np.dot() for dot product in NumPy

Introduction

We use np.dot() to multiply arrays in a way that combines rows and columns. It helps us find relationships between data points.

To calculate the total sales by multiplying price and quantity arrays.

To find the angle between two vectors in physics or machine learning.

To combine features in data for predictions using linear algebra.

To compute weighted sums in neural networks or statistics.

Syntax

NumPy

np.dot(a, b)

a and b can be numbers, 1D or 2D arrays.

For 1D arrays, it calculates the sum of products (like a scalar product).

Examples

This multiplies corresponding elements and sums them: 1*4 + 2*5 + 3*6 = 32.

NumPy

import numpy as np

# Dot product of two 1D arrays
v1 = np.array([1, 2, 3])
v2 = np.array([4, 5, 6])
result = np.dot(v1, v2)
print(result)

This multiplies matrices: rows of m1 with columns of m2.

NumPy

import numpy as np

# Dot product of 2D arrays (matrices)
m1 = np.array([[1, 2], [3, 4]])
m2 = np.array([[5, 6], [7, 8]])
result = np.dot(m1, m2)
print(result)

Here, the 1D array acts like a row vector multiplied by the matrix.

NumPy

import numpy as np

# Dot product of 1D and 2D array
v = np.array([1, 2])
m = np.array([[3, 4], [5, 6]])
result = np.dot(v, m)
print(result)

Sample Program

This program calculates the dot product of two simple vectors. It multiplies each pair of elements and sums them up.

NumPy

import numpy as np

# Define two vectors
vector_a = np.array([2, 3, 4])
vector_b = np.array([1, 0, -1])

# Calculate dot product
dot_product = np.dot(vector_a, vector_b)

print(f"Dot product of {vector_a} and {vector_b} is {dot_product}")

Important Notes

If arrays have incompatible shapes, np.dot() will raise an error.

For 2D arrays, np.dot() performs matrix multiplication.

For higher dimensions, consider using np.matmul() or the @ operator.

Summary

np.dot() multiplies arrays to find sums of products.

It works for vectors (1D) and matrices (2D).

Useful for combining data, physics calculations, and machine learning.

Practice

1. What does the np.dot() function do when applied to two 1D arrays (vectors)?

easy

A. Multiplies each element of the first array by the second array as a whole.

B. Adds the two arrays element-wise.

C. Returns the cross product of the two vectors.

D. Calculates the sum of products of corresponding elements (dot product).

Solution

1. Step 1: Understand np.dot() with 1D arrays

   When given two 1D arrays, np.dot() multiplies each pair of elements and sums them up.

2. Step 2: Compare with other operations

   Adding element-wise or cross product are different operations; np.dot() specifically does the sum of products.

3. Final Answer:

   Calculates the sum of products of corresponding elements (dot product). -> Option D

4. Quick Check:

   np.dot(vector1, vector2) = sum of element-wise products [OK]

Hint: Dot product sums element-wise multiplications [OK]

Common Mistakes:

• Confusing dot product with element-wise addition
• Thinking np.dot() returns cross product for 1D arrays
• Assuming np.dot() multiplies arrays element-wise without summing

2. Which of the following is the correct syntax for the np.dot() function to compute the dot product of two numpy arrays a and b?

easy

A. np.dot(a, b)

B. a.dot(b)

C. np.dot(a + b)

D. np.dot(a * b)

Solution

1. Step 1: Recall np.dot() syntax

   The function np.dot() takes two arguments: the first and second arrays to multiply.

2. Step 2: Check each option

   np.dot(a, b) correctly calls np.dot(a, b). a.dot(b) is valid but uses method syntax, not the function. Options A and C misuse the function by passing one argument or element-wise multiplication.

3. Final Answer:

   np.dot(a, b) -> Option A

4. Quick Check:

   np.dot(array1, array2) is correct syntax [OK]

Hint: Use np.dot(a, b) with two arguments [OK]

Common Mistakes:

• Passing only one argument to np.dot()
• Using addition or multiplication inside np.dot() incorrectly
• Confusing method call with function call

3. What is the output of the following code?

import numpy as np
x = np.array([1, 2, 3])
y = np.array([4, 5, 6])
result = np.dot(x, y)
print(result)

medium

A. [4 10 18]

B. 32

C. 15

D. Error

Solution

1. Step 1: Calculate element-wise products

   Multiply corresponding elements: 1*4=4, 2*5=10, 3*6=18.

2. Step 2: Sum the products

   Sum: 4 + 10 + 18 = 32.

3. Final Answer:

   32 -> Option B

4. Quick Check:

   Sum of products = 32 [OK]

Hint: Multiply and sum elements for dot product [OK]

Common Mistakes:

• Printing element-wise multiplication instead of sum
• Confusing dot product with addition
• Expecting a vector output instead of a scalar

4. Identify the error in this code snippet:

import numpy as np
A = np.array([[1, 2], [3, 4]])
B = np.array([5, 6, 7])
result = np.dot(A, B)
print(result)

medium

A. Syntax error in np.dot() call.

B. np.dot() requires three arguments.

C. Shape mismatch: cannot multiply 2x2 matrix with length 3 vector.

D. No error; output is [17 39].

Solution

1. Step 1: Check shapes of arrays

   Matrix A is 2x2, vector B has length 3. For dot product, inner dimensions must match.

2. Step 2: Identify mismatch

   2 (columns of A) does not equal 3 (length of B), so multiplication is invalid.

3. Final Answer:

   Shape mismatch: cannot multiply 2x2 matrix with length 3 vector. -> Option C

4. Quick Check:

   Matrix columns must match vector length [OK]

Hint: Check matrix columns match vector length [OK]

Common Mistakes:

• Ignoring shape mismatch and expecting output
• Thinking np.dot() can auto-adjust shapes
• Confusing syntax error with shape error

5. Given two matrices:

A = np.array([[1, 0, 2], [3, 1, 0]])
B = np.array([[2, 1], [0, 3], [1, 4]])

What is the result of np.dot(A, B)?

hard

A. [[4 9]
[6 6]]

B. [[2 1 8]
[6 3 0]]

C. [[2 1]
[0 3]
[1 4]]

D. Error due to shape mismatch

Solution

1. Step 1: Verify shapes for multiplication

   A is 2x3, B is 3x2, so multiplication is valid (3 matches 3).

2. Step 2: Calculate dot product manually

   Row 1 of A and column 1 of B: 1*2 + 0*0 + 2*1 = 2 + 0 + 2 = 4
   Row 1 of A and column 2 of B: 1*1 + 0*3 + 2*4 = 1 + 0 + 8 = 9
   Row 2 of A and column 1 of B: 3*2 + 1*0 + 0*1 = 6 + 0 + 0 = 6
   Row 2 of A and column 2 of B: 3*1 + 1*3 + 0*4 = 3 + 3 + 0 = 6

3. Final Answer:

   [[4 9] [6 6]] -> Option A

4. Quick Check:

   Matrix multiplication sums products of rows and columns [OK]

Hint: Multiply rows of A by columns of B and sum [OK]

Common Mistakes:

• Mixing up rows and columns during multiplication
• Expecting element-wise multiplication output
• Ignoring shape compatibility rules