Day 32 - Visualizing Data with Python

Visualizing Data with Pandas and Matplotlib

1. Introduction (10 minutes)

• Overview:
  • Recap the restaurant orders DataFrame and our computed columns from the previous lecture.
  • Today we will visualize our data to better understand restaurant performance.
• Learning Goals:
  • Learn to generate plots with Matplotlib.
  • Understand how to work with wide versus tall data.
  • Explore how to reformat data (using melt) for visualization.
• Do Now:
  • Ask: “Why is visualizing data important in data analysis? How might a chart help you see trends that raw numbers do not?”

2. Plotting with Matplotlib (20 minutes)

• Setting Up Matplotlib:
  • Example Code:

    import matplotlib.pyplot as plt
    # Ensure converters are registered for dates if needed
    from pandas.plotting import register_matplotlib_converters
    register_matplotlib_converters()
• Line Plot of Daily Total Sales:
  • First, compute total sales per day:

    daily_sales = orders.groupby('date')['total_price'].sum().reset_index()
    print(daily_sales)
  • Plotting:

    plt.figure()
    plt.plot(daily_sales['date'], daily_sales['total_price'], marker='o')
    plt.xlabel('Date')
    plt.ylabel('Total Sales ($)')
    plt.title('Daily Total Sales')
    plt.xticks(rotation=45)
    plt.tight_layout()
    plt.show()
• Bar Chart for Sales by Dish:
  • Example Code:

    dish_sales = orders.groupby('dish')['total_price'].sum().reset_index()
    plt.figure()
    plt.bar(dish_sales['dish'], dish_sales['total_price'], color='skyblue')
    plt.xlabel('Dish')
    plt.ylabel('Total Sales ($)')
    plt.title('Sales by Dish')
    plt.show()
• Interactive Exercise:
  • Ask: “Try modifying the plots—change markers, colors, or add gridlines. What effect do these changes have?”

3. Reshaping Data: Wide vs. Tall Format (15 minutes)

• Concepts:
  • Explain that data can be organized in different shapes. Wide data has many columns; tall data has more rows and fewer columns.
  • In Pandas, the melt function converts wide data into tall (long) format, which can simplify plotting.
• Example:
  • Suppose we have a wide table of weekly sales for different regions:

    wide_sales = pd.DataFrame({
        'week': ['Week1', 'Week2', 'Week3'],
        'north': [150, 200, 180],
        'south': [130, 170, 160],
        'east': [140, 210, 190],
        'west': [120, 160, 150]
    })
    print(wide_sales)
  • Convert to Tall Format:

    tall_sales = wide_sales.melt(id_vars=['week'], var_name='region', value_name='sales')
    print(tall_sales)
• Using the Tall Data for Plotting:
  • Example:
    Plot sales by region for each week:

    plt.figure()
    sns.barplot(x='week', y='sales', hue='region', data=tall_sales)
    plt.title('Weekly Sales by Region')
    plt.show()
• Discussion:
  • Ask students: “What are some advantages of having data in tall format for plotting?”

4. Wrap-Up (5 minutes)

• Recap Key Points:
  • We generated plots using Matplotlib to visualize daily sales and sales by dish.
  • We learned about wide vs. tall data and used the melt function to reformat data.
  • Visualizations help us communicate insights and detect trends that aren’t obvious from raw numbers.