Matplotlib Fundamentals: Creating Publication-Quality Plots

TIP

Learning Objectives: After this lesson, you'll master the fundamentals of data visualization with matplotlib—creating, customizing, and combining various plot types for effective data communication.

Why Matplotlib?

Matplotlib is the grandfather of Python visualization libraries. While newer libraries like seaborn and plotly offer convenience, understanding matplotlib gives you complete control over every pixel of your plots.

Basic Plotting

Line Plots

Scatter Plots

Bar Charts

Histograms

Pie Charts

Customizing Plots

Colors, Markers, and Line Styles

Titles, Labels, and Legends

Axis Customization

Subplots and Figure Layout

Multiple Plots in One Figure

Sharing Axes

Saving Figures

Object-Oriented Interface

The object-oriented approach gives you more control:

Practical Example: Complete Dashboard

Key Takeaways

✅ Basic plots: plt.plot(), plt.scatter(), plt.bar(), plt.hist(), plt.pie()

✅ Customization: Colors, markers, line styles, labels, titles, legends

✅ Subplots: Use plt.subplots() for multiple plots in one figure

✅ Object-oriented: fig, ax = plt.subplots() gives more control

✅ Saving: plt.savefig() with DPI and format options

✅ Layout: plt.tight_layout() prevents overlapping elements

Connections: Matplotlib in the Ecosystem

🔗 Connection to Other Libraries

🔗 Connection to Data Science

Visualization is essential for:

• EDA: Understanding data distributions
• Communication: Presenting findings
• Model evaluation: Learning curves, confusion matrices
• Publication: Journal-quality figures

Practice Exercises

Exercise 1: Multi-Plot Figure

Next Steps

In the next lesson, we'll explore seaborn—a high-level library that creates beautiful statistical visualizations with minimal code.

Ready for easier, more beautiful plots? Seaborn is next!

Further Reading

Visual Galleries (Steal Code from These)

• Matplotlib Gallery — every plot type with copy-pasteable source. The single most useful matplotlib resource.
• Python Graph Gallery — chart-type-first organization across matplotlib + seaborn + plotly.
• The Data Visualisation Catalogue — pick a chart type by what you're trying to communicate, then see the matplotlib equivalent.

Official Docs

• Matplotlib — Quick Start Guide — the official walkthrough.
• Matplotlib — pyplot vs Object-Oriented API — once you outgrow plt.plot, learn the OO fig, ax = plt.subplots() pattern.
• Matplotlib — Styling and Themes — rcParams, style sheets, themes.

Modern Best Practices

• Effective matplotlib (Practical Business Python) — Chris Moffitt. The post that taught everyone the OO API.
• Matplotlib Cyberpunk style — drop-in dark theme with neon glow effects. Surprisingly delightful.
• matplotx — modern theme set; matplotx.styles.dracula, nord, aura.

Beyond matplotlib

• Plotly Python — interactive, zoomable, exportable to standalone HTML.
• Altair — declarative grammar of graphics. Concise, principled, JSON-spec under the hood.
• Bokeh — for huge interactive dashboards.

Books

• Book: Python Data Science Handbook — Chapter 4. Free online.
• Book: Storytelling with Data — Cole Nussbaumer Knaflic. The why of visualization; transformative, library-agnostic.