Mic validation dataset

import random
import csv
from datetime import datetime, timedelta

def generate_minting_events(n_events=1000, n_actors=50):
    """Generate synthetic minting events for testing"""

    SHARD_WEIGHTS = {
        'reflection': 1.0,
        'learning': 1.0,
        'civic': 1.5,
        'stability': 2.0,
        'stewardship': 2.0,
        'innovation': 2.5,
        'guardian': 3.0
    }

    THRESHOLD = 0.95
    ALPHA = 1.0
    KS_PER_MIC = 1_000_000

    events = []
    start_time = datetime(2025, 12, 1, 8, 0, 0)

    for i in range(n_events):
        actor_id = f"CIVIC_{random.randint(1, n_actors):03d}"
        shard_type = random.choice(list(SHARD_WEIGHTS.keys()))
        shard_weight = SHARD_WEIGHTS[shard_type]

        # Generate MII with bias toward threshold
        mii_before = random.gauss(0.94, 0.03)
        mii_before = max(0.80, min(1.0, mii_before))

        # Action improves MII
        improvement = random.uniform(0.01, 0.05)
        mii_after = min(1.0, mii_before + improvement)

        quality = random.uniform(0.5, 1.0)

        # Calculate MIC
        if mii_after >= THRESHOLD:
            delta = mii_after - THRESHOLD
            mic = ALPHA * delta * shard_weight * quality
        else:
            mic = 0.0

        ks = round(mic * KS_PER_MIC)

        events.append({
            'event_id': f'E{i+1:06d}',
            'timestamp': (start_time + timedelta(minutes=i*5)).isoformat() + 'Z',
            'actor_id': actor_id,
            'action_type': shard_type,
            'mii_before': round(mii_before, 6),
            'mii_after': round(mii_after, 6),
            'shard_type': shard_type,
            'shard_weight': shard_weight,
            'mic_minted': round(mic, 6),
            'ks_minted': ks,
            'quality_score': round(quality, 2)
        })

    return events

def save_to_csv(events, filename='minting_events.csv'):
    """Save events to CSV file"""
    if not events:
        return

    with open(filename, 'w', newline='') as f:
        writer = csv.DictWriter(f, fieldnames=events[0].keys())
        writer.writeheader()
        writer.writerows(events)

if __name__ == '__main__':
    events = generate_minting_events(1000, 50)
    save_to_csv(events)
    print(f"Generated {len(events)} events")