File size: 7,003 Bytes
74f2af5 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 | """
Phase 6 + Consciousness Stack Baseline Benchmark
Quick benchmark WITHOUT requiring Llama model or full server.
Tests core improvement metrics:
- Semantic tension quality
- Specialization tracking
- Conflict detection
- State vector consistency
"""
import sys
import json
import time
from pathlib import Path
sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
from framework_definitions import StateVector, CoherenceMetrics, TensionDefinition
from semantic_tension import SemanticTensionEngine
from specialization_tracker import SpecializationTracker
print("\n" + "="*70)
print("PHASE 6 + CONSCIOUSNESS STACK QUICK BASELINE")
print("="*70)
# Test 1: Framework initialization
print("\n[TEST 1] Framework Initialization")
start = time.time()
state = StateVector(psi=0.8, tau=0.6, chi=1.2, phi=0.3, lam=0.7)
t1 = time.time() - start
print(f" StateVector creation: {t1*1000:.2f}ms")
# Test 2: Semantic tension computation
print("\n[TEST 2] Semantic Tension Computation")
engine = SemanticTensionEngine()
test_pairs = [
("Newton's laws are fundamental.", "Quantum mechanics reveals indeterminacy."),
("The universe is deterministic.", "Quantum effects introduce randomness."),
("Consciousness is an illusion.", "Consciousness is fundamental to reality."),
("AI will surpass human intelligence.", "AI and humans will collaborate."),
("Ethics are absolute.", "Ethics are relative and contextual."),
]
tensions = []
start = time.time()
for claim_a, claim_b in test_pairs:
tension = engine.compute_semantic_tension(claim_a, claim_b)
polarity = engine.compute_polarity(claim_a, claim_b)
tensions.append({
"claim_a": claim_a[:40],
"claim_b": claim_b[:40],
"semantic_tension": round(tension, 3),
"polarity": polarity
})
t2 = time.time() - start
print(f" {len(test_pairs)} pairs processed in {t2*1000:.2f}ms ({t2/len(test_pairs)*1000:.1f}ms per pair)")
for i, t in enumerate(tensions, 1):
print(f" [{i}] Tension={t['semantic_tension']}, Polarity={t['polarity']}")
# Test 3: Specialization tracking
print("\n[TEST 3] Specialization Tracking")
tracker = SpecializationTracker()
# Simulate 5 adapters, 3 domains each
adapters = ["Newton", "Quantum", "Ethics", "Creativity", "Systems"]
domains_per_adapter = 3
samples_per_domain = 4
start = time.time()
total_recordings = 0
for adapter in adapters:
queries = [
f"Physics query about {adapter}",
f"Ethics question for {adapter}",
f"Systems analysis with {adapter}",
]
for query in queries:
for _ in range(samples_per_domain):
coherence = 0.75 + (hash(f"{adapter}{query}") % 100) / 500 # Random 0.75-0.95
tracker.record_adapter_performance(adapter, query, coherence)
total_recordings += 1
t3 = time.time() - start
print(f" {total_recordings} recordings in {t3*1000:.2f}ms")
# Compute specialization
specialization_scores = {}
for adapter in adapters:
spec = tracker.compute_specialization(adapter)
specialization_scores[adapter] = spec
print(f" {adapter}: {spec}")
# Test 4: Coherence metrics
print("\n[TEST 4] System Coherence Metrics")
test_states = [
("Healthy", (0.75, 0.65, 0.3, 0.6)),
("Collapsing", (0.1, 0.2, 0.9, 0.05)),
("Groupthinking", (0.95, 0.95, 0.0, 0.95)),
("Balanced", (0.6, 0.6, 0.4, 0.6)),
]
start = time.time()
health_results = []
for name, (div, tension, variance, resolution) in test_states:
gamma, health = CoherenceMetrics.compute_gamma(div, tension, variance, resolution)
health_results.append({
"name": name,
"gamma": round(gamma, 3),
"health": health,
"diversity": div,
"tension_health": tension
})
t4 = time.time() - start
print(f" {len(test_states)} states computed in {t4*1000:.2f}ms")
for h in health_results:
print(f" {h['name']}: gamma={h['gamma']}, health={h['health']}")
# Test 5: State distance (structural tension)
print("\n[TEST 5] State Space Distance (Structural Tension)")
state_a = StateVector(psi=0.8, tau=0.6, chi=1.2, phi=0.3, lam=0.7)
state_b = StateVector(psi=0.5, tau=0.7, chi=0.8, phi=-0.2, lam=0.6)
state_c = StateVector(psi=0.1, tau=0.1, chi=0.1, phi=0.1, lam=0.1)
start = time.time()
dist_ab = StateVector.distance(state_a, state_b)
dist_ac = StateVector.distance(state_a, state_c)
dist_bc = StateVector.distance(state_b, state_c)
t5 = time.time() - start
print(f" 3 distances computed in {t5*1000:.2f}ms")
print(f" State A-B distance: {dist_ab:.3f}")
print(f" State A-C distance: {dist_ac:.3f}")
print(f" State B-C distance: {dist_bc:.3f}")
# SUMMARY
print("\n" + "="*70)
print("BASELINE RESULTS SUMMARY")
print("="*70)
summary = {
"tests_run": 5,
"total_time_ms": (t1+t2+t3+t4+t5)*1000,
"tests": {
"framework_init_ms": t1*1000,
"semantic_tension_ms": t2*1000,
"specialization_ms": t3*1000,
"coherence_ms": t4*1000,
"state_distance_ms": t5*1000,
},
"results": {
"semantic_tensions": tensions,
"specialization": {k: str(v) for k, v in specialization_scores.items()},
"coherence": health_results,
"state_distances": {
"A-B": dist_ab,
"A-C": dist_ac,
"B-C": dist_bc,
}
}
}
print(f"Total execution time: {summary['total_time_ms']:.1f}ms")
print(f" - Framework init: {t1*1000:.2f}ms")
print(f" - Semantic tension: {t2*1000:.2f}ms")
print(f" - Specialization: {t3*1000:.2f}ms")
print(f" - Coherence metrics: {t4*1000:.2f}ms")
print(f" - State distance: {t5*1000:.2f}ms")
# Save results
with open('phase6_baseline_results.json', 'w') as f:
# Convert non-serializable objects
summary_clean = {
"tests_run": summary["tests_run"],
"total_time_ms": summary["total_time_ms"],
"tests": summary["tests"],
"results": {
"semantic_tensions": summary["results"]["semantic_tensions"],
"coherence": summary["results"]["coherence"],
"state_distances": {str(k): float(v) for k, v in summary["results"]["state_distances"].items()},
"note": "Specialization scores stored as string due to nested dict structure"
}
}
json.dump(summary_clean, f, indent=2)
print("\nResults saved to: phase6_baseline_results.json")
print("\n" + "="*70)
print("QUALITY METRICS")
print("="*70)
print(f"Average semantic tension: {sum(t['semantic_tension'] for t in tensions)/len(tensions):.3f}")
print(f"Min/Max semantic tension: {min(t['semantic_tension'] for t in tensions):.3f} - {max(t['semantic_tension'] for t in tensions):.3f}")
print(f"Coherence (Healthy) Gamma: {health_results[0]['gamma']:.3f}")
print(f"Coherence (Collapsing) Gamma: {health_results[1]['gamma']:.3f}")
print(f"Coherence (Groupthink) Gamma: {health_results[2]['gamma']:.3f}")
print(f"Max structural tension: {max(dist_ab, dist_ac, dist_bc):.3f}")
print("\n✅ Phase 6 baseline benchmark complete!")
print("="*70 + "\n")
|