np.ones() for one-filled arrays in NumPy

Introduction

We use np.ones() to quickly create arrays filled with the number one. This helps when you need a starting point or a simple array full of ones for calculations.

When you want to create a mask or filter array filled with ones.

When initializing weights or biases in machine learning models with ones.

When you need a simple array of ones to add or multiply with other arrays.

When testing functions that require input arrays filled with ones.

When you want to create a placeholder array with all elements set to one.

Syntax

NumPy

import numpy as np

array = np.ones(shape, dtype=None, order='C')

shape can be an integer or a tuple of integers defining the array size.

dtype sets the data type of the array elements (default is float).

Examples

This creates a simple 1D array with 5 elements, all set to 1.0.

NumPy

import numpy as np

# Create a 1D array of length 5 filled with ones
array_1d = np.ones(5)
print(array_1d)

This creates a 2D array with 3 rows and 4 columns, all elements are 1.0.

NumPy

import numpy as np

# Create a 2D array (3 rows, 4 columns) filled with ones
array_2d = np.ones((3, 4))
print(array_2d)

This creates an empty array because the size is zero.

NumPy

import numpy as np

# Create an empty array with zero size
empty_array = np.ones(0)
print(empty_array)

This creates a 1D array of integers instead of floats.

NumPy

import numpy as np

# Create a 1D array of integers filled with ones
int_array = np.ones(4, dtype=int)
print(int_array)

Sample Program

This program shows an initial array of zeros, then creates a new array of the same length filled with ones using np.ones(). It prints both arrays to compare.

NumPy

import numpy as np

# Create a 1D array of length 6 filled with ones
array_before = np.array([0, 0, 0, 0, 0, 0])
print('Before creating ones array:')
print(array_before)

# Use np.ones to create a new array filled with ones
array_ones = np.ones(6)
print('\nAfter creating ones array:')
print(array_ones)

Important Notes

Time complexity: Creating an array with np.ones() is very fast and runs in O(n) where n is the number of elements.

Space complexity: The array uses memory proportional to the number of elements.

A common mistake is to forget that the default data type is float, so the ones are 1.0, not integer 1, unless you specify dtype=int.

Use np.ones() when you need an array filled with ones quickly. For zeros, use np.zeros(). For other values, use np.full().

Summary

np.ones() creates arrays filled with the number one.

You can specify the shape and data type of the array.

This function is useful for initializing arrays or masks in data science tasks.

Practice

1. What does the np.ones() function do in NumPy?

easy

A. Creates an array filled with ones

B. Creates an array filled with zeros

C. Creates an array filled with random numbers

D. Creates an empty array

Solution

1. Step 1: Understand the function purpose

   np.ones() is designed to create arrays filled with the number one.

2. Step 2: Compare with other array creation functions

   Functions like np.zeros() create arrays filled with zeros, but np.ones() specifically fills with ones.

3. Final Answer:

   Creates an array filled with ones -> Option A

4. Quick Check:

   np.ones() = array of ones [OK]

Hint: Remember: ones means all elements are 1 [OK]

Common Mistakes:

• Confusing np.ones() with np.zeros()
• Thinking it creates random values
• Assuming it creates an empty array

2. Which of the following is the correct syntax to create a 2x3 array of ones with integer type using np.ones()?

easy

A. np.ones((2, 3), dtype=int)

B. np.ones([2, 3], type=int)

C. np.ones(2, 3, dtype='int')

D. np.ones(2, 3, int)

Solution

1. Step 1: Check the shape argument format

   The shape must be given as a tuple, so (2, 3) is correct.

2. Step 2: Check the dtype argument syntax

   The correct keyword is dtype, and the value can be int or 'int'.

3. Final Answer:

   np.ones((2, 3), dtype=int) -> Option A

4. Quick Check:

   Shape as tuple and dtype keyword correct [OK]

Hint: Use tuple for shape and 'dtype' keyword [OK]

Common Mistakes:

• Using list instead of tuple for shape
• Using 'type' instead of 'dtype'
• Passing shape as separate arguments

3. What is the output of the following code?

import numpy as np
arr = np.ones((2, 2), dtype=float)
print(arr)

medium

A. [[1 1]
[1 1]]

B. [[1. 1.]
[1. 1.]]

C. [[1 1 1]
[1 1 1]]

D. [[0. 0.]
[0. 0.]]

Solution

1. Step 1: Understand the shape and dtype

   The shape is (2, 2), so the array has 2 rows and 2 columns. The dtype is float, so elements are floats.

2. Step 2: Predict the printed output

   Each element is 1.0, printed as 1. with a decimal point. So the output is a 2x2 array of 1.0 values.

3. Final Answer:

   [[1. 1.] [1. 1.]] -> Option B

4. Quick Check:

   Shape (2,2) and float type means 1.0 values [OK]

Hint: Float dtype shows decimal points in output [OK]

Common Mistakes:

• Confusing integer and float output formats
• Misreading shape as (2,3)
• Expecting zeros instead of ones

4. Identify the error in this code snippet:

import numpy as np
arr = np.ones(3, 4)
print(arr)

medium

A. print() cannot display numpy arrays

B. Missing dtype argument

C. np.ones() cannot create 2D arrays

D. Shape must be a tuple, so use np.ones((3, 4))

Solution

1. Step 1: Check the shape argument

   The code passes two separate arguments 3 and 4, but np.ones() expects a single shape argument as a tuple.

2. Step 2: Correct the shape argument

   Use np.ones((3, 4)) to create a 3 by 4 array of ones.

3. Final Answer:

   Shape must be a tuple, so use np.ones((3, 4)) -> Option D

4. Quick Check:

   Shape argument must be tuple [OK]

Hint: Shape is always a tuple in np.ones() [OK]

Common Mistakes:

• Passing shape as separate arguments
• Forgetting parentheses around shape tuple
• Thinking dtype is mandatory

5. You want to create a 3D mask array of shape (2, 3, 4) filled with ones of boolean type using np.ones(). Which code snippet correctly does this?

hard

A. np.ones((2, 3, 4), type=bool)

B. np.ones([2, 3, 4], type=bool)

C. np.ones((2, 3, 4), dtype=bool)

D. np.ones(2, 3, 4, dtype=bool)

Solution

1. Step 1: Confirm the shape argument format

   The shape must be a tuple (2, 3, 4) to create a 3D array.

2. Step 2: Use correct dtype keyword and value

   The dtype keyword must be dtype and the value bool to create a boolean array.

3. Final Answer:

   np.ones((2, 3, 4), dtype=bool) -> Option C

4. Quick Check:

   Tuple shape and dtype=bool correct [OK]

Hint: Use tuple shape and dtype=bool for boolean arrays [OK]

Common Mistakes:

• Using list instead of tuple for shape
• Using 'type' instead of 'dtype'
• Passing shape as separate arguments