Module 6 Lesson 9: Seaborn: Statistical Visualization

Matplotlib is great for basics, but for advanced statistics, we use Seaborn. Built on top of Matplotlib, Seaborn makes your charts look professional and "publication-ready" out of the box. It understands Pandas DataFrames natively, making it a favorite for data scientists.

Lesson Overview

In this lesson, we will cover:

• The Power of Themes: Instant beauty with sns.set_theme().
• Aesthetic Plots: Boxplots, Violin plots, and Heatmaps.
• Categorical Data: Visualizing groups and distributions.
• Simplifying Complexity: Multi-plot grids.

---

1. Setting the Mood

Seaborn automatically makes your charts look better with its default themes.

import seaborn as sns
import matplotlib.pyplot as plt

# Set the style to 'darkgrid' or 'whitegrid'
sns.set_theme(style="darkgrid")

# Load a built-in dataset for testing
tips = sns.load_dataset("tips")

# Create a simple Scatter Plot
sns.scatterplot(data=tips, x="total_bill", y="tip", hue="day")
plt.show()

(Note: hue automatically colors the points based on a category—like 'day'!)

---

2. Visualizing Distributions (Boxplots)

If you want to see how spread out your data is and find "outliers" (weird values), use a Boxplot.

sns.boxplot(data=tips, x="day", y="total_bill")
plt.show()

---

3. The Famous Heatmap

Heatmaps are the best way to visualize "Correlation"—how much two variables relate to each other.

# Calculate correlation between numerical columns
corr = tips.select_dtypes('number').corr()

sns.heatmap(corr, annot=True, cmap="coolwarm")
plt.show()

---

4. Why Seaborn?

1. Less Code: What takes 10 lines in Matplotlib often takes 1 line in Seaborn.
2. Pandas Friendly: You don't have to extract lists; just pass the DataFrame and the column names.
3. Beautiful Colors: It comes with sophisticated color palettes (husl, viridis, magma).

---

Practice Exercise: The Fitness Tracker Viz

1. Load a dataset (or create one) with columns: Day, Steps, and Mood (Happy, Tired, Neutral).
2. Use Seaborn to create a Bar Plot of Day vs Steps.
3. Add hue="Mood" to see if your mood correlates with how much you walked.
4. Change the color palette to "magma" (hint: palette="magma").

---

Quick Knowledge Check

1. What library is Seaborn built on top of?
2. What does the hue parameter do in Seaborn?
3. Which plot is better for seeing the density of data: a Boxplot or a Scatter plot?
4. How do you apply a global style to all your charts at once?

---

Key Takeaways

• Seaborn is for high-level, beautiful statistical visualization.
• It works perfectly with Pandas DataFrames.
• hue is a powerful tool for adding a 3rd dimension to a 2D chart.
• Use Seaborn when you want to look like a pro with minimal effort.

---

What’s Next?

You’ve learned the tools. Now it's time to do some real detective work! In Lesson 10, we’ll start our Exploratory Data Analysis (EDA) Project, where we’ll take a raw dataset and find the hidden stories inside it!