import pandas as pd
import numpy as np
print(pd.__version__)

rng = np.random.default_rng(seed=77)
dates = pd.date_range('2026-02-01', periods=180, freq='D')
is_weekend = dates.dayofweek >= 5

active_minutes = np.where(is_weekend, rng.normal(70, 25, size=180), rng.normal(45, 18, size=180))
active_minutes = np.clip(active_minutes, 5, None).round(0)
print(active_minutes[:5])

steps = (active_minutes * rng.uniform(90, 130, size=180) + rng.normal(0, 800, size=180)).round(0).astype(int)
steps = np.clip(steps, 500, None)
calories_burned = (1600 + steps * 0.04 + active_minutes * 3 + rng.normal(0, 80, size=180)).round(0)
print(steps[:5])

sleep_hours = np.clip(rng.normal(7.0, 1.0, size=180), 3.5, 10.5).round(1)
resting_heart_rate = (68 - (sleep_hours - 7) * 2 + rng.normal(0, 3, size=180)).round(0).astype(int)

activity = pd.DataFrame({
    'date': dates,
    'steps': steps,
    'active_minutes': active_minutes,
    'calories_burned': calories_burned,
    'sleep_hours': sleep_hours,
    'resting_heart_rate': resting_heart_rate
})
print(activity.shape)

activity.to_csv('fitness_log.csv', index=False)
activity = pd.read_csv('fitness_log.csv', parse_dates=['date'])
activity.head()

print(activity.shape)
activity.info()

activity[['steps', 'active_minutes', 'calories_burned', 'sleep_hours', 'resting_heart_rate']].describe().round(1)

activity['weekday'] = activity['date'].dt.day_name()
activity['is_weekend'] = activity['date'].dt.dayofweek >= 5
activity[['date', 'weekday', 'is_weekend']].head(3)

weekly = activity.set_index('date').resample('W').agg(
    total_steps=('steps', 'sum'),
    avg_sleep=('sleep_hours', 'mean'),
    avg_resting_hr=('resting_heart_rate', 'mean')
).round(1)
weekly.head()

activity['steps_rolling_7d'] = activity['steps'].rolling(window=7).mean().round(0)
activity[['date', 'steps', 'steps_rolling_7d']].tail()

activity['hit_goal'] = activity['steps'] >= 8000
print(f"Days goal was hit: {activity['hit_goal'].sum()} out of {len(activity)}")
print(f"Goal hit rate: {activity['hit_goal'].mean() * 100:.1f}%")

streak_id = (activity['hit_goal'] != activity['hit_goal'].shift()).cumsum()
streak_lengths = activity[activity['hit_goal']].groupby(streak_id).size()
print(f'Longest goal streak: {streak_lengths.max()} days')

numeric_cols = ['steps', 'active_minutes', 'calories_burned', 'sleep_hours', 'resting_heart_rate']
activity[numeric_cols].corr().round(2)

weekday_comparison = activity.groupby('is_weekend').agg(
    avg_steps=('steps', 'mean'),
    avg_active_minutes=('active_minutes', 'mean'),
    avg_sleep=('sleep_hours', 'mean')
).round(1)
weekday_comparison.index = ['Weekday', 'Weekend']
weekday_comparison

import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

fig, ax = plt.subplots(figsize=(10, 5))
ax.plot(activity['date'], activity['steps'], color='gray', alpha=0.4, label='Daily steps')
ax.plot(activity['date'], activity['steps_rolling_7d'], color='seagreen', linewidth=2, label='7-day rolling average')
ax.axhline(8000, color='crimson', linestyle='--', linewidth=1, label='Goal (8,000)')
ax.set_title('Daily Steps Over Six Months')
ax.set_ylabel('Steps')
ax.legend()
plt.tight_layout()
plt.savefig('steps_trend.png', dpi=150)
plt.close(fig)
print('Saved steps_trend.png')

fig, ax = plt.subplots(figsize=(8, 5))
ax.scatter(activity['sleep_hours'], activity['resting_heart_rate'], alpha=0.5, color='mediumpurple')
ax.set_title('Sleep Hours vs. Resting Heart Rate')
ax.set_xlabel('Sleep Hours')
ax.set_ylabel('Resting Heart Rate (bpm)')
plt.tight_layout()
plt.savefig('sleep_vs_heart_rate.png', dpi=150)
plt.close(fig)
print('Saved sleep_vs_heart_rate.png')