Slicing with start:stop:step in NumPy - Time & Space Complexity

Time Complexity: Slicing with start:stop:step

O(1)

Understanding Time Complexity

We want to understand how the time it takes to slice a numpy array changes as the array gets bigger.

Specifically, how does using start, stop, and step in slicing affect the work done?

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

import numpy as np
arr = np.arange(1000)
sliced = arr[10:900:5]

This code creates an array of 1000 numbers and then slices it from index 10 to 899, taking every 5th element.

Identify Repeating Operations

• Primary operation: Adjusting metadata (offset, shape, strides) to create a view of the original array. No data copying.
• How many times: Constant time, performed once regardless of array size or slice parameters.

How Execution Grows With Input

As the array size grows, the slicing operation remains constant time because it creates a view, not a copy.

Input Size (n)	Approx. Operations
10	1 (metadata adjustment)
100	1
1000	1

Pattern observation: The work is constant, independent of array size, slice length, or step size.

Final Time Complexity

Time Complexity: O(1)

This means the time is constant, as NumPy slicing creates a zero-copy view by adjusting metadata only.

Common Mistake

[X] Wrong: "Slicing copies the selected elements, so time proportional to output size (O(k))."

[OK] Correct: NumPy slicing returns a view with adjusted strides—no data is copied. Use .copy() for actual copying, which is O(k).

Interview Connect

Knowing slicing is O(1) helps optimize code by avoiding unnecessary copies, crucial for large datasets in data science pipelines.

Self-Check

"What if we changed the step size to 1? How would the time complexity change?"

Answer: Still O(1), as it's always a view.