Matplotlib vs Seaborn — Python Visualization Guide | DataPath

📊

Matplotlib vs Seaborn — Quick Overview

Python has two dominant visualization libraries for data analysis: Matplotlib and Seaborn. They're not competitors — Seaborn is built on top of Matplotlib. Here's the relationship:

Matplotlib is the low-level foundation:

Fine-grained control over every element (colors, fonts, axes, legends)
Verbose — requires more code for styled charts
Works with NumPy arrays and basic Python data structures
Best for custom visualizations and publication-ready figures

Seaborn is the high-level interface:

Beautiful defaults out of the box
Built for statistical visualizations (distributions, relationships, categories)
Designed for Pandas DataFrames
Less code for common chart types

The 80/20 Rule for Analysts

Use Seaborn for:

Exploratory data analysis (quick insights)
Statistical plots (distributions, correlations, regressions)
Multi-plot grids (faceted charts)
When you want charts that look good immediately

Use Matplotlib when:

You need pixel-perfect customization
You're creating charts Seaborn doesn't offer
You're fine-tuning a Seaborn plot (Seaborn returns Matplotlib objects)
You're building dashboards or custom visualizations

Think of it this way...

Matplotlib is a blank canvas with brushes. Seaborn is a paint-by-numbers kit with beautiful defaults. Most analysts start with Seaborn (fast, pretty), then customize with Matplotlib (control, precision).

📈

Creating Basic Charts — Side by Side

Let's compare the same chart in both libraries to see the difference in code and output.

Bar Chart: Revenue by City

Matplotlib:

code.pyPython

import matplotlib.pyplot as plt
import pandas as pd

# Sample data
cities = ['Mumbai', 'Delhi', 'Bangalore', 'Chennai']
revenue = [450000, 380000, 350000, 280000]

# Create chart
plt.figure(figsize=(10, 6))
plt.bar(cities, revenue, color='steelblue')
plt.xlabel('City', fontsize=12)
plt.ylabel('Revenue (₹)', fontsize=12)
plt.title('Revenue by City', fontsize=14, fontweight='bold')
plt.xticks(rotation=45)
plt.grid(axis='y', alpha=0.3)
plt.tight_layout()
plt.show()

Seaborn:

code.pyPython

import seaborn as sns
import pandas as pd

# Sample data
df = pd.DataFrame({
    'city': ['Mumbai', 'Delhi', 'Bangalore', 'Chennai'],
    'revenue': [450000, 380000, 350000, 280000]
})

# Create chart (fewer lines, beautiful by default)
sns.set_style('whitegrid')
plt.figure(figsize=(10, 6))
sns.barplot(data=df, x='city', y='revenue', palette='Blues_d')
plt.xlabel('City', fontsize=12)
plt.ylabel('Revenue (₹)', fontsize=12)
plt.title('Revenue by City', fontsize=14, fontweight='bold')
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

Key Differences:

Seaborn requires DataFrame (more natural for analysis)
Seaborn's default styling is more polished
Matplotlib has slightly less code for simple bar charts
Both require similar customization for labels and titles

Line Chart: Daily Orders Over Time

Matplotlib:

code.pyPython

dates = pd.date_range('2026-03-01', periods=10)
orders = [120, 135, 128, 145, 150, 142, 138, 155, 160, 148]

plt.figure(figsize=(12, 6))
plt.plot(dates, orders, marker='o', linewidth=2, color='#2E86AB')
plt.xlabel('Date', fontsize=12)
plt.ylabel('Orders', fontsize=12)
plt.title('Daily Orders — March 2026', fontsize=14)
plt.xticks(rotation=45)
plt.grid(True, alpha=0.3)
plt.tight_layout()
plt.show()

Seaborn:

code.pyPython

df = pd.DataFrame({'date': dates, 'orders': orders})

sns.set_style('darkgrid')
plt.figure(figsize=(12, 6))
sns.lineplot(data=df, x='date', y='orders', marker='o', linewidth=2)
plt.xlabel('Date', fontsize=12)
plt.ylabel('Orders', fontsize=12)
plt.title('Daily Orders — March 2026', fontsize=14)
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()

Scatter Plot: Delivery Time vs Rating

Matplotlib:

code.pyPython

delivery_time = [25, 30, 35, 40, 45, 50, 55, 60]
rating = [4.8, 4.7, 4.5, 4.3, 4.0, 3.8, 3.5, 3.2]

plt.figure(figsize=(10, 6))
plt.scatter(delivery_time, rating, s=100, alpha=0.6, color='coral')
plt.xlabel('Delivery Time (minutes)', fontsize=12)
plt.ylabel('Customer Rating', fontsize=12)
plt.title('Delivery Time vs Customer Rating', fontsize=14)
plt.grid(True, alpha=0.3)
plt.show()

Seaborn (with regression line):

code.pyPython

df = pd.DataFrame({'delivery_time': delivery_time, 'rating': rating})

plt.figure(figsize=(10, 6))
sns.regplot(data=df, x='delivery_time', y='rating', scatter_kws={'s': 100, 'alpha': 0.6})
plt.xlabel('Delivery Time (minutes)', fontsize=12)
plt.ylabel('Customer Rating', fontsize=12)
plt.title('Delivery Time vs Customer Rating', fontsize=14)
plt.show()

Notice: Seaborn's regplot() automatically adds a regression line and confidence interval — showing the negative correlation without extra code.

⚠️ CheckpointQuiz error: Missing or invalid options array

📊

Where Seaborn Excels — Statistical Plots

Seaborn shines for statistical visualizations that would require significant code in pure Matplotlib.

Distribution Plots

code.pyPython

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

# PhonePe transaction amounts
np.random.seed(42)
amounts = np.random.lognormal(7, 1.5, 10000)  # Right-skewed distribution

# Histogram with KDE (kernel density estimate)
plt.figure(figsize=(12, 5))

plt.subplot(1, 2, 1)
sns.histplot(amounts, bins=50, kde=True, color='skyblue')
plt.xlabel('Transaction Amount (₹)')
plt.title('Distribution of Transaction Amounts')

plt.subplot(1, 2, 2)
sns.boxplot(y=amounts, color='lightcoral')
plt.ylabel('Transaction Amount (₹)')
plt.title('Boxplot — Outliers and Quartiles')

plt.tight_layout()
plt.show()

Categorical Plots

code.pyPython

# Flipkart orders by category and city
df = pd.DataFrame({
    'city': ['Mumbai', 'Mumbai', 'Delhi', 'Delhi', 'Bangalore', 'Bangalore'] * 50,
    'category': ['Electronics', 'Fashion', 'Electronics', 'Fashion', 'Electronics', 'Fashion'] * 50,
    'orders': np.random.randint(50, 300, 300)
})

# Grouped bar chart with error bars
plt.figure(figsize=(10, 6))
sns.barplot(data=df, x='city', y='orders', hue='category', errorbar='sd', palette='Set2')
plt.title('Orders by City and Category')
plt.xlabel('City')
plt.ylabel('Average Orders')
plt.legend(title='Category')
plt.show()

Correlation Heatmap

code.pyPython

# Zomato restaurant metrics
df = pd.DataFrame({
    'avg_rating': np.random.uniform(3.5, 4.8, 100),
    'delivery_time': np.random.randint(20, 60, 100),
    'order_count': np.random.randint(50, 500, 100),
    'price_range': np.random.randint(1, 5, 100)
})

# Correlation matrix
corr_matrix = df.corr()

plt.figure(figsize=(8, 6))
sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', center=0, fmt='.2f', linewidths=1)
plt.title('Correlation Matrix — Restaurant Metrics')
plt.show()

Pair Plot (Multi-Variable Relationships)

code.pyPython

# Visualize all pairwise relationships
sns.pairplot(df, diag_kind='kde', corner=True)
plt.suptitle('Pairwise Relationships — Zomato Metrics', y=1.01)
plt.show()

Why Seaborn Wins Here: These plots require 50+ lines in Matplotlib. Seaborn does them in 1-2 lines with beautiful styling.

🎨

Where Matplotlib Excels — Customization

Matplotlib gives you pixel-level control over every element. Use it when Seaborn's defaults aren't enough.

Multiple Subplots with Custom Layouts

code.pyPython

# Dashboard-style layout
fig, axes = plt.subplots(2, 2, figsize=(14, 10))

# Top-left: Line chart
axes[0, 0].plot(dates, orders, color='#2E86AB', linewidth=2)
axes[0, 0].set_title('Daily Orders', fontweight='bold')
axes[0, 0].grid(True, alpha=0.3)

# Top-right: Bar chart
axes[0, 1].bar(cities, revenue, color=['#FF6B6B', '#4ECDC4', '#45B7D1', '#FFA07A'])
axes[0, 1].set_title('Revenue by City', fontweight='bold')

# Bottom-left: Scatter
axes[1, 0].scatter(delivery_time, rating, s=100, alpha=0.6, c=delivery_time, cmap='RdYlGn_r')
axes[1, 0].set_title('Delivery Time vs Rating', fontweight='bold')

# Bottom-right: Histogram
axes[1, 1].hist(amounts[:1000], bins=30, color='mediumpurple', alpha=0.7)
axes[1, 1].set_title('Transaction Distribution', fontweight='bold')

plt.tight_layout()
plt.show()

Annotations and Custom Text

code.pyPython

plt.figure(figsize=(12, 6))
plt.plot(dates, orders, marker='o', linewidth=2, color='#2E86AB')

# Annotate the highest point
max_idx = orders.index(max(orders))
plt.annotate(f'Peak: {max(orders)} orders',
             xy=(dates[max_idx], orders[max_idx]),
             xytext=(dates[max_idx] + pd.Timedelta(days=1), orders[max_idx] + 5),
             arrowprops=dict(arrowstyle='->', color='red', lw=2),
             fontsize=11, color='red', fontweight='bold')

plt.xlabel('Date')
plt.ylabel('Orders')
plt.title('Daily Orders with Peak Annotation')
plt.grid(True, alpha=0.3)
plt.show()

Dual Y-Axis (Two Metrics, One Chart)

code.pyPython

# Orders and revenue on the same chart
fig, ax1 = plt.subplots(figsize=(12, 6))

# First y-axis: Orders
ax1.plot(dates, orders, color='#2E86AB', marker='o', linewidth=2, label='Orders')
ax1.set_xlabel('Date', fontsize=12)
ax1.set_ylabel('Orders', color='#2E86AB', fontsize=12)
ax1.tick_params(axis='y', labelcolor='#2E86AB')

# Second y-axis: Revenue
ax2 = ax1.twinx()
revenue_daily = [o * 350 for o in orders]  # Assume ₹350 per order
ax2.plot(dates, revenue_daily, color='#FF6B6B', marker='s', linewidth=2, label='Revenue')
ax2.set_ylabel('Revenue (₹)', color='#FF6B6B', fontsize=12)
ax2.tick_params(axis='y', labelcolor='#FF6B6B')

plt.title('Daily Orders and Revenue', fontsize=14, fontweight='bold')
fig.tight_layout()
plt.show()

Info

Pro Tip: Start with Seaborn for quick charts. When you need to customize, use plt.gca() to access the underlying Matplotlib axes object and apply custom styling.

🤝

Combining Both — Best of Both Worlds

You don't have to choose one or the other. Use Seaborn for the chart, Matplotlib for customization.

Example: Seaborn Chart + Matplotlib Customization

code.pyPython

# Create a Seaborn chart
df = pd.DataFrame({
    'city': ['Mumbai', 'Delhi', 'Bangalore', 'Chennai'] * 3,
    'month': ['Jan', 'Jan', 'Jan', 'Jan', 'Feb', 'Feb', 'Feb', 'Feb', 'Mar', 'Mar', 'Mar', 'Mar'],
    'revenue': [45000, 38000, 35000, 28000, 48000, 40000, 37000, 30000, 52000, 43000, 39000, 32000]
})

plt.figure(figsize=(12, 6))

# Seaborn does the heavy lifting
sns.barplot(data=df, x='city', y='revenue', hue='month', palette='viridis')

# Matplotlib adds custom touches
plt.title('Q1 2026 Revenue by City and Month', fontsize=16, fontweight='bold', pad=20)
plt.xlabel('City', fontsize=13)
plt.ylabel('Revenue (₹)', fontsize=13)
plt.legend(title='Month', fontsize=11, title_fontsize=12, loc='upper right')
plt.axhline(40000, color='red', linestyle='--', linewidth=1.5, alpha=0.7, label='Target: ₹40K')
plt.text(3.5, 41000, 'Target', color='red', fontsize=10, fontweight='bold')

plt.tight_layout()
plt.show()

Setting Global Styles

code.pyPython

# Set Seaborn style globally
sns.set_style('whitegrid')
sns.set_palette('husl')
sns.set_context('talk')  # Larger fonts for presentations

# Now all plots (Seaborn or Matplotlib) use these defaults

Available Styles:

darkgrid — dark background, white grid
whitegrid — white background, gray grid
dark — dark background, no grid
white — white background, no grid
ticks — white background with ticks

Contexts (font sizes):

paper — smallest (for papers)
notebook — default
talk — larger (for presentations)
poster — largest (for posters)

⚠️ FinalQuiz error: Missing or invalid questions array

⚠️ SummarySection error: Missing or invalid items array

Received: {"hasItems":false,"isArray":false}

Matplotlib vs Seaborn — Which Visualization Library to Use?

Matplotlib vs Seaborn — Quick Overview

The 80/20 Rule for Analysts

Creating Basic Charts — Side by Side

Bar Chart: Revenue by City

Line Chart: Daily Orders Over Time

Scatter Plot: Delivery Time vs Rating

Where Seaborn Excels — Statistical Plots

Distribution Plots

Categorical Plots

Correlation Heatmap

Pair Plot (Multi-Variable Relationships)

Where Matplotlib Excels — Customization

Multiple Subplots with Custom Layouts

Annotations and Custom Text

Dual Y-Axis (Two Metrics, One Chart)

Combining Both — Best of Both Worlds

Example: Seaborn Chart + Matplotlib Customization

Setting Global Styles