np.sign() for sign detection in NumPy - Step-by-Step Execution

Concept Flow - np.sign() for sign detection

Input array

↓

Apply np.sign()

↓

Check each element's sign

↓

Output array with -1, 0, or 1

↓

Use sign info for analysis

np.sign() takes each number and returns -1 if negative, 0 if zero, and 1 if positive, helping detect the sign of each element.

Execution Sample

NumPy

import numpy as np
arr = np.array([-3, 0, 4, -1, 5])
signs = np.sign(arr)
print(signs)

This code finds the sign of each number in the array, outputting -1, 0, or 1 for each element.

Execution Table

Step	Input Element	np.sign() Result	Explanation
1	-3	-1	Negative number returns -1
2	0	0	Zero returns 0
3	4	1	Positive number returns 1
4	-1	-1	Negative number returns -1
5	5	1	Positive number returns 1
6	All elements processed	[-1 0 1 -1 1]	Final output array with signs

💡 All elements processed, np.sign() returned sign for each element

Variable Tracker

Variable	Start	After 1	After 2	After 3	After 4	After 5	Final
arr	[-3, 0, 4, -1, 5]	[-3, 0, 4, -1, 5]	[-3, 0, 4, -1, 5]	[-3, 0, 4, -1, 5]	[-3, 0, 4, -1, 5]	[-3, 0, 4, -1, 5]	[-3, 0, 4, -1, 5]
signs	[]	[-1]	[-1, 0]	[-1, 0, 1]	[-1, 0, 1, -1]	[-1, 0, 1, -1, 1]	[-1, 0, 1, -1, 1]

Key Moments - 2 Insights

Why does np.sign() return 0 for zero but not for other numbers?

np.sign() returns 0 only when the input element is exactly zero, as shown in execution_table step 2. This distinguishes zero from positive or negative numbers.

Can np.sign() handle arrays with mixed positive, negative, and zero values?

Yes, as shown in the execution_table, np.sign() processes each element independently and returns the correct sign for each, producing an array of -1, 0, or 1.

Visual Quiz - 3 Questions

Look at the execution_table, what is the np.sign() result for the input element -1 at step 4?

A -1

B 0

C 1

D None

💡 Hint

Check execution_table row with Step 4 for input -1 and its np.sign() result.

At which step does np.sign() return 0 according to the execution_table?

A Step 1

B Step 3

C Step 2

D Step 5

💡 Hint

Look for the step where the input element is zero and np.sign() returns 0.

If the input array had an extra element 7 at the end, how would the final signs array change?

A [-1, 0, 1, -1, 1, -1]

B [-1, 0, 1, -1, 1, 1]

C [-1, 0, 1, -1, 1, 0]

D [-1, 0, 1, -1, 1]

💡 Hint

Positive number 7 would add a 1 at the end of the signs array.

Concept Snapshot

np.sign(array) returns an array with -1 for negatives, 0 for zeros, and 1 for positives.
Useful to quickly detect the sign of each element.
Works element-wise on numpy arrays.
Output array has same shape as input.
Simple way to classify numbers by sign.

Full Transcript

This lesson shows how np.sign() works in numpy. It takes each number in an array and returns -1 if the number is negative, 0 if it is zero, and 1 if it is positive. We traced an example array with values -3, 0, 4, -1, and 5. Step by step, np.sign() returned -1, 0, 1, -1, and 1 respectively. The final output is an array of these sign values. This helps us quickly understand the sign of each number in data. We also answered common questions about zero handling and mixed values. Finally, quiz questions tested understanding of the sign results at different steps and how adding a positive number changes the output.

Practice

1. What does the np.sign() function return when applied to a negative number?

easy

A. -1

B. 0

C. 1

D. The original number

Solution

1. Step 1: Understand np.sign() behavior

   The function returns -1 for negative numbers, 0 for zero, and 1 for positive numbers.

2. Step 2: Apply to a negative number

   Since the input is negative, np.sign() returns -1.

3. Final Answer:

   -1 -> Option A

4. Quick Check:

   Negative number sign = -1 [OK]

Hint: Negative input always gives -1 from np.sign() [OK]

Common Mistakes:

• Confusing negative with zero
• Expecting original number as output
• Thinking it returns boolean

2. Which of the following is the correct syntax to get the sign of each element in a numpy array arr?

easy

A. np.sign(arr)

B. arr.sign()

C. sign(arr)

D. np.sign_of(arr)

Solution

1. Step 1: Recall numpy function usage

   Functions in numpy are called with the syntax np.function_name(arguments).

2. Step 2: Identify correct function call

   The correct function to get sign is np.sign(), so np.sign(arr) is correct.

3. Final Answer:

   np.sign(arr) -> Option A

4. Quick Check:

   Correct numpy function call = np.sign(arr) [OK]

Hint: Use np.sign(array) to get signs of all elements [OK]

Common Mistakes:

• Using method on array like arr.sign()
• Calling sign() without np prefix
• Using non-existent np.sign_of()

3. What is the output of the following code?

import numpy as np
arr = np.array([-3, 0, 4])
sign_arr = np.sign(arr)
print(sign_arr)

medium

A. [-3 0 4]

B. [3 0 4]

C. [0 0 0]

D. [-1 0 1]

Solution

1. Step 1: Understand input array values

   The array has values -3 (negative), 0 (zero), and 4 (positive).

2. Step 2: Apply np.sign() to each element

   np.sign(-3) = -1, np.sign(0) = 0, np.sign(4) = 1, so the output array is [-1, 0, 1].

3. Final Answer:

   [-1 0 1] -> Option D

4. Quick Check:

   Signs of [-3,0,4] = [-1,0,1] [OK]

Hint: np.sign() maps negative to -1, zero to 0, positive to 1 [OK]

Common Mistakes:

• Expecting original values
• Confusing zero with positive
• Outputting boolean instead of sign

4. The following code throws an error. What is the mistake?

import numpy as np
arr = [-1, 2, 0]
signs = np.sign arr
print(signs)

medium

A. print() syntax error

B. Using list instead of numpy array

C. Missing parentheses in function call

D. np.sign does not exist

Solution

1. Step 1: Check function call syntax

   The code uses np.sign arr without parentheses, which is invalid syntax in Python.

2. Step 2: Correct the syntax

   It should be np.sign(arr) with parentheses to call the function properly.

3. Final Answer:

   Missing parentheses in function call -> Option C

4. Quick Check:

   Function calls need parentheses [OK]

Hint: Always use parentheses when calling functions [OK]

Common Mistakes:

• Omitting parentheses
• Thinking lists cause error here
• Assuming np.sign is undefined

5. You have a numpy array data = np.array([-5, 0, 3, -2, 7]). How can you create a new array that replaces all negative values with 0, using np.sign()?

hard

A. data * np.sign(data)

B. data * (np.sign(data) + 1) / 2

C. np.sign(data) * 2

D. data + np.sign(data)

Solution

1. Step 1: Understand np.sign() output

   np.sign(data) gives -1 for negatives, 0 for zero, 1 for positives.

2. Step 2: Transform sign to mask for positives and zero

   Adding 1 to sign gives 0 for -1, 1 for 0, 2 for 1. Dividing by 2 maps negatives to 0, zero to 0.5, positives to 1.

3. Step 3: Multiply original data by this mask

   Multiplying data by this mask sets negative values to 0, keeps zero and positive values unchanged (zero times 0.5 is 0).

4. Final Answer:

   data * (np.sign(data) + 1) / 2 -> Option B

5. Quick Check:

   Mask negatives to zero using (sign+1)/2 [OK]

Hint: Use (sign+1)/2 as mask to zero negatives [OK]

Common Mistakes:

• Using sign directly multiplies negatives
• Adding sign to data changes values wrongly
• Confusing mask calculation