Singular matrix issue with Numpy

1.3K    Asked by jackBurgess in Python , Asked on Apr 14, 2021

 I am trying to multiply a vector(3 by 1) by its transpose(1 by 3). I get a (3 by 3) array but I cannot get its inverse. Any idea why?

import numpy as np

c=array([1, 8, 50])
np.transpose(c[np.newaxis]) * c
array([[   1,    8,   50],
   [   8,   64,  400],
   [  50,  400, 2500]])
np.linalg.inv(np.transpose(c[np.newaxis]) * c)
Traceback (most recent call last):
  File "", line 1, in
  File "C:Python26libsite-packagesnumpylinalglinalg.py", line 445, in inv
    return wrap(solve(a, identity(a.shape[0], dtype=a.dtype)))
  File "C:Python26libsite-packagesnumpylinalglinalg.py", line 328, in solve

    raise LinAlgError, 'Singular matrix'

LinAlgError: Singular matrix

Answered by jack Burgess

The matrix you pasted:

[[   1,    8,   50],
 [ 8, 64, 400], 
       [  50,  400, 2500]

Has a determinant of zero. This is known as linalgerror: singular matrix (one for which an inverse does not exist)



Your Answer

Interviews

Parent Categories