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

rng = np.random.default_rng(seed=33)
dates = pd.date_range('2026-01-01', periods=365, freq='D')
day_of_year = np.arange(365)

seasonal_swing = 12 * np.sin(2 * np.pi * (day_of_year - 80) / 365)
avg_temp = 18 + seasonal_swing + rng.normal(0, 2.5, size=365)
avg_temp = avg_temp.round(1)
print(avg_temp[:5])

min_temp = (avg_temp - rng.uniform(3, 7, size=365)).round(1)
max_temp = (avg_temp + rng.uniform(3, 7, size=365)).round(1)
precipitation = np.where(rng.random(365) < 0.3, rng.exponential(6, size=365).round(1), 0.0)
humidity = np.clip(rng.normal(60, 12, size=365), 20, 100).round(0)

weather = pd.DataFrame({
    'date': dates,
    'avg_temp_c': avg_temp,
    'min_temp_c': min_temp,
    'max_temp_c': max_temp,
    'precipitation_mm': precipitation,
    'humidity_pct': humidity
})
print(weather.shape)

missing_idx = rng.choice(weather.index, size=8, replace=False)
weather.loc[missing_idx, 'avg_temp_c'] = np.nan
weather.to_csv('daily_weather.csv', index=False)
weather = pd.read_csv('daily_weather.csv', parse_dates=['date'])
weather.head()

print(weather.shape)
weather.info()

weather.isna().sum()

weather = weather.sort_values('date').reset_index(drop=True)
weather['avg_temp_c'] = weather['avg_temp_c'].interpolate()
weather['avg_temp_c'].isna().sum()

weather = weather.set_index('date')
weather.loc['2026-07'].head()

print(weather.loc['2026-07', 'avg_temp_c'].mean().round(1))
print(weather.loc['2026-01', 'avg_temp_c'].mean().round(1))

monthly = weather.resample('ME').agg(
    avg_temp_c=('avg_temp_c', 'mean'),
    total_precip_mm=('precipitation_mm', 'sum'),
    avg_humidity_pct=('humidity_pct', 'mean')
).round(1)
monthly

weather['rolling_7d_temp'] = weather['avg_temp_c'].rolling(window=7).mean()
weather[['avg_temp_c', 'rolling_7d_temp']].tail()

weather['is_extreme'] = (weather['max_temp_c'] > 28) | (weather['precipitation_mm'] > 15)
print(weather['is_extreme'].sum())
weather[weather['is_extreme']][['max_temp_c', 'precipitation_mm']].head()

weather_reset = weather.reset_index()
weather_reset['month'] = weather_reset['date'].dt.month_name()
monthly_avg_temp = weather_reset.groupby('month')['avg_temp_c'].mean().round(1)
monthly_avg_temp = monthly_avg_temp.reindex(['January', 'February', 'March', 'April', 'May', 'June', 'July', 'August', 'September', 'October', 'November', 'December'])
monthly_avg_temp

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

fig, ax = plt.subplots(figsize=(10, 5))
ax.plot(weather.index, weather['avg_temp_c'], color='gray', alpha=0.4, label='Daily avg temp')
ax.plot(weather.index, weather['rolling_7d_temp'], color='crimson', linewidth=2, label='7-day rolling average')
ax.set_title('Daily Temperature Across the Year')
ax.set_ylabel('Temperature (C)')
ax.legend()
plt.tight_layout()
plt.savefig('temperature_trend.png', dpi=150)
plt.close(fig)
print('Saved temperature_trend.png')

fig, ax = plt.subplots(figsize=(8, 5))
monthly['total_precip_mm'].plot(kind='bar', ax=ax, color='steelblue')
ax.set_title('Total Precipitation by Month')
ax.set_ylabel('Precipitation (mm)')
ax.set_xticklabels([d.strftime('%b') for d in monthly.index], rotation=45)
plt.tight_layout()
plt.savefig('monthly_precipitation.png', dpi=150)
plt.close(fig)
print('Saved monthly_precipitation.png')