[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: MacOSX: Octave.app 2.9.14
From: |
John W. Eaton |
Subject: |
Re: MacOSX: Octave.app 2.9.14 |
Date: |
Wed, 03 Oct 2007 17:36:29 -0400 |
On 3-Oct-2007, Quentin Spencer wrote:
| OK, here's another version with special cases for p==2 and real and
| complex inputs.
|
| Quentin
|
| Index: scripts/linear-algebra/norm.m
| ===================================================================
| RCS file: /cvs/octave/scripts/linear-algebra/norm.m,v
| retrieving revision 1.31
| diff -u -r1.31 norm.m
| --- scripts/linear-algebra/norm.m 1 Oct 2007 15:59:33 -0000 1.31
| +++ scripts/linear-algebra/norm.m 3 Oct 2007 21:19:56 -0000
| @@ -93,6 +93,16 @@
| retval = max (abs (x));
| elseif (p == -Inf)
| retval = min (abs (x));
| + elseif (p == 1)
| + retval = sum (abs (x));
| + elseif (p == 2)
| + if (iscomplex (x))
| + y = abs(x);
| + y .*= y;
| + retval = sqrt (sum (y));
| + else
| + retval = sqrt (sum (x .* x));
| + endif
| else
| retval = sum (abs (x) .^ p) ^ (1/p);
| endif
OK, I used
elseif (p == 2)
if (iscomplex (x))
x = abs (x);
endif
retval = sqrt (sum (x .* x));
else
Is
sum (x .* x)
really faster here than
sumsq (x)
?
jwe
Re: MacOSX: Octave.app 2.9.14, David Bateman, 2007/10/03