Python Docs

Seaborn Tutorial

Seaborn is a Python data visualization library based on Matplotlib. It provides a high-level interface for drawing attractive and informative statistical graphics and makes it easy to create complex visualizations with minimal code.

Introduction to Seaborn

Seaborn builds on top of Matplotlib and integrates closely with Pandas data structures, making it ideal for statistical visualizations and quick exploratory analysis.

Key Features

Built-in themes for styling Matplotlib graphics
Tools for choosing color palettes
Functions for visualizing univariate and bivariate distributions
Functions for visualizing linear regression models
Tools for visualizing matrices and statistical time series
High-level abstractions for structuring multi-plot grids

Installation

pip install seaborn

Import Seaborn and common companions:

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

Getting Started

Setting Themes

# Set default theme
sns.set_theme()

# Set specific style
sns.set_style("darkgrid")  # Options: darkgrid, whitegrid, dark, white, ticks

# Set context (scaling)
sns.set_context("notebook")  # Options: paper, notebook, talk, poster

# Set color palette
sns.set_palette("husl")

Loading Sample Datasets

# Seaborn comes with built-in datasets
tips = sns.load_dataset('tips')
iris = sns.load_dataset('iris')
titanic = sns.load_dataset('titanic')
planets = sns.load_dataset('planets')

# View dataset
print(tips.head())

Distribution Plots

Histogram and KDE

tips = sns.load_dataset('tips')

# Histogram
sns.histplot(data=tips, x='total_bill', bins=20)
plt.title('Distribution of Total Bill')
plt.show()

# KDE plot
sns.kdeplot(data=tips, x='total_bill')
plt.title('KDE of Total Bill')
plt.show()

# Combined histogram and KDE
sns.histplot(data=tips, x='total_bill', kde=True)
plt.show()

Distribution Plot (displot)

# Create a figure with distribution
sns.displot(data=tips, x='total_bill', kde=True, bins=20)
plt.show()

# Multiple distributions
sns.displot(data=tips, x='total_bill', hue='sex', kde=True)
plt.show()

# Faceted distribution
sns.displot(data=tips, x='total_bill', col='time', kde=True)
plt.show()

Rug Plot

# Add rug plot to show individual observations
sns.kdeplot(data=tips, x='total_bill')
sns.rugplot(data=tips, x='total_bill')
plt.show()

Categorical Plots

Bar Plot

# Bar plot with error bars
sns.barplot(data=tips, x='day', y='total_bill')
plt.title('Average Bill by Day')
plt.show()

# With hue
sns.barplot(data=tips, x='day', y='total_bill', hue='sex')
plt.show()

Count Plot

# Count occurrences
sns.countplot(data=tips, x='day')
plt.title('Count of Visits by Day')
plt.show()

# With hue
sns.countplot(data=tips, x='day', hue='sex')
plt.show()

Box Plot

# Box plot
sns.boxplot(data=tips, x='day', y='total_bill')
plt.title('Bill Distribution by Day')
plt.show()

# With hue
sns.boxplot(data=tips, x='day', y='total_bill', hue='smoker')
plt.show()

Violin Plot

# Violin plot (combination of box and KDE)
sns.violinplot(data=tips, x='day', y='total_bill')
plt.show()

# Split violins
sns.violinplot(data=tips, x='day', y='total_bill', hue='sex', split=True)
plt.show()

Swarm Plot

# Swarm plot (shows all points)
sns.swarmplot(data=tips, x='day', y='total_bill')
plt.show()

# Combine with violin plot
sns.violinplot(data=tips, x='day', y='total_bill', inner=None, color='lightgray')
sns.swarmplot(data=tips, x='day', y='total_bill', size=2, color='black')
plt.show()

Relational Plots

Scatter Plot

# Basic scatter plot
sns.scatterplot(data=tips, x='total_bill', y='tip')
plt.show()

# With hue and size
sns.scatterplot(data=tips, x='total_bill', y='tip', hue='time', size='size')
plt.show()

# With style
sns.scatterplot(data=tips, x='total_bill', y='tip', hue='day', style='time')
plt.show()

Line Plot

# Line plot with confidence interval
fmri = sns.load_dataset('fmri')
sns.lineplot(data=fmri, x='timepoint', y='signal', hue='event')
plt.show()

# Multiple lines
sns.lineplot(data=fmri, x='timepoint', y='signal', hue='region', style='event')
plt.show()

relplot (Figure-level)

# Scatter plot with facets
sns.relplot(data=tips, x='total_bill', y='tip', col='time', hue='smoker')
plt.show()

# Line plot with facets
sns.relplot(data=fmri, x='timepoint', y='signal', hue='event', 
            col='region', kind='line')
plt.show()

Regression Plots

Linear Regression Plot

# Regression plot with confidence interval
sns.regplot(data=tips, x='total_bill', y='tip')
plt.show()

# Polynomial regression
sns.regplot(data=tips, x='total_bill', y='tip', order=2)
plt.show()

# Without confidence interval
sns.regplot(data=tips, x='total_bill', y='tip', ci=None)
plt.show()

Residual Plot

# Check model assumptions
sns.residplot(data=tips, x='total_bill', y='tip')
plt.axhline(0, color='red', linestyle='--')
plt.show()

lmplot (Figure-level)

# Regression with facets
sns.lmplot(data=tips, x='total_bill', y='tip', hue='smoker')
plt.show()

# With columns and rows
sns.lmplot(data=tips, x='total_bill', y='tip', col='time', row='sex')
plt.show()

Matrix Plots

Heatmap

# Correlation matrix
iris = sns.load_dataset('iris')
corr = iris.corr(numeric_only=True)

# Create heatmap
sns.heatmap(corr, annot=True, cmap='coolwarm', center=0)
plt.title('Correlation Heatmap')
plt.show()

# Custom color map and formatting
sns.heatmap(corr, annot=True, fmt='.2f', cmap='RdYlGn', 
            linewidths=0.5, cbar_kws={'label': 'Correlation'})
plt.show()

Clustermap

# Hierarchical clustering
sns.clustermap(corr, cmap='viridis', annot=True)
plt.show()

Pair Plots and Joint Plots

Pair Plot

# Pairwise relationships
iris = sns.load_dataset('iris')
sns.pairplot(iris)
plt.show()

# With hue
sns.pairplot(iris, hue='species')
plt.show()

# Custom diagonal and off-diagonal plots
sns.pairplot(iris, hue='species', diag_kind='kde', 
             plot_kws={'alpha': 0.6})
plt.show()

Joint Plot

# Joint distribution
sns.jointplot(data=tips, x='total_bill', y='tip')
plt.show()

# With regression
sns.jointplot(data=tips, x='total_bill', y='tip', kind='reg')
plt.show()

# Hexbin for large datasets
sns.jointplot(data=tips, x='total_bill', y='tip', kind='hex')
plt.show()

# KDE
sns.jointplot(data=tips, x='total_bill', y='tip', kind='kde')
plt.show()

Figure-Level vs Axes-Level Functions

Axes-Level (for subplots)

# Create multiple plots in one figure
fig, axes = plt.subplots(2, 2, figsize=(12, 10))

sns.scatterplot(data=tips, x='total_bill', y='tip', ax=axes[0, 0])
axes[0, 0].set_title('Scatter Plot')

sns.boxplot(data=tips, x='day', y='total_bill', ax=axes[0, 1])
axes[0, 1].set_title('Box Plot')

sns.violinplot(data=tips, x='day', y='total_bill', ax=axes[1, 0])
axes[1, 0].set_title('Violin Plot')

sns.barplot(data=tips, x='day', y='total_bill', ax=axes[1, 1])
axes[1, 1].set_title('Bar Plot')

plt.tight_layout()
plt.show()

Figure-Level (creates own figure)

# Figure-level functions create their own figure
# relplot, displot, catplot, lmplot, pairplot, jointplot

# Example: catplot
sns.catplot(data=tips, x='day', y='total_bill', hue='sex', 
            kind='bar', col='time', height=5)
plt.show()

Color Palettes

# View available palettes
sns.color_palette()

# Categorical palettes
sns.set_palette("husl")
sns.set_palette("Set2")
sns.set_palette("Paired")

# Sequential palettes
sns.set_palette("Blues")
sns.set_palette("rocket")

# Diverging palettes
sns.set_palette("coolwarm")
sns.set_palette("vlag")

# Custom palette
custom_palette = ["#FF6B6B", "#4ECDC4", "#45B7D1"]
sns.set_palette(custom_palette)

# Show palette
sns.palplot(sns.color_palette("husl", 8))
plt.show()

Customization

Styling

# Set style
sns.set_style("whitegrid")

# Temporarily set style
with sns.axes_style("darkgrid"):
    sns.scatterplot(data=tips, x='total_bill', y='tip')
    plt.show()

# Remove spines
sns.despine()

# Custom parameters
sns.set_theme(rc={
    'axes.facecolor': '#f5f5f5',
    'figure.facecolor': 'white',
    'axes.edgecolor': 'black',
    'grid.color': 'white'
})

Context Scaling

# For different media
sns.set_context("paper")     # Smallest
sns.set_context("notebook")  # Default
sns.set_context("talk")      # For presentations
sns.set_context("poster")    # Largest

# Custom scaling
sns.set_context("notebook", font_scale=1.5, rc={"lines.linewidth": 2.5})

Practical Example: Comprehensive Data Analysis

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd

# Load dataset
tips = sns.load_dataset('tips')

# Set style
sns.set_theme(style="whitegrid", context="notebook")

# Create comprehensive analysis figure
fig = plt.figure(figsize=(16, 12))

# 1. Distribution of total bill
ax1 = plt.subplot(3, 3, 1)
sns.histplot(data=tips, x='total_bill', kde=True, ax=ax1)
ax1.set_title('Total Bill Distribution', fontweight='bold')

# 2. Tip distribution by day
ax2 = plt.subplot(3, 3, 2)
sns.boxplot(data=tips, x='day', y='tip', ax=ax2)
ax2.set_title('Tips by Day', fontweight='bold')

# 3. Bill by time of day
ax3 = plt.subplot(3, 3, 3)
sns.violinplot(data=tips, x='time', y='total_bill', hue='sex', split=True, ax=ax3)
ax3.set_title('Bill by Time & Gender', fontweight='bold')

# 4. Tip percentage by party size
ax4 = plt.subplot(3, 3, 4)
tips['tip_pct'] = tips['tip'] / tips['total_bill'] * 100
sns.barplot(data=tips, x='size', y='tip_pct', ax=ax4)
ax4.set_title('Tip % by Party Size', fontweight='bold')

# 5. Regression: Bill vs Tip
ax5 = plt.subplot(3, 3, 5)
sns.regplot(data=tips, x='total_bill', y='tip', ax=ax5, scatter_kws={'alpha':0.5})
ax5.set_title('Bill vs Tip Regression', fontweight='bold')

# 6. Count of visits
ax6 = plt.subplot(3, 3, 6)
sns.countplot(data=tips, x='day', hue='time', ax=ax6)
ax6.set_title('Visit Counts', fontweight='bold')

# 7. Smoker vs Non-smoker bills
ax7 = plt.subplot(3, 3, 7)
sns.swarmplot(data=tips, x='smoker', y='total_bill', hue='sex', ax=ax7)
ax7.set_title('Bills: Smoker vs Non-smoker', fontweight='bold')

# 8. Correlation heatmap
ax8 = plt.subplot(3, 3, 8)
corr_data = tips[['total_bill', 'tip', 'size', 'tip_pct']].corr()
sns.heatmap(corr_data, annot=True, fmt='.2f', cmap='coolwarm', ax=ax8)
ax8.set_title('Correlation Matrix', fontweight='bold')

# 9. Tip by day and time
ax9 = plt.subplot(3, 3, 9)
pivot_data = tips.pivot_table(values='tip', index='day', columns='time', aggfunc='mean')
sns.heatmap(pivot_data, annot=True, fmt='.2f', cmap='YlOrRd', ax=ax9)
ax9.set_title('Avg Tip by Day & Time', fontweight='bold')

plt.suptitle('Restaurant Tips Analysis Dashboard', 
             fontsize=16, fontweight='bold', y=0.995)
plt.tight_layout()
plt.savefig('tips_analysis.png', dpi=300, bbox_inches='tight')
plt.show()

Best Practices

Use appropriate plot types: Choose plots that best represent your data.
Leverage hue and style: Add dimensions to your visualizations.
Set themes early: Apply styling at the beginning of your script.
Use figure-level functions for faceting: They handle subplot creation automatically.
Choose colorblind-friendly palettes: Ensure accessibility.
Add context with titles and labels: Make plots self-explanatory.

Summary

Seaborn simplifies the creation of beautiful statistical visualizations in Python. Built on top of Matplotlib, it provides intuitive high-level functions that make complex plots easy to create. Master Seaborn to quickly explore and communicate insights from your data.