bill lam wrote:
> In the test script of lapack, the 2 line for heev failed and were
> commented
> out to make the rest of test continue.  Does anyone have any idea why they
> fail?

The matrices m and mz are not necessarily symmetric or Hermitian.  I see
that the test matrices in heev.ijs are all adjusted to be, using (+ |:) or
(+ (+@|:)).  Could this be the problem, or does heev adapt to this?

Best wishes,

John

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to