linalg:inv

(linalg:inv matrix)

Returns the inverse of a square matrix, computed by Gauss-Jordan elimination on the augmented matrix [a | I]. The result is a packed double-float array -- linalg computes in floating point (speed over exactness), so a general inverse carries the usual rounding. To solve a linear system it is usually clearer to call linalg:solve directly.

A singular matrix (one whose linalg:det is 0) has no inverse and signals an error:

> (linalg:inv #2A((1 2) (2 4)))
Error: linalg: inv of a singular matrix