Inconsistent results when using np.linalg.inv on singular matrices - how to handle this?
I'm working on a project where I need to compute the inverse of several matrices using NumPy's `np.linalg.inv` function. However, I'm experiencing inconsistent results with singular matrices. For example, the following code: ```python import numpy as np A = np.array([[1, 2], [2, 4]]) try: inv_A = np.linalg.inv(A) except np.linalg.LinAlgError as e: print(f"behavior: {e}") ``` throws the behavior: `behavior: Singular matrix`. I understand that a singular matrix doesn't have an inverse, but I expected this to return a specific warning or output instead of raising an exception. I've tried using `np.linalg.pinv(A)` to compute the pseudo-inverse, but I still want to figure out how to handle this situation more gracefully. Is there a recommended way to check if a matrix is singular before attempting to compute its inverse? Additionally, what are the best practices for managing such exceptions in a larger application, especially when dealing with multiple matrices in a batch operation? I'm using NumPy version 1.23.1. Any insights would be greatly appreciated!