Solving linear systems (solve) in SciPy - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the purpose of the scipy.linalg.solve function?

It solves a system of linear equations Ax = b for x, where A is a square matrix and b is a vector or matrix of constants.

beginner

What are the input requirements for scipy.linalg.solve?

The matrix A must be square (same number of rows and columns), and b must have a compatible shape (same number of rows as A).

intermediate

What happens if the matrix A is singular or nearly singular when using scipy.linalg.solve?

The function will raise a LinAlgError because the system does not have a unique solution or is unstable to solve.

intermediate

How does scipy.linalg.solve differ from using matrix inversion to solve Ax = b?

It is more efficient and numerically stable to use solve directly rather than computing the inverse of A and multiplying by b.

beginner

Write a simple example code snippet using scipy.linalg.solve to solve the system:<br>2x + 3y = 8<br>5x + 4y = 13

import numpy as np
from scipy.linalg import solve

A = np.array([[2, 3], [5, 4]])
b = np.array([8, 13])
x = solve(A, b)
print(x)  # Output: [1. 2.]

What shape must matrix A have to use scipy.linalg.solve?

A Any rectangular matrix

B Square matrix

C Only diagonal matrix

D Only identity matrix

The matrix A must be square to solve Ax = b uniquely.

If A is singular, what does scipy.linalg.solve do?

A Raises a <code>LinAlgError</code>

B Automatically inverts <code>A</code>

C Returns zeros

D Returns a solution anyway

Singular matrices do not have unique solutions, so LinAlgError is raised.

Which is better for solving Ax = b numerically?

A Using <code>solve</code> function

B Calculating <code>A</code> inverse and multiplying by <code>b</code>

C Guessing the solution

D Using random numbers

solve is more efficient and stable than calculating the inverse.

What does the output of scipy.linalg.solve(A, b) represent?

A The matrix <code>A</code> multiplied by <code>b</code>

B The inverse of <code>A</code>

C The vector <code>x</code> that satisfies <code>Ax = b</code>

D The determinant of <code>A</code>

The output is the solution vector x.

Which library provides the solve function used here?

A pandas

B numpy.linalg

C matplotlib

D scipy.linalg

solve is from scipy.linalg.

Explain how to use scipy.linalg.solve to solve a system of linear equations.

Think about the inputs and outputs of the function.

Describe why using scipy.linalg.solve is preferred over calculating the inverse of a matrix to solve linear systems.

Consider performance and accuracy.