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

rng = np.random.default_rng(seed=11)

stores = ['Downtown', 'Westside', 'Northgate', 'Eastview']
category_products = {
    'Grocery': ['Bread', 'Milk', 'Eggs', 'Coffee', 'Rice'],
    'Electronics': ['Headphones', 'Charger', 'Mouse', 'Webcam'],
    'Apparel': ['T-Shirt', 'Jeans', 'Jacket', 'Socks'],
    'Home & Garden': ['Lamp', 'Planter', 'Rug', 'Tool Kit'],
    'Toys': ['Puzzle', 'Board Game', 'Action Figure']
}
categories = list(category_products.keys())
print(stores)
print(categories)

n_rows = 400
dates = pd.date_range('2026-01-01', periods=90, freq='D')

rows = []
for _ in range(n_rows):
    date = rng.choice(dates)
    store = rng.choice(stores)
    category = rng.choice(categories)
    product = rng.choice(category_products[category])
    quantity = int(rng.integers(1, 6))
    unit_price = round(float(rng.uniform(3, 120)), 2)
    rows.append([date, store, category, product, quantity, unit_price])

sales = pd.DataFrame(rows, columns=['date', 'store', 'category', 'product', 'quantity', 'unit_price'])
sales.insert(0, 'order_id', range(1001, 1001 + len(sales)))
print(sales.shape)

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

print(sales.shape)
print(sales.dtypes)

sales.info()

sales.describe()

sales['store'].value_counts()

sales['revenue'] = sales['quantity'] * sales['unit_price']
sales[['product', 'quantity', 'unit_price', 'revenue']].head()

print(f"Total revenue: ${sales['revenue'].sum():,.2f}")
print(f"Average order value: ${sales['revenue'].mean():,.2f}")

big_orders = sales[sales['revenue'] > 300]
print(len(big_orders))
big_orders.sort_values('revenue', ascending=False).head()

electronics_orders = sales[(sales['category'] == 'Electronics') & (sales['store'] == 'Downtown')]
electronics_orders.head()

revenue_by_category = sales.groupby('category')['revenue'].sum().sort_values(ascending=False)
revenue_by_category

revenue_by_store = sales.groupby('store')['revenue'].sum().sort_values(ascending=False)
revenue_by_store

category_summary = sales.groupby('category').agg(
    total_revenue=('revenue', 'sum'),
    avg_unit_price=('unit_price', 'mean'),
    orders=('revenue', 'count')
).sort_values('total_revenue', ascending=False)
category_summary

pivot = pd.pivot_table(sales, values='revenue', index='store', columns='category', aggfunc='sum', fill_value=0)
pivot

top_products = sales.groupby('product')['revenue'].sum().nlargest(5)
top_products

daily_revenue = sales.groupby('date')['revenue'].sum()
print(daily_revenue.head())
print(daily_revenue.sort_values(ascending=False).head(3))

rolling_avg = daily_revenue.rolling(window=7).mean()
print(rolling_avg.tail())

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

fig, ax = plt.subplots(figsize=(8, 5))
revenue_by_category.plot(kind='bar', ax=ax, color='steelblue')
ax.set_title('Total Revenue by Category')
ax.set_ylabel('Revenue ($)')
plt.tight_layout()
plt.savefig('revenue_by_category.png', dpi=150)
plt.close(fig)
print('Saved revenue_by_category.png')

fig, ax = plt.subplots(figsize=(9, 5))
daily_revenue.plot(ax=ax, color='gray', alpha=0.5, label='Daily revenue')
rolling_avg.plot(ax=ax, color='crimson', linewidth=2, label='7-day rolling average')
ax.set_title('Daily Revenue with 7-Day Rolling Average')
ax.set_ylabel('Revenue ($)')
ax.legend()
plt.tight_layout()
plt.savefig('daily_revenue_trend.png', dpi=150)
plt.close(fig)
print('Saved daily_revenue_trend.png')