I have a square matrix with very small values of the order of 10^-9 or 10^-20. I need to compute the inverse of this matrix. But when I try to compute the inverse of the function using Python's inbuilt numpy.linalg.inv(.) function, I get an inverse matrix with all entries 0. Is there any transformation that can be applied to the matrix to extract the inverse approximation out of the original matrix.

  • 2
    $\begingroup$ Are all of the entries small (say, smaller than $10^{-9}$ in absolute value)? If so, then Morgan's trick should work $\endgroup$ Mar 30, 2019 at 16:34
  • $\begingroup$ Obligatory: Are you sure you need to compute the inverse? There might be more appropriate tools for the task you have in mind. $\endgroup$
    – Elle Najt
    Mar 30, 2019 at 16:41
  • $\begingroup$ @Lorenzo Yes it is inverse itself. However, even if not exact, any approximations will also do. $\endgroup$ Mar 30, 2019 at 23:15
  • $\begingroup$ This may help:geeksforgeeks.org/precision-handling-python and mpmath.org or stackoverflow.com/questions/11522933/… $\endgroup$
    – NoChance
    Mar 31, 2019 at 5:46


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy