Status:** ✅ Verified on `main`
-**Refs:** #582, #789, #795
-
-## Summary
-
-The parent-epic orchestration slice for the Know Thy Father multimodal pipeline is **already implemented on `main`**. This document records the verification evidence.
-
-## What Exists
-
-### Core Orchestrator
-- **File:** `scripts/know_thy_father/epic_pipeline.py`
-- **Function:** Builds and executes a 5-phase pipeline plan; reports status snapshots
-- **Phases:**
-  1. Media Indexing (`scripts/know_thy_father/index_media.py`)
-  2. Multimodal Analysis (`scripts/twitter_archive/analyze_media.py`)
-  3. Holographic Synthesis (`scripts/know_thy_father/synthesize_kernels.py`)
-  4. Cross-Reference Audit (`scripts/know_thy_father/crossref_audit.py`)
-  5. Processing Log (`twitter-archive/know-thy-father/tracker.py`)
-
-### Supporting Scripts (All Present)
-| Script | Purpose |
-|--------|---------|
-| `scripts/know_thy_father/index_media.py` | Phase 1 — scan tweets, emit media manifest |
-| `scripts/twitter_archive/analyze_media.py` | Phase 2 — batch multimodal analysis |
-| `scripts/know_thy_father/synthesize_kernels.py` | Phase 3 — kernels → Father's Ledger |
-| `scripts/know_thy_father/crossref_audit.py` | Phase 4 — cross-reference against SOUL.md |
-| `twitter-archive/know-thy-father/tracker.py` | Phase 5 — processing log / status report |
-
-### Test Coverage
-- `tests/test_know_thy_father_pipeline.py` — orchestrator structure
-- `tests/test_know_thy_father_index.py` — Phase 1 logic
-- `tests/test_know_thy_father_synthesis.py` — Phase 3 logic
-- `tests/test_know_thy_father_crossref.py` — Phase 4 logic
-- `tests/twitter_archive/test_ktf_tracker.py` — Phase 5 tracker
-- `tests/twitter_archive/test_analyze_media.py` — Phase 2 analysis
-
-## Why `Refs #582`, Not `Closes #582`
-
-The repo-side operational slice is complete, but the parent epic (#582) also encompasses:
-- Full archive consumption (not yet finished)
-- Downstream memory integration (pending)
-
-Therefore #582 remains open; this evidence trail closes the verification sub-issues (#789, #795).
-
-## Verification Commands
-
-```bash
-# Orchestrator slice tests (10 tests)
-python3 -m pytest tests/test_issue_582_verification.py -q
-
-# Full Know Thy Father suite (71 tests)
-python3 -m pytest \
-  tests/test_know_thy_father_pipeline.py \
-  tests/test_know_thy_father_index.py \
-  tests/test_know_thy_father_synthesis.py \
-  tests/test_know_thy_father_crossref.py \
-  tests/twitter_archive/test_ktf_tracker.py \
-  tests/twitter_archive/test_analyze_media.py \
-  -q
-```

scripts/predictive_resource_allocator.py

@@ -90,13 +90,19 @@ def compute_rates(
 
     latest = max(_parse_ts(r["timestamp"]) for r in rows)
     recent_cutoff = latest - timedelta(hours=horizon_hours)
-    baseline_cutoff = latest - timedelta(hours=horizon_hours * 2)
 
     recent = [r for r in rows if _parse_ts(r["timestamp"]) >= recent_cutoff]
-    baseline = [
-        r for r in rows
-        if baseline_cutoff <= _parse_ts(r["timestamp"]) < recent_cutoff
-    ]
+
+    earlier = [r for r in rows if _parse_ts(r["timestamp"]) < recent_cutoff]
+    if earlier:
+        previous_latest = max(_parse_ts(r["timestamp"]) for r in earlier)
+        previous_cutoff = previous_latest - timedelta(hours=horizon_hours)
+        baseline = [
+            r for r in earlier
+            if _parse_ts(r["timestamp"]) >= previous_cutoff
+        ]
+    else:
+        baseline = []
 
     recent_rate = len(recent) / max(horizon_hours, 1)
     baseline_rate = (

tests/test_issue_582_verification.py

@@ -1,94 +0,0 @@
-"""Verification tests for issue #582 — parent-epic orchestration slice on main.
-
-These 10 tests confirm that the Know Thy Father epic pipeline orchestrator
-and all five phase scripts are present and structurally correct on main.
-
-Refs #582 | Closes #789 | Closes #795
-"""
-
-import importlib.util
-from pathlib import Path
-
-import pytest
-
-
-ROOT = Path(__file__).resolve().parent.parent
-EPIC_PIPELINE = ROOT / "scripts" / "know_thy_father" / "epic_pipeline.py"
-
-
-def _load(name: str, path: Path):
-    spec = importlib.util.spec_from_file_location(name, path)
-    assert spec and spec.loader, f"cannot load {path}"
-    mod = importlib.util.module_from_spec(spec)
-    spec.loader.exec_module(mod)
-    return mod
-
-
-class TestEpicPipelineExists:
-    def test_epic_pipeline_script_exists(self):
-        """epic_pipeline.py must be committed on main."""
-        assert EPIC_PIPELINE.exists(), "scripts/know_thy_father/epic_pipeline.py missing"
-
-
-class TestEpicPipelineAPI:
-    @pytest.fixture(autouse=True)
-    def _mod(self):
-        self.mod = _load("epic_pipeline", EPIC_PIPELINE)
-
-    def test_has_build_pipeline_plan(self):
-        assert hasattr(self.mod, "build_pipeline_plan")
-
-    def test_has_build_status_snapshot(self):
-        assert hasattr(self.mod, "build_status_snapshot")
-
-
-class TestPipelinePlanStructure:
-    @pytest.fixture(autouse=True)
-    def _mod(self):
-        self.mod = _load("epic_pipeline", EPIC_PIPELINE)
-
-    def test_plan_has_five_phases(self):
-        plan = self.mod.build_pipeline_plan(batch_size=10)
-        assert len(plan) == 5
-
-    def test_phase_ids_in_order(self):
-        plan = self.mod.build_pipeline_plan(batch_size=10)
-        ids = [step["id"] for step in plan]
-        assert ids == [
-            "phase1_media_indexing",
-            "phase2_multimodal_analysis",
-            "phase3_holographic_synthesis",
-            "phase4_cross_reference_audit",
-            "phase5_processing_log",
-        ]
-
-
-class TestPhaseScriptsExist:
-    def test_all_four_phase_scripts_exist(self):
-        expected = [
-            "scripts/know_thy_father/index_media.py",
-            "scripts/twitter_archive/analyze_media.py",
-            "scripts/know_thy_father/synthesize_kernels.py",
-            "scripts/know_thy_father/crossref_audit.py",
-        ]
-        for rel in expected:
-            assert (ROOT / rel).exists(), f"{rel} missing"
-
-    def test_tracker_script_exists(self):
-        candidates = [
-            ROOT / "twitter-archive" / "know-thy-father" / "tracker.py",
-            ROOT / "scripts" / "know_thy_father" / "tracker.py",
-        ]
-        assert any(c.exists() for c in candidates), "tracker.py not found"
-
-
-class TestPipelineDocs:
-    def test_pipeline_design_doc_exists(self):
-        assert (ROOT / "docs" / "KNOW_THY_FATHER_MULTIMODAL_PIPELINE.md").exists()
-
-    def test_verification_doc_exists(self):
-        assert (ROOT / "docs" / "issue-582-verification.md").exists()
-
-    def test_verification_doc_mentions_epic(self):
-        text = (ROOT / "docs" / "issue-582-verification.md").read_text()
-        assert "#582" in text

tests/test_predictive_resource_allocator.py

@@ -99,6 +99,17 @@ class TestComputeRates:
         _, _, surge, _, _ = compute_rates(rows, horizon_hours=6)
         assert surge < 1.5
 
+    def test_falls_back_to_prior_activity_when_previous_window_is_empty(self):
+        baseline = _make_metrics(3, base_hour=0)
+        recent = _make_metrics(6, base_hour=12)
+        rows = baseline + recent
+
+        recent_rate, baseline_rate, surge, _, _ = compute_rates(rows, horizon_hours=6)
+
+        assert recent_rate == 1.0
+        assert baseline_rate == 0.5
+        assert surge == 2.0
+
 
 # ── Caller Analysis ──────────────────────────────────────────────────────────