Saving Plots as Images

Why saving matters

Saved plots are used in:

• Reports
• Notebooks
• Websites
• PowerPoint/Docs

McKinney’s original explanation is refreshingly simple: fig.savefig(path) writes the active figure to disk, and matplotlib infers the file type from the extension you give it — .png for a raster image, .svg/.pdf for vector formats that stay sharp at any zoom level.

diagram Choosing a save format mermaid

  flowchart LR
  A["fig.savefig(path)"] --> B{"File extension"}
  B -->|".png"| C["Raster image - web, slides"]
  B -->|".pdf / .svg"| D["Vector image - print, infinite zoom"]
  C --> E["dpi controls resolution"]
  D --> F["Scales without pixelation"]

The file extension you pass to savefig() decides the output format and how it should be used.

Save to PNG

savefig

import matplotlib.pyplot as plt
 
plt.figure(figsize=(7, 4))
plt.plot([1, 2, 3], [3, 2, 5])
plt.title("Saved plot")
 
plt.savefig("plot.png", dpi=150, bbox_inches="tight")

Notes

• dpi controls resolution
• bbox_inches="tight" avoids cutting off labels

Save to PDF (for print)

Save PDF

import matplotlib.pyplot as plt
 
plt.plot([1, 2, 3], [3, 2, 5])
plt.title("PDF plot")
plt.savefig("plot.pdf", bbox_inches="tight")

From the book

McKinney flags dpi as the option he uses most often: it’s the dots-per-inch resolution, and it defaults to a fairly low value out of the box (100 in IPython, 72 in Jupyter). For a publication-quality PNG, he reaches for fig.savefig("path.png", dpi=400).

Next

You’ve covered every core chart type — continue to the Matplotlib Mini Project to combine them into one EDA charts pack.

🧪 Try It Yourself

Exercise 1 – Save a PNG at higher DPI

pch.exerciseTag pch.exerciseTitle

Exercise 2 – Infer format from the extension

pch.exerciseTag pch.exerciseTitle

Exercise 3 – Avoid clipped labels

pch.exerciseTag pch.exerciseTitle