generate_data_now.py

import pandas as pd
import random
import numpy as np

# Generate 2000 traffic records
random.seed(42)
np.random.seed(42)

data = []
for i in range(2000):
    road_id = random.randint(1, 20)
    hour = random.randint(0, 23)
    day_of_week = random.randint(0, 6)
    
    # Base traffic pattern
    base_traffic = 200 + (road_id * 15)
    
    # Rush hour factor
    if hour in [7, 8, 17, 18, 19]:
        rush_factor = 2.5
    elif hour in [9, 10, 16, 20]:
        rush_factor = 1.8
    elif hour in [11, 12, 13, 14, 15]:
        rush_factor = 1.3
    elif hour in [0, 1, 2, 3, 4, 5]:
        rush_factor = 0.3
    else:
        rush_factor = 0.8
    
    # Weekend factor
    weekend_factor = 0.7 if day_of_week >= 5 else 1.0
    
    # Calculate traffic with noise
    traffic = base_traffic * rush_factor * weekend_factor
    noise = random.gauss(0, traffic * 0.1)
    traffic = max(0, traffic + noise)
    
    data.append({
        'road_id': road_id,
        'time_of_day': hour,
        'day_of_week': day_of_week,
        'traffic_volume': round(traffic, 2)
    })

df = pd.DataFrame(data)
df.to_csv('data/traffic_data.csv', index=False)
print(f"Generated {len(df)} traffic records")
print(df.head())