np.exp() and np.log() in NumPy - Time & Space Complexity

Time Complexity: np.exp() and np.log()

O(n)

Understanding Time Complexity

We want to understand how the time to run np.exp() and np.log() changes as the input size grows.

How does the cost of these functions grow when we apply them to bigger arrays?

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

import numpy as np

arr = np.random.rand(n)
exp_arr = np.exp(arr)
log_arr = np.log(arr + 1e-10)  # avoid log(0)

This code creates an array of size n, then applies the exponential and logarithm functions element-wise.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: Applying np.exp() and np.log() to each element of the array.
• How many times: Once for each of the n elements in the array.

How Execution Grows With Input

As the array size n grows, the number of operations grows roughly the same way, because each element needs one exponential and one logarithm calculation.

Input Size (n)	Approx. Operations
10	About 20 operations (10 exp + 10 log)
100	About 200 operations
1000	About 2000 operations

Pattern observation: The total work grows directly with n, doubling n doubles the work.

Final Time Complexity

Time Complexity: O(n)

This means the time to run these functions grows linearly with the size of the input array.

Common Mistake

[X] Wrong: "np.exp() and np.log() run in constant time no matter the input size."

[OK] Correct: These functions are applied to every element in the array, so the total time grows with the number of elements.

Interview Connect

Knowing how vectorized functions like np.exp() and np.log() scale helps you understand performance when working with large datasets.

Self-Check

"What if we applied np.exp() only to a fixed-size subset of the array instead of the whole array? How would the time complexity change?"

Practice

1. What does the function np.exp(x) compute in NumPy?

easy

A. The square root of x

B. The value of e raised to the power x

C. The natural logarithm of x

D. The sine of x

Solution

1. Step 1: Understand the purpose of np.exp()

   The function np.exp() calculates e (Euler's number, approximately 2.718) raised to the power of the input value.

2. Step 2: Compare with other options

   Other options describe different functions: natural log is np.log(), square root is np.sqrt(), sine is np.sin().

3. Final Answer:

   The value of e raised to the power x -> Option B

4. Quick Check:

   np.exp(x) = e^x [OK]

Hint: Remember: exp means e to the power x [OK]

Common Mistakes:

• Confusing np.exp() with np.log()
• Thinking np.exp() calculates logarithm
• Mixing up with square root or trigonometric functions

2. Which of the following is the correct syntax to compute the natural logarithm of a NumPy array arr?

easy

A. np.log(arr)

B. np.exp(arr)

C. np.ln(arr)

D. np.log10(arr)

Solution

1. Step 1: Identify the function for natural logarithm

   The natural logarithm in NumPy is computed using np.log().

2. Step 2: Check other options for correctness

   np.exp() calculates exponentials, np.ln() does not exist, and np.log10() calculates base-10 logarithm.

3. Final Answer:

   np.log(arr) -> Option A

4. Quick Check:

   Natural log = np.log() [OK]

Hint: Natural log uses np.log(), not np.ln() or np.log10() [OK]

Common Mistakes:

• Using np.ln() which is not a valid NumPy function
• Confusing natural log with base-10 log
• Using np.exp() instead of np.log()

3. What is the output of the following code?

import numpy as np
arr = np.array([1, 2, 3])
result = np.log(np.exp(arr))
print(result)

medium

A. [1. 2. 3.]

B. [0. 0. 0.]

C. [2.718 7.389 20.086]

D. Error: invalid input

Solution

1. Step 1: Understand the inner function np.exp(arr)

   Applying np.exp() to [1, 2, 3] gives [e^1, e^2, e^3] ≈ [2.718, 7.389, 20.086].

2. Step 2: Apply np.log() to the result

   Taking the natural log of these values returns the original array [1, 2, 3] because log and exp are inverse functions.

3. Final Answer:

   [1. 2. 3.] -> Option A

4. Quick Check:

   np.log(np.exp(x)) = x [OK]

Hint: log(exp(x)) returns x because they undo each other [OK]

Common Mistakes:

• Expecting the exponential values instead of original
• Confusing output with zeros
• Thinking it causes an error

4. Identify the error in the following code snippet:

import numpy as np
arr = np.array([-1, 0, 1])
result = np.log(arr)
print(result)

medium

A. np.log() should be np.exp()

B. np.array() syntax is incorrect

C. np.log() cannot take zero or negative values

D. No error, code runs fine

Solution

1. Step 1: Check input values for np.log()

   Natural logarithm is undefined for zero and negative numbers. The array contains -1 and 0, which cause errors or warnings.

2. Step 2: Understand the error behavior

   NumPy will return -inf or NaN for zero or negative inputs, which is usually an error or warning in calculations.

3. Final Answer:

   np.log() cannot take zero or negative values -> Option C

4. Quick Check:

   Log input must be positive [OK]

Hint: Logarithm inputs must be positive numbers only [OK]

Common Mistakes:

• Ignoring domain restrictions of log function
• Confusing np.log() with np.exp()
• Assuming code runs without warnings or errors

5. You have a dataset of positive values stored in a NumPy array data. You want to normalize it by applying the natural logarithm, then reverse the transformation after some processing. Which sequence of operations correctly achieves this?

hard

A. Apply np.exp(data) first, then np.log() on the result to reverse

B. Apply np.sqrt(data) first, then np.log() on the result to reverse

C. Apply np.log10(data) first, then np.exp() on the result to reverse

D. Apply np.log(data) first, then np.exp() on the result to reverse

Solution

1. Step 1: Understand the normalization step

   Applying np.log(data) transforms data to a logarithmic scale, useful for normalization.

2. Step 2: Reverse transformation

   To get back original data, apply np.exp() to the logged data because exp is the inverse of log.

3. Step 3: Check other options

   Apply np.exp(data) first, then np.log() on the result to reverse reverses the order incorrectly, Apply np.log10(data) first, then np.exp() on the result to reverse mixes log base 10 with exp (base e), Apply np.sqrt(data) first, then np.log() on the result to reverse uses sqrt which is unrelated.

4. Final Answer:

   Apply np.log(data) first, then np.exp() on the result to reverse -> Option D

5. Quick Check:

   log then exp returns original data [OK]

Hint: Log then exp reverses; order matters [OK]

Common Mistakes:

• Reversing the order of log and exp
• Mixing log base 10 with exp
• Using unrelated functions like sqrt