Scruff AI
57 lines
1.8 KiB
Python
57 lines
1.8 KiB
Python
#!/usr/bin/env python3
|
|
import json, glob, statistics
|
|
|
|
files = glob.glob('/mnt/d/fractal-brain/beast-build/sweep_results/run_*_status.json')
|
|
data = []
|
|
for f in files:
|
|
try:
|
|
with open(f) as fp:
|
|
d = json.load(fp)
|
|
if 'asymmetry' in d and 'coherence' in d:
|
|
data.append({'asymmetry': d['asymmetry'], 'coherence': d['coherence']})
|
|
except:
|
|
pass
|
|
|
|
if not data:
|
|
print("No data")
|
|
exit()
|
|
|
|
asym = [d['asymmetry'] for d in data]
|
|
coh = [d['coherence'] for d in data]
|
|
|
|
print("="*60)
|
|
print("SWEEP DATA SUMMARY")
|
|
print("="*60)
|
|
print(f"Total samples: {len(data)}")
|
|
print(f"\nAsymmetry:")
|
|
print(f" Range: {min(asym):.4f} - {max(asym):.4f}")
|
|
print(f" Mean: {statistics.mean(asym):.4f}")
|
|
print(f" Median: {statistics.median(asym):.4f}")
|
|
print(f"\nCoherence:")
|
|
print(f" Range: {min(coh):.4f} - {max(coh):.4f}")
|
|
print(f" Mean: {statistics.mean(coh):.4f}")
|
|
print(f" Median: {statistics.median(coh):.4f}")
|
|
|
|
print(f"\nDistribution by asymmetry bands:")
|
|
for threshold in [13.2, 13.4, 13.6, 13.8, 14.0, 14.2, 14.4]:
|
|
count = sum(1 for a in asym if threshold <= a < threshold + 0.2)
|
|
pct = 100 * count / len(asym)
|
|
print(f" {threshold:.1f}-{threshold+0.2:.1f}: {count:4d} samples ({pct:.1f}%)")
|
|
|
|
# Find peaks
|
|
print(f"\nTop 10 asymmetry values:")
|
|
sorted_asym = sorted(asym, reverse=True)
|
|
for i, a in enumerate(sorted_asym[:10]):
|
|
print(f" {i+1}. {a:.4f}")
|
|
|
|
print(f"\nCorrelation (asymmetry vs coherence):")
|
|
if len(asym) > 1:
|
|
mean_a = statistics.mean(asym)
|
|
mean_c = statistics.mean(coh)
|
|
cov = sum((a - mean_a) * (c - mean_c) for a, c in zip(asym, coh)) / len(asym)
|
|
var_a = statistics.variance(asym) if len(asym) > 1 else 0
|
|
var_c = statistics.variance(coh) if len(coh) > 1 else 0
|
|
if var_a > 0 and var_c > 0:
|
|
corr = cov / (var_a * var_c) ** 0.5
|
|
print(f" Pearson r: {corr:.4f}")
|