Why Figure size and DPI in Matplotlib? - Purpose & Use Cases

The Big Idea

What if your charts could always look perfect no matter where you show them?

The Scenario

Imagine you want to create a chart to share with your team. You draw it by hand or use a basic tool without control over size or clarity. When you print or show it on different screens, the chart looks blurry or parts get cut off.

The Problem

Manually adjusting images or charts without knowing about figure size and DPI means you waste time guessing the right dimensions. The output can be blurry, too small, or too large, making your work look unprofessional and hard to understand.

The Solution

Using figure size and DPI settings in matplotlib lets you control exactly how big and sharp your charts are. This means your visuals look clear on any screen or print, saving time and making your data easy to understand.

Before vs After

✗ Before

plt.plot(data)
plt.show()

✓ After

plt.figure(figsize=(8,6), dpi=100)
plt.plot(data)
plt.show()

What It Enables

You can create crisp, perfectly sized charts that fit any presentation or report, making your data story clear and professional.

Real Life Example

A data analyst preparing a sales report can set figure size and DPI to ensure charts fit neatly on slides and print clearly in handouts, impressing the audience with sharp visuals.

Key Takeaways

Manual chart sizing often leads to blurry or poorly fitting visuals.

Figure size and DPI settings give precise control over chart dimensions and clarity.

This control helps create professional, easy-to-read data visuals for any use.

Practice

1. What does the figsize parameter control in a matplotlib figure?

easy

A. The physical size of the figure in inches

B. The color of the figure background

C. The resolution of the figure in dots per inch

D. The font size of the plot labels

Solution

1. Step 1: Understand the role of figsize

   The figsize parameter sets the width and height of the figure in inches.

2. Step 2: Differentiate from DPI

   DPI controls resolution, not size. Color and font size are unrelated to figsize.

3. Final Answer:

   The physical size of the figure in inches -> Option A

4. Quick Check:

   Figure size = physical dimensions [OK]

Hint: Remember: figsize = inches size, DPI = sharpness [OK]

Common Mistakes:

• Confusing DPI with figure size
• Thinking figsize controls resolution
• Mixing up color or font size with figsize

2. Which of the following is the correct way to create a matplotlib figure with size 8 inches wide and 4 inches tall?

easy

A. plt.figure(dpi=(8, 4))

B. plt.figure(size=(8, 4))

C. plt.figure(figsize=(8, 4))

D. plt.figure(shape=(8, 4))

Solution

1. Step 1: Recall correct parameter name

   The parameter to set figure size is figsize, which takes a tuple (width, height).

2. Step 2: Check syntax correctness

   Only plt.figure(figsize=(8, 4)) uses figsize correctly. Others use invalid parameter names.

3. Final Answer:

   plt.figure(figsize=(8, 4)) -> Option C

4. Quick Check:

   Use figsize=(width, height) [OK]

Hint: Use figsize=(width, height) tuple in plt.figure() [OK]

Common Mistakes:

• Using wrong parameter names like size or shape
• Passing DPI values instead of size
• Not using a tuple for figsize

3. What will be the DPI of the figure created by this code?

fig = plt.figure(figsize=(6, 3), dpi=150)

medium

A. 100

B. 150

C. 6

D. 3

Solution

1. Step 1: Identify the DPI parameter value

   The code explicitly sets dpi=150 when creating the figure.

2. Step 2: Understand DPI meaning

   DPI means dots per inch, so the figure resolution is 150 dots per inch.

3. Final Answer:

   150 -> Option B

4. Quick Check:

   DPI parameter sets resolution [OK]

Hint: DPI is set by dpi= value in plt.figure() [OK]

Common Mistakes:

• Confusing DPI with figure size
• Choosing width or height as DPI
• Assuming default DPI without checking code

4. Identify the error in this code snippet:

plt.figure(figsize=8, dpi=100)

medium

A. No error, code is correct

B. dpi must be a tuple, not an integer

C. plt.figure does not accept figsize parameter

D. figsize must be a tuple, not a single integer

Solution

1. Step 1: Check the type of figsize argument

   figsize expects a tuple (width, height), but 8 is a single integer.

2. Step 2: Verify dpi argument

   dpi accepts an integer, so dpi=100 is correct.

3. Final Answer:

   figsize must be a tuple, not a single integer -> Option D

4. Quick Check:

   figsize = (width, height) tuple [OK]

Hint: Always pass figsize as (width, height) tuple [OK]

Common Mistakes:

• Passing single int instead of tuple to figsize
• Thinking dpi needs a tuple
• Assuming figsize can be a single number

5. You want to save a plot image that is 10 inches wide and 5 inches tall with a resolution of 300 DPI. Which code snippet correctly creates and saves this figure?

hard

A. fig = plt.figure(figsize=(10, 5))
plt.plot([1, 2, 3])
fig.savefig('plot.png', dpi=300)

B. fig = plt.figure(figsize=(10, 5), dpi=100)
plt.plot([1, 2, 3])
fig.savefig('plot.png')

C. fig = plt.figure(dpi=300)
plt.plot([1, 2, 3])
fig.savefig('plot.png', figsize=(10, 5))

D. fig = plt.figure(figsize=10, dpi=300)
plt.plot([1, 2, 3])
fig.savefig('plot.png')

Solution

1. Step 1: Understand figure creation and saving

   Setting figsize in plt.figure() controls size on screen, but saving with specific DPI requires passing dpi to savefig().

2. Step 2: Analyze each option

   fig = plt.figure(figsize=(10, 5), dpi=100) plt.plot([1, 2, 3]) fig.savefig('plot.png') sets DPI in figure but saving uses default DPI, so output may differ. fig = plt.figure(figsize=(10, 5)) plt.plot([1, 2, 3]) fig.savefig('plot.png', dpi=300) sets size in figure and DPI in savefig, ensuring correct size and resolution. fig = plt.figure(dpi=300) plt.plot([1, 2, 3]) fig.savefig('plot.png', figsize=(10, 5)) wrongly passes figsize to savefig, which is invalid. fig = plt.figure(figsize=10, dpi=300) plt.plot([1, 2, 3]) fig.savefig('plot.png') passes single int to figsize, causing error.

3. Final Answer:

   fig = plt.figure(figsize=(10, 5)) plt.plot([1, 2, 3]) fig.savefig('plot.png', dpi=300) -> Option A

4. Quick Check:

   Set size in figure, DPI in savefig for correct output [OK]

Hint: Set figsize in figure, dpi in savefig to control size and resolution [OK]

Common Mistakes:

• Setting dpi only in figure but not in savefig
• Passing figsize to savefig instead of figure
• Using single int instead of tuple for figsize