Cumulative histograms in Matplotlib - Time & Space Complexity

Time Complexity: Cumulative histograms

O(n)

Understanding Time Complexity

We want to understand how the time to create a cumulative histogram changes as the data size grows.

How does the number of data points affect the work matplotlib does to draw the histogram?

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

import matplotlib.pyplot as plt
import numpy as np

data = np.random.randn(1000)
plt.hist(data, bins=50, cumulative=True)
plt.show()

This code creates a cumulative histogram from 1000 random data points divided into 50 bins.

Identify Repeating Operations

• Primary operation: Counting how many data points fall into each bin.
• How many times: Each of the n data points is checked once to find its bin.

How Execution Grows With Input

As the number of data points increases, the work to count them into bins grows roughly in direct proportion.

Input Size (n)	Approx. Operations
10	About 10 checks
100	About 100 checks
1000	About 1000 checks

Pattern observation: Doubling the data roughly doubles the counting work.

Final Time Complexity

Time Complexity: O(n)

This means the time to build the cumulative histogram grows linearly with the number of data points.

Common Mistake

[X] Wrong: "The number of bins affects the time complexity more than the number of data points."

[OK] Correct: The bins are fixed and small compared to data size; the main work is checking each data point once, so data size dominates time.

Interview Connect

Understanding how data size affects plotting time helps you explain performance in data visualization tasks clearly and confidently.

Self-Check

"What if we increased the number of bins significantly? How would the time complexity change?"

Practice

1. What does setting cumulative=true in plt.hist() do to the histogram?

easy

A. It shows the total counts accumulated up to each bin.

B. It changes the histogram to a bar chart.

C. It normalizes the histogram to show probabilities.

D. It colors the bars based on their height.

Solution

1. Step 1: Understand the cumulative parameter

   Setting cumulative=True in plt.hist() changes the histogram to show accumulated counts.

2. Step 2: Effect on histogram output

   Instead of showing counts per bin, it shows the sum of counts up to that bin, making the histogram cumulative.

3. Final Answer:

   It shows the total counts accumulated up to each bin. -> Option A

4. Quick Check:

   cumulative=True means accumulation [OK]

Hint: Remember cumulative=True sums counts up to each bin [OK]

Common Mistakes:

• Thinking it normalizes the histogram
• Confusing cumulative with coloring
• Assuming it changes chart type to bar chart

2. Which of the following is the correct syntax to create a cumulative histogram of data array data using matplotlib?

easy

A. plt.hist(data, cumulative=True)

B. plt.hist(data, cumulative=0)

C. plt.hist(data, cumulate=True)

D. plt.hist(data, cumulative)

Solution

1. Step 1: Check correct parameter name and usage

   The correct parameter to make a histogram cumulative is cumulative=True.

2. Step 2: Validate syntax correctness

   plt.hist(data, cumulative=True) uses the exact correct syntax. plt.hist(data, cumulative=0) uses 0 instead of True (no cumulative effect). plt.hist(data, cumulate=True) misspells the parameter. plt.hist(data, cumulative) is invalid syntax.

3. Final Answer:

   plt.hist(data, cumulative=True) -> Option A

4. Quick Check:

   Use cumulative=True exactly [OK]

Hint: Use cumulative=True exactly for cumulative histograms [OK]

Common Mistakes:

• Misspelling 'cumulative' parameter
• Using cumulative without assignment
• Using cumulative=0 (disables cumulative)

3. What will be the output counts of the cumulative histogram bins for data = [1, 2, 2, 3, 4] with bins [1, 2, 3, 4, 5] using plt.hist(data, bins=[1,2,3,4,5], cumulative=True)?

medium

A. [0, 1, 3, 5]

B. [1, 3, 4, 5]

C. [1, 1, 2, 3]

D. [1, 2, 3, 4]

Solution

1. Step 1: Count data points in each bin

   Bins: [1-2), [2-3), [3-4), [4-5). Counts per bin: 1 in first, 2 in second, 1 in third, 1 in fourth.

2. Step 2: Calculate cumulative counts

   Cumulative sums: 1, 1+2=3, 3+1=4, 4+1=5.

3. Final Answer:

   [1, 3, 4, 5] -> Option B

4. Quick Check:

   Cumulative sums of counts = [1,3,4,5] [OK]

Hint: Add counts bin by bin for cumulative totals [OK]

Common Mistakes:

• Counting bins inclusively instead of half-open
• Not summing counts cumulatively
• Miscounting data points per bin

4. The code below is intended to plot a cumulative histogram but does not produce the expected cumulative effect. What is the error?

import matplotlib.pyplot as plt
data = [1,2,2,3,4]
plt.hist(data, bins=4, cumulative=False)
plt.show()

medium

A. The bins parameter must be a list, not an integer.

B. plt.show() is missing and must be added.

C. The parameter cumulative should be set to True, not False.

D. Data list must be sorted before plotting.

Solution

1. Step 1: Identify the cumulative parameter usage

   The code sets cumulative=False, so the histogram is not cumulative.

2. Step 2: Correct the parameter to get cumulative histogram

   Changing cumulative=False to cumulative=True will produce the cumulative histogram as intended.

3. Final Answer:

   The parameter cumulative should be set to True, not False. -> Option C

4. Quick Check:

   cumulative=True needed for cumulative histogram [OK]

Hint: Set cumulative=True to enable cumulative histogram [OK]

Common Mistakes:

• Leaving cumulative=False by mistake
• Thinking bins must be a list always
• Believing data must be sorted
• Forgetting plt.show() (though present here)

5. You have two datasets: data1 = [1,2,3,4,5] and data2 = [2,3,4,5,6]. You want to plot their cumulative histograms on the same plot to compare their distributions. Which code snippet correctly does this?

hard

A. plt.hist([data1, data2], bins=6, cumulative=False, label=['Data1', 'Data2']); plt.legend(); plt.show()

B. plt.hist(data1 + data2, bins=6, cumulative=True); plt.show()

C. plt.hist(data1, bins=6); plt.hist(data2, bins=6); plt.show()

D. plt.hist([data1, data2], bins=6, cumulative=True, label=['Data1', 'Data2'], histtype='step'); plt.legend(); plt.show()

Solution

1. Step 1: Understand plotting multiple cumulative histograms

   To compare cumulative histograms, pass both datasets as a list, set cumulative=True, and use histtype='step' for clarity.

2. Step 2: Check options for correct parameters

   plt.hist([data1, data2], bins=6, cumulative=True, label=['Data1', 'Data2'], histtype='step'); plt.legend(); plt.show() correctly uses plt.hist([data1, data2], bins=6, cumulative=True, label=['Data1', 'Data2'], histtype='step') with legend and show. plt.hist(data1 + data2, bins=6, cumulative=True); plt.show() incorrectly combines the datasets into one histogram. plt.hist(data1, bins=6); plt.hist(data2, bins=6); plt.show() is not cumulative. plt.hist([data1, data2], bins=6, cumulative=False, label=['Data1', 'Data2']); plt.legend(); plt.show() is not cumulative.

3. Final Answer:

   plt.hist([data1, data2], bins=6, cumulative=True, label=['Data1', 'Data2'], histtype='step'); plt.legend(); plt.show() -> Option D

4. Quick Check:

   Multiple datasets + cumulative=True + histtype='step' = plt.hist([data1, data2], bins=6, cumulative=True, label=['Data1', 'Data2'], histtype='step'); plt.legend(); plt.show() [OK]

Hint: Use list of datasets and cumulative=True with histtype='step' [OK]

Common Mistakes:

• Combining datasets into a single histogram
• Forgetting cumulative=True
• Not using histtype='step' for clear lines
• Using cumulative=False by mistake