Dumbbell charts in Matplotlib - Step-by-Step Execution

Concept Flow - Dumbbell charts

Prepare data with two values per category

↓

Plot points for first values

↓

Plot points for second values

↓

Draw lines connecting paired points

↓

Customize labels, colors, and axes

↓

Show chart

Dumbbell charts show two values per category connected by a line, highlighting differences visually.

Execution Sample

Matplotlib

import matplotlib.pyplot as plt
categories = ['A', 'B', 'C']
val1 = [10, 20, 30]
val2 = [15, 18, 35]
plt.scatter(val1, categories)
plt.scatter(val2, categories)
plt.hlines(categories, val1, val2)
plt.show()

This code plots two sets of points per category and connects them with horizontal lines to form a dumbbell chart.

Execution Table

Step	Action	Data/Variables	Plot Elements Created	Result
1	Import matplotlib.pyplot as plt	plt module loaded	None	Ready to plot
2	Define categories	categories = ['A', 'B', 'C']	None	Categories set
3	Define first values	val1 = [10, 20, 30]	None	First values set
4	Define second values	val2 = [15, 18, 35]	None	Second values set
5	Plot first points	plt.scatter(val1, categories)	Scatter points at (10,A), (20,B), (30,C)	First points visible
6	Plot second points	plt.scatter(val2, categories)	Scatter points at (15,A), (18,B), (35,C)	Second points visible
7	Draw horizontal lines	plt.hlines(categories, val1, val2)	Lines connecting (10,A)-(15,A), (20,B)-(18,B), (30,C)-(35,C)	Lines visible connecting points
8	Show plot	plt.show()	Full dumbbell chart displayed	Chart visible on screen

💡 Plot displayed and execution ends

Variable Tracker

Variable	Start	After Step 2	After Step 3	After Step 4	Final
categories	undefined	['A', 'B', 'C']	['A', 'B', 'C']	['A', 'B', 'C']	['A', 'B', 'C']
val1	undefined	undefined	[10, 20, 30]	[10, 20, 30]	[10, 20, 30]
val2	undefined	undefined	undefined	[15, 18, 35]	[15, 18, 35]

Key Moments - 3 Insights

Why do we use plt.hlines with categories as y-values?

Because categories are on the y-axis, plt.hlines draws horizontal lines at each category level connecting the two x-values (val1 and val2). See execution_table step 7.

What happens if val1 and val2 have different lengths?

Matplotlib will raise an error because scatter and hlines require matching lengths for x and y data. All arrays must align as shown in execution_table steps 5-7.

Why do we plot scatter points twice?

Each scatter plots one set of values (val1 and val2) so we see both points per category. This is shown in execution_table steps 5 and 6.

Visual Quiz - 3 Questions

Look at the execution_table at step 7, what does plt.hlines do?

A Draws vertical lines connecting categories

B Draws horizontal lines connecting val1 and val2 at each category

C Plots scatter points for val1

D Shows the plot window

💡 Hint

Check the 'Plot Elements Created' column at step 7 in execution_table

According to variable_tracker, what is the value of val2 after step 4?

A [10, 20, 30]

B undefined

C [15, 18, 35]

D ['A', 'B', 'C']

💡 Hint

Look at the val2 row and After Step 4 column in variable_tracker

If val1 was [10, 20] instead of length 3, what would happen?

A Matplotlib would raise an error due to length mismatch

B The third category would be plotted with zero value

C The plot would show only two categories

D The plot would ignore extra val2 values

💡 Hint

Refer to key_moments about array length matching and execution_table steps 5-7

Concept Snapshot

Dumbbell charts show two values per category connected by lines.
Use plt.scatter() twice for points, plt.hlines() to connect.
Categories on y-axis, values on x-axis.
Ensure val1, val2, and categories have same length.
Visualize differences clearly between paired values.

Full Transcript

Dumbbell charts help compare two values per category by plotting points and connecting them with lines. First, we prepare data arrays for categories and two sets of values. Then, we plot scatter points for each value set at the categories on the y-axis. Next, horizontal lines connect each pair of points to highlight differences. Finally, we show the plot. Variables categories, val1, and val2 hold the data. Scatter plots create points, and hlines draw connecting lines. Matching array lengths are important to avoid errors. This visual method clearly shows differences between paired data points.

Practice

1. What is the main purpose of a dumbbell chart in data visualization?

easy

A. To compare two related values for each category using connected points

B. To show the distribution of a single variable using bars

C. To display the relationship between two continuous variables as a scatter plot

D. To visualize hierarchical data using nested rectangles

Solution

1. Step 1: Understand dumbbell chart structure

   Dumbbell charts show two points per category connected by a line to highlight differences or changes.

2. Step 2: Compare with other chart types

   Bar charts show single values, scatter plots show relationships, and treemaps show hierarchies, so they don't fit dumbbell chart purpose.

3. Final Answer:

   To compare two related values for each category using connected points -> Option A

4. Quick Check:

   Dumbbell chart = compare two values [OK]

Hint: Dumbbell charts connect two points per category [OK]

Common Mistakes:

• Confusing dumbbell charts with bar charts
• Thinking dumbbell charts show distributions
• Mixing up scatter plots with dumbbell charts

2. Which matplotlib functions are typically used together to create a dumbbell chart?

easy

A. bar and plot

B. pie and bar

C. hist and scatter

D. hlines and scatter

Solution

1. Step 1: Identify dumbbell chart components

   Dumbbell charts use horizontal lines to connect two points and scatter points to mark values.

2. Step 2: Match matplotlib functions

   hlines draws horizontal lines, scatter plots points, so these two create dumbbell charts.

3. Final Answer:

   hlines and scatter -> Option D

4. Quick Check:

   Lines + points = hlines & scatter [OK]

Hint: Use hlines for lines and scatter for points [OK]

Common Mistakes:

• Using bar instead of hlines for connecting lines
• Confusing hist with scatter
• Trying pie charts for dumbbell visualization

3. What will be the output of this code snippet creating a dumbbell chart?

import matplotlib.pyplot as plt
categories = ['A', 'B']
start = [1, 3]
end = [4, 5]
plt.hlines(y=categories, xmin=start, xmax=end, color='gray')
plt.scatter(start, categories, color='blue')
plt.scatter(end, categories, color='red')
plt.show()

medium

A. A horizontal dumbbell chart with two categories showing blue and red points connected by gray lines

B. A vertical bar chart with categories on x-axis and values on y-axis

C. An error because categories cannot be used as y-values

D. A scatter plot with categories on y-axis and points only at start values

Solution

1. Step 1: Analyze hlines command

   plt.hlines(y=categories, xmin=start, xmax=end, color='gray') fails because y-coordinates must be numeric, but categories=['A', 'B'] are strings. Raises ValueError: could not convert string to float.

2. Step 2: Confirm scatter commands also fail

   plt.scatter(start, categories) and subsequent scatter require numeric y-values, leading to the same error.

3. Final Answer:

   An error because categories cannot be used as y-values -> Option C

4. Quick Check:

   Strings as y-values in hlines/scatter -> error [OK]

Hint: hlines and scatter require numeric y-positions; map categories to indices [OK]

Common Mistakes:

• Assuming strings work directly as y-values
• Expecting horizontal dumbbell chart output
• Confusing with bar chart output

4. Identify the error in this dumbbell chart code and select the fix:

import matplotlib.pyplot as plt
categories = ['X', 'Y']
start = [2, 4]
end = [5, 7]
plt.hlines(y=categories, xmin=start, xmax=end)
plt.scatter(start, categories)
plt.scatter(end, categories)
plt.show()

medium

A. No error; code runs correctly

B. Error because categories are strings; convert categories to numeric indices

C. Error because scatter requires numeric y-values; swap x and y in scatter

D. Error due to missing color argument in hlines

Solution

1. Step 1: Check data types for plotting

   hlines and scatter both require numeric y-values. Passing string categories as y causes ValueError (cannot convert string to float).

2. Step 2: Fix by using numeric indices

   Create y_pos = list(range(len(categories))), use y=y_pos in hlines and scatters, then plt.yticks(y_pos, categories).

3. Final Answer:

   Error because categories are strings; convert categories to numeric indices -> Option B

4. Quick Check:

   Categories (str) -> numeric indices [OK]

Hint: Use numeric y_pos = range(len(categories)); set labels with yticks [OK]

Common Mistakes:

• Assuming hlines and scatter accept strings equally
• Ignoring axis data type requirements
• Missing that scatter needs numeric coordinates

5. You have sales data for two years for 4 products: products = ['P1', 'P2', 'P3', 'P4'], sales_2022 = [100, 150, 200, 130], sales_2023 = [120, 140, 210, 160]. How would you create a dumbbell chart to compare sales changes using matplotlib?

hard

A. Use plt.hlines with y as products, xmin as sales_2022, xmax as sales_2023, and plt.scatter for points

B. Use plt.bar for both years side by side

C. Use plt.plot with products on x-axis and sales on y-axis

D. Use plt.scatter only with sales_2023 values

Solution

1. Step 1: Understand dumbbell chart requirements

   We need to show two values per product connected by a line to compare sales in 2022 and 2023.

2. Step 2: Map data to plotting functions

   hlines draws lines between sales_2022 and sales_2023 at y=products; scatter plots points for each year's sales.

3. Final Answer:

   Use plt.hlines with y as products, xmin as sales_2022, xmax as sales_2023, and plt.scatter for points -> Option A

4. Quick Check:

   Dumbbell = hlines + scatter with two values [OK]

Hint: Connect two year values with hlines; mark points with scatter [OK]

Common Mistakes:

• Using bar charts instead of dumbbell charts
• Plotting only one year's data
• Not connecting points with lines