Alois Schlögl wrote:
Søren Hauberg wrote:
fre, 21 05 2010 kl. 09:53 +0200, skrev Alois Schlögl:
Søren Hauberg wrote:
lør, 20 03 2010 kl. 19:36 +0100, skrev Michael Creel:
I think that replacing list with cell arrays is pretty
straightforward. I did that for samin and bfgsmin quite a while ago,
when lists were deprecated. I suggest that the authors of the
functions in question take care of this before Octave 3.4 appears. If
they don't, then functions that stop working can be moved to a "deep
freeze" directory, so that they can be recovered when someone has time
to work on them.
This is fine by me.
The following functions seem to be using lists at the moment. Unless
they are fixed by the time Octave 3.4 is released, they will be put in a
directory that is not installed as part of the package.
d2_min.m
fminunc_compat.m
line_min.m
minimize.m
nelder_mead_min.m
test_d2_min_2.m
test_d2_min_3.m
test_fminunc_1.m
test_min_3.m
test_min_4.m
test_minimize_1.m
test_nelder_mead_min_1.m
deriv.m
optimset_compat.m
__semi_bracket.m
So, if you care about any of these functions, please speak up.
I've replaced all list elements in optim. According to my tests, these
tests work fine:
test_d2_min_1 % ok
test_d2_min_2 % ok
test_d2_min_3 % ok
test_nelder_mead_min_2% ok
test_wpolyfit % ok
test_min_3 % ok
I guess, this means that most other functions (like d2_min, line_min,
deriv, minimize, nelder_mead_min, and fminunc_compat are also fine. )
The following tests are still failing
test_min_4 % ffailed
test_minimize_1 % failed
test_nelder_mead_min_1% failed
test_fminunc_1 % failed
test_min_4 fails with the following usage error:
usage: bfgsmin: 1st element of 3rd argument (iteration limit) must be
Inf or positive integer.
Perhaps, this test is broken?
The other tests fail with a segmentation fault - so there is at least a
problem somewhere else, too.
I've investigated the segmentation faults in the functions
test_minimize_1
test_nelder_mead_min_1
test_fminunc_1
and filed a bug report with octave
https://savannah.gnu.org/bugs/index.php?29960
This bug was fixed in octave. I run the tests of optim with the most
recent development version of octave. A few minor corrections were
needed in the source of optim
http://octave.svn.sourceforge.net/viewvc/octave?view=rev&revision=7348
clear; verbose=1; test_d2_min_1
clear; verbose=1; test_d2_min_2
clear; verbose=1; test_d2_min_3
clear; verbose=1; test_fminunc_1
clear; verbose=1; test_min_1
clear; verbose=1; test_min_2
clear; verbose=1; test_min_3
clear; verbose=1; test_min_4
clear; verbose=1; test_minimize_1
clear; verbose=1; test_nelder_mead_min_1
clear; verbose=1; test_nelder_mead_min_2
clear; verbose=1; test_wpolyfit
In order to run these tests, I limited maxev in nelder_mead_min.m,
otherwise it would not terminate (within a short time), and remove the
test on using the "jac" - option in test_fminunc_1.m otherwise one gets
"error: Option 'jac' doesn't exist any more. Sorry."
(see patches below).
The resulting diary file is attached. All tests terminated without any
seg-faults. Many tests report an ok result; several cases report "Too
many iterations for this function" and/or "Error is too big". I guess
these issues are not related to the conversion from lists to cells.
Alois
Index: nelder_mead_min.m
===================================================================
--- nelder_mead_min.m (revision 7359)
+++ nelder_mead_min.m (working copy)
@@ -93,7 +93,7 @@
crit = 0; # Stopping criterion ctl(1)
tol = 10*eps; # Stopping test's threshold ctl(2)
narg = 1; # Position of minimized arg ctl(3)
-maxev = inf; # Max num of func evaluations ctl(4)
+maxev = 1e4; # Max num of func evaluations ctl(4)
isz = 1; # Initial size ctl(5)
rst = 0; # Max # of restarts
Index: test_fminunc_1.m
===================================================================
--- test_fminunc_1.m (revision 7359)
+++ test_fminunc_1.m (working copy)
@@ -106,7 +106,8 @@
## Run, w/ differential returned by function ('jac' option) ##########
## Minimum wrt 'x' is y0
-
+try,
+ ## option 'jac' is not supported anymore
opt = optimset_compat ("GradO","on");
[xlev,vlev,nlev] = fminunc_compat ("d2ff",x0,opt,y0,1);
@@ -118,8 +119,8 @@
ok = 0;
elseif verbose, prn ("ok %i\n",cnt);
end
+end;
octave:57> clear; verbose=1; test_d2_min_1
Testing d2_min () on a quadratic programming problem
Going to call d2_min
d2_min should find in one iteration + one more to check
d2_min : niter=0 nev=2 nobs=21 ,nparams=21
time=0 errx=3.75 minv=NaN
Too many iterations for this function
Error is too big : 3.75
Going to call d2_min()
d2_min should find in one iteration + one more to check
d2_min : niter=0 nev=2 nobs=21 ,nparams=21
time=0 errx=3.75 minv=NaN
Too many iterations for this function
Error is too big : 3.75
Some tests failed
octave:58> clear; verbose=1; test_d2_min_2
Testing d2_min () on a quadratic programming problem
Going to call d2_min()
d2_min should find in one iteration + one more to check
d2_min : niter=0 nev=2 nobs=21 nparams=19
time=0 errx=3.47 minv=NaN
Too many iterations for this function
Error is too big : 3.47
octave:59> clear; verbose=1; test_d2_min_3
Testing d2_min () on a strange 2-dimensional function
Going to call d2_min
d2_min should find minv = 0 (plus a little error)
d2_min : niter=0 nev=2 nparams=2
time=0 errx=2.92 minv=NaN
Error is too big : 2.92
octave:60> clear; verbose=1; test_fminunc_1
Testing that fminunc_compat() works as it should
Nparams = N = 2
nelder_mead : Too many iterations. Returning
ok 1
ok 2
nelder_mead : Too many iterations. Returning
ok 3
Error is too big : 2.07
octave:61> clear; verbose=1; test_min_1
Testing 'bfgsmin' on a quadratic programming problem
Set 'optim_func' to the name of the optimization
function you want to test (must have same synopsis
as 'bfgs')
Dimension is 28
Condition is 121.000000
nev=1 N=28 errx=3.29e-06 errv=3.02e-12
ok 1
ok 2
All tests ok
octave:62> clear; verbose=1; test_min_2
Testing bfgsmin on a quadratic problem
Set 'optim_func' to the name of the optimization
function you want to test (must have same synopsis
as 'bfgs')
Nparams = P = 15, Nobses = R = 20
ok 1
Costs : init= 177, final= 1, best= 1
time : 0.23
All tests ok (there's just one test)
octave:63> clear; verbose=1; test_min_3
Checking that extra arguments are accepted
Set 'optim_func' to the name of the optimization
function you want to test (must have same synopsis
as 'bfgs')
Tested function : bfgsmin
Nparams = P = 2, Nobses = R = 3
Costs : init= 9.67, final= 1, best= 1
time : 0.02
All tests ok
octave:64> clear; verbose=1; test_min_4
Testing that bfgsmin accepts struct control variable
Set 'optim_func' to the name of the optimization
function you want to test (must have same synopsis
as 'bfgsmin')
Nparams = N = 2
ok 1
ok 2
ok 3
All tests ok
octave:65> clear; verbose=1; test_minimize_1
Testing that minimize() works as it should
Nparams = N = 2
nelder_mead : Too many iterations. Returning
ok 1
ok 2
nelder_mead : Too many iterations. Returning
ok 3
Error is too big : 1.42
Error is too big : 1.42
Error is too big : 1.42
Error is too big : 1.42
ok 8
ok 9
ok 10
octave:66> clear; verbose=1; test_nelder_mead_min_1
test_nelder_mead :
optimization problem has dimension 6
trying to minimize 'my_quad_func'
nelder_mead : Too many iterations. Returning
ok 1
function evaluations = 10002
nelder_mead : Too many iterations. Returning
ok 2
function evaluations = 10002
nelder_mead : Too many iterations. Returning
ok 3
function evaluations = 10001
nelder_mead : Too many iterations. Returning
ok 4
function evaluations = 10001
trying to minimize 'non_quad_func_1'
ok 5
function evaluations = 1112
ok 6
function evaluations = 1112
ok 7
function evaluations = 1016
ok 8
function evaluations = 1016
trying to minimize 'non_quad_func_2'
nelder_mead : Too many iterations. Returning
ok 9
function evaluations = 10002
nelder_mead : Too many iterations. Returning
ok 10
function evaluations = 10002
nelder_mead : Too many iterations. Returning
ok 11
function evaluations = 10009
nelder_mead : Too many iterations. Returning
ok 12
function evaluations = 10009
All tests ok
octave:67> clear; verbose=1; test_nelder_mead_min_2
test_nelder_mead_2
Check whether nelder_mead_min accepts options properly
nev=3 imin=2 ymin=0 done=1. Done
ok 1
nev=3 imin=2 ymin=0 done=0
nev=5 imin=2 ymin=0 done=1. Done
ok 2
nev=3 imin=2 ymin=0 done=1. Done
ok 3
nev=3 imin=2 ymin=0 done=0
nev=101 imin=2 ymin=0 done=0
nev=201 imin=2 ymin=0 done=0
nev=301 imin=2 ymin=0 done=0
nev=401 imin=2 ymin=0 done=0
nev=501 imin=2 ymin=0 done=0
nev=601 imin=2 ymin=0 done=0
nev=701 imin=2 ymin=0 done=0
nev=801 imin=2 ymin=0 done=0
nev=901 imin=2 ymin=0 done=0
nev=1001 imin=2 ymin=0 done=0
nev=1101 imin=1 ymin=0 done=0
nev=1201 imin=1 ymin=0 done=0
nev=1301 imin=1 ymin=0 done=0
nev=1401 imin=1 ymin=0 done=0
nev=1501 imin=1 ymin=0 done=0
nev=1601 imin=1 ymin=0 done=0
nev=1701 imin=1 ymin=0 done=0
nev=1801 imin=1 ymin=0 done=0
nev=1901 imin=1 ymin=0 done=0
nev=2001 imin=1 ymin=0 done=0
nev=2101 imin=1 ymin=0 done=0
nev=2201 imin=1 ymin=0 done=0
nev=2301 imin=1 ymin=0 done=0
nev=2401 imin=1 ymin=0 done=0
nev=2501 imin=1 ymin=0 done=0
nev=2601 imin=1 ymin=0 done=0
nev=2701 imin=1 ymin=0 done=0
nev=2801 imin=1 ymin=0 done=0
nev=2901 imin=1 ymin=0 done=0
nev=3001 imin=1 ymin=0 done=0
nev=3101 imin=1 ymin=0 done=0
nev=3201 imin=1 ymin=0 done=0
nev=3301 imin=1 ymin=0 done=0
nev=3401 imin=1 ymin=0 done=0
nev=3501 imin=1 ymin=0 done=0
nev=3601 imin=1 ymin=0 done=0
nev=3701 imin=1 ymin=0 done=0
nev=3801 imin=1 ymin=0 done=0
nev=3901 imin=1 ymin=0 done=0
nev=4001 imin=1 ymin=0 done=0
nev=4101 imin=1 ymin=0 done=0
nev=4201 imin=1 ymin=0 done=0
nev=4301 imin=1 ymin=0 done=0
nev=4401 imin=1 ymin=0 done=0
nev=4501 imin=1 ymin=0 done=0
nev=4601 imin=1 ymin=0 done=0
nev=4701 imin=1 ymin=0 done=0
nev=4801 imin=1 ymin=0 done=0
nev=4901 imin=1 ymin=0 done=0
nev=5001 imin=1 ymin=0 done=0
nev=5101 imin=1 ymin=0 done=0
nev=5201 imin=1 ymin=0 done=0
nev=5301 imin=1 ymin=0 done=0
nev=5401 imin=1 ymin=0 done=0
nev=5501 imin=1 ymin=0 done=0
nev=5601 imin=1 ymin=0 done=0
nev=5701 imin=1 ymin=0 done=0
nev=5801 imin=1 ymin=0 done=0
nev=5901 imin=1 ymin=0 done=0
nev=6001 imin=1 ymin=0 done=0
nev=6101 imin=1 ymin=0 done=0
nev=6201 imin=1 ymin=0 done=0
nev=6301 imin=1 ymin=0 done=0
nev=6401 imin=1 ymin=0 done=0
nev=6501 imin=1 ymin=0 done=0
nev=6601 imin=1 ymin=0 done=0
nev=6701 imin=1 ymin=0 done=0
nev=6801 imin=1 ymin=0 done=0
nev=6901 imin=1 ymin=0 done=0
nev=7001 imin=1 ymin=0 done=0
nev=7101 imin=1 ymin=0 done=0
nev=7201 imin=1 ymin=0 done=0
nev=7301 imin=1 ymin=0 done=0
nev=7401 imin=1 ymin=0 done=0
nev=7501 imin=1 ymin=0 done=0
nev=7601 imin=1 ymin=0 done=0
nev=7701 imin=1 ymin=0 done=0
nev=7801 imin=1 ymin=0 done=0
nev=7901 imin=1 ymin=0 done=0
nev=8001 imin=1 ymin=0 done=0
nev=8101 imin=1 ymin=0 done=0
nev=8201 imin=1 ymin=0 done=0
nev=8301 imin=1 ymin=0 done=0
nev=8401 imin=1 ymin=0 done=0
nev=8501 imin=1 ymin=0 done=0
nev=8601 imin=1 ymin=0 done=0
nev=8701 imin=1 ymin=0 done=0
nev=8801 imin=1 ymin=0 done=0
nev=8901 imin=1 ymin=0 done=0
nev=9001 imin=1 ymin=0 done=0
nev=9101 imin=1 ymin=0 done=0
nev=9201 imin=1 ymin=0 done=0
nev=9301 imin=1 ymin=0 done=0
nev=9401 imin=1 ymin=0 done=0
nev=9501 imin=1 ymin=0 done=0
nev=9601 imin=1 ymin=0 done=0
nev=9701 imin=1 ymin=0 done=0
nev=9801 imin=1 ymin=0 done=0
nev=9901 imin=1 ymin=0 done=0
nelder_mead : Too many iterations. Returning
ok 4
nev=3 imin=2 ymin=0 done=0
nev=101 imin=2 ymin=0 done=0
nev=201 imin=2 ymin=0 done=0
nev=301 imin=2 ymin=0 done=0
nev=401 imin=2 ymin=0 done=0
nev=501 imin=2 ymin=0 done=0
nev=601 imin=2 ymin=0 done=0
nev=701 imin=2 ymin=0 done=0
nev=801 imin=2 ymin=0 done=0
nev=901 imin=2 ymin=0 done=0
nev=1001 imin=2 ymin=0 done=0
nev=1101 imin=1 ymin=0 done=0
nev=1201 imin=1 ymin=0 done=0
nev=1301 imin=1 ymin=0 done=0
nev=1401 imin=1 ymin=0 done=0
nev=1501 imin=1 ymin=0 done=0
nev=1601 imin=1 ymin=0 done=0
nev=1701 imin=1 ymin=0 done=0
nev=1801 imin=1 ymin=0 done=0
nev=1901 imin=1 ymin=0 done=0
nev=2001 imin=1 ymin=0 done=0
nev=2101 imin=1 ymin=0 done=0
nev=2201 imin=1 ymin=0 done=0
nev=2301 imin=1 ymin=0 done=0
nev=2401 imin=1 ymin=0 done=0
nev=2501 imin=1 ymin=0 done=0
nev=2601 imin=1 ymin=0 done=0
nev=2701 imin=1 ymin=0 done=0
nev=2801 imin=1 ymin=0 done=0
nev=2901 imin=1 ymin=0 done=0
nev=3001 imin=1 ymin=0 done=0
nev=3101 imin=1 ymin=0 done=0
nev=3201 imin=1 ymin=0 done=0
nev=3301 imin=1 ymin=0 done=0
nev=3401 imin=1 ymin=0 done=0
nev=3501 imin=1 ymin=0 done=0
nev=3601 imin=1 ymin=0 done=0
nev=3701 imin=1 ymin=0 done=0
nev=3801 imin=1 ymin=0 done=0
nev=3901 imin=1 ymin=0 done=0
nev=4001 imin=1 ymin=0 done=0
nev=4101 imin=1 ymin=0 done=0
nev=4201 imin=1 ymin=0 done=0
nev=4301 imin=1 ymin=0 done=0
nev=4401 imin=1 ymin=0 done=0
nev=4501 imin=1 ymin=0 done=0
nev=4601 imin=1 ymin=0 done=0
nev=4701 imin=1 ymin=0 done=0
nev=4801 imin=1 ymin=0 done=0
nev=4901 imin=1 ymin=0 done=0
nev=5001 imin=1 ymin=0 done=0
nev=5101 imin=1 ymin=0 done=0
nev=5201 imin=1 ymin=0 done=0
nev=5301 imin=1 ymin=0 done=0
nev=5401 imin=1 ymin=0 done=0
nev=5501 imin=1 ymin=0 done=0
nev=5601 imin=1 ymin=0 done=0
nev=5701 imin=1 ymin=0 done=0
nev=5801 imin=1 ymin=0 done=0
nev=5901 imin=1 ymin=0 done=0
nev=6001 imin=1 ymin=0 done=0
nev=6101 imin=1 ymin=0 done=0
nev=6201 imin=1 ymin=0 done=0
nev=6301 imin=1 ymin=0 done=0
nev=6401 imin=1 ymin=0 done=0
nev=6501 imin=1 ymin=0 done=0
nev=6601 imin=1 ymin=0 done=0
nev=6701 imin=1 ymin=0 done=0
nev=6801 imin=1 ymin=0 done=0
nev=6901 imin=1 ymin=0 done=0
nev=7001 imin=1 ymin=0 done=0
nev=7101 imin=1 ymin=0 done=0
nev=7201 imin=1 ymin=0 done=0
nev=7301 imin=1 ymin=0 done=0
nev=7401 imin=1 ymin=0 done=0
nev=7501 imin=1 ymin=0 done=0
nev=7601 imin=1 ymin=0 done=0
nev=7701 imin=1 ymin=0 done=0
nev=7801 imin=1 ymin=0 done=0
nev=7901 imin=1 ymin=0 done=0
nev=8001 imin=1 ymin=0 done=0
nev=8101 imin=1 ymin=0 done=0
nev=8201 imin=1 ymin=0 done=0
nev=8301 imin=1 ymin=0 done=0
nev=8401 imin=1 ymin=0 done=0
nev=8501 imin=1 ymin=0 done=0
nev=8601 imin=1 ymin=0 done=0
nev=8701 imin=1 ymin=0 done=0
nev=8801 imin=1 ymin=0 done=0
nev=8901 imin=1 ymin=0 done=0
nev=9001 imin=1 ymin=0 done=0
nev=9101 imin=1 ymin=0 done=0
nev=9201 imin=1 ymin=0 done=0
nev=9301 imin=1 ymin=0 done=0
nev=9401 imin=1 ymin=0 done=0
nev=9501 imin=1 ymin=0 done=0
nev=9601 imin=1 ymin=0 done=0
nev=9701 imin=1 ymin=0 done=0
nev=9801 imin=1 ymin=0 done=0
nev=9901 imin=1 ymin=0 done=0
nelder_mead : Too many iterations. Returning
not ok 5
nev=3 imin=2 ymin=0 done=0
nev=101 imin=2 ymin=0 done=0
nev=201 imin=2 ymin=0 done=0
nev=301 imin=2 ymin=0 done=0
nev=401 imin=2 ymin=0 done=0
nev=501 imin=2 ymin=0 done=0
nev=601 imin=2 ymin=0 done=0
nev=701 imin=2 ymin=0 done=0
nev=801 imin=2 ymin=0 done=0
nev=901 imin=2 ymin=0 done=0
nev=1001 imin=2 ymin=0 done=0
nev=1101 imin=1 ymin=0 done=0
nev=1201 imin=1 ymin=0 done=0
nev=1301 imin=1 ymin=0 done=0
nev=1401 imin=1 ymin=0 done=0
nev=1501 imin=1 ymin=0 done=0
nev=1601 imin=1 ymin=0 done=0
nev=1701 imin=1 ymin=0 done=0
nev=1801 imin=1 ymin=0 done=0
nev=1901 imin=1 ymin=0 done=0
nev=2001 imin=1 ymin=0 done=0
nev=2101 imin=1 ymin=0 done=0
nev=2201 imin=1 ymin=0 done=0
nev=2301 imin=1 ymin=0 done=0
nev=2401 imin=1 ymin=0 done=0
nev=2501 imin=1 ymin=0 done=0
nev=2601 imin=1 ymin=0 done=0
nev=2701 imin=1 ymin=0 done=0
nev=2801 imin=1 ymin=0 done=0
nev=2901 imin=1 ymin=0 done=0
nev=3001 imin=1 ymin=0 done=0
nev=3101 imin=1 ymin=0 done=0
nev=3201 imin=1 ymin=0 done=0
nev=3301 imin=1 ymin=0 done=0
nev=3401 imin=1 ymin=0 done=0
nev=3501 imin=1 ymin=0 done=0
nev=3601 imin=1 ymin=0 done=0
nev=3701 imin=1 ymin=0 done=0
nev=3801 imin=1 ymin=0 done=0
nev=3901 imin=1 ymin=0 done=0
nev=4001 imin=1 ymin=0 done=0
nev=4101 imin=1 ymin=0 done=0
nev=4201 imin=1 ymin=0 done=0
nev=4301 imin=1 ymin=0 done=0
nev=4401 imin=1 ymin=0 done=0
nev=4501 imin=1 ymin=0 done=0
nev=4601 imin=1 ymin=0 done=0
nev=4701 imin=1 ymin=0 done=0
nev=4801 imin=1 ymin=0 done=0
nev=4901 imin=1 ymin=0 done=0
nev=5001 imin=1 ymin=0 done=0
nev=5101 imin=1 ymin=0 done=0
nev=5201 imin=1 ymin=0 done=0
nev=5301 imin=1 ymin=0 done=0
nev=5401 imin=1 ymin=0 done=0
nev=5501 imin=1 ymin=0 done=0
nev=5601 imin=1 ymin=0 done=0
nev=5701 imin=1 ymin=0 done=0
nev=5801 imin=1 ymin=0 done=0
nev=5901 imin=1 ymin=0 done=0
nev=6001 imin=1 ymin=0 done=0
nev=6101 imin=1 ymin=0 done=0
nev=6201 imin=1 ymin=0 done=0
nev=6301 imin=1 ymin=0 done=0
nev=6401 imin=1 ymin=0 done=0
nev=6501 imin=1 ymin=0 done=0
nev=6601 imin=1 ymin=0 done=0
nev=6701 imin=1 ymin=0 done=0
nev=6801 imin=1 ymin=0 done=0
nev=6901 imin=1 ymin=0 done=0
nev=7001 imin=1 ymin=0 done=0
nev=7101 imin=1 ymin=0 done=0
nev=7201 imin=1 ymin=0 done=0
nev=7301 imin=1 ymin=0 done=0
nev=7401 imin=1 ymin=0 done=0
nev=7501 imin=1 ymin=0 done=0
nev=7601 imin=1 ymin=0 done=0
nev=7701 imin=1 ymin=0 done=0
nev=7801 imin=1 ymin=0 done=0
nev=7901 imin=1 ymin=0 done=0
nev=8001 imin=1 ymin=0 done=0
nev=8101 imin=1 ymin=0 done=0
nev=8201 imin=1 ymin=0 done=0
nev=8301 imin=1 ymin=0 done=0
nev=8401 imin=1 ymin=0 done=0
nev=8501 imin=1 ymin=0 done=0
nev=8601 imin=1 ymin=0 done=0
nev=8701 imin=1 ymin=0 done=0
nev=8801 imin=1 ymin=0 done=0
nev=8901 imin=1 ymin=0 done=0
nev=9001 imin=1 ymin=0 done=0
nev=9101 imin=1 ymin=0 done=0
nev=9201 imin=1 ymin=0 done=0
nev=9301 imin=1 ymin=0 done=0
nev=9401 imin=1 ymin=0 done=0
nev=9501 imin=1 ymin=0 done=0
nev=9601 imin=1 ymin=0 done=0
nev=9701 imin=1 ymin=0 done=0
nev=9801 imin=1 ymin=0 done=0
nev=9901 imin=1 ymin=0 done=0
nelder_mead : Too many iterations. Returning
ok 6
octave:68> clear; verbose=1; test_wpolyfit
Polynomial: 115.16*x^1 + 0.047028 [ p(chi^2>observed)=0.00% ]
See residuals? [y,n]
computing parameter uncertainty from monte carlo simulation...
Coefficient Error
117.874 1.74422
0.0407012 0.00395322
Press <Enter> to see some sample regression lines:
Press <Enter> to continue with the tests:
Filippelli, A., NIST.
parameter certified value rel. error
ans =
-4.0296e-05 -4.0296e-05 2.5815e-08
-2.4678e-03 -2.4678e-03 2.4846e-08
-6.7019e-02 -6.7019e-02 2.3962e-08
-1.0622e+00 -1.0622e+00 2.3160e-08
-1.0875e+01 -1.0875e+01 2.2437e-08
-7.5124e+01 -7.5124e+01 2.1789e-08
-3.5448e+02 -3.5448e+02 2.1210e-08
-1.1280e+03 -1.1280e+03 2.0694e-08
-2.3164e+03 -2.3164e+03 2.0234e-08
-2.7722e+03 -2.7722e+03 1.9823e-08
-1.4675e+03 -1.4675e+03 1.9467e-08
p-error certified value rel. error
ans =
8.9663e-06 8.9663e-06 2.2489e-08
5.3562e-04 5.3562e-04 2.2232e-08
1.4236e-02 1.4236e-02 2.1931e-08
2.2162e-01 2.2162e-01 2.1585e-08
2.2369e+00 2.2369e+00 2.1197e-08
1.5290e+01 1.5290e+01 2.0768e-08
7.1648e+01 7.1648e+01 2.0301e-08
2.2720e+02 2.2720e+02 1.9801e-08
4.6648e+02 4.6648e+02 1.9270e-08
5.5978e+02 5.5978e+02 1.8716e-08
2.9808e+02 2.9808e+02 1.8143e-08
Press <Enter> to proceed to the next test
Pontius, P., NIST
parameter certified value rel. error
ans =
-3.1608e-15 -3.1608e-15 1.9467e-14
7.3206e-07 7.3206e-07 4.3390e-16
6.7357e-04 6.7357e-04 2.7316e-13
p-error certified value rel. error
ans =
4.8665e-17 4.8665e-17 3.9005e-14
1.5782e-10 1.5782e-10 4.1112e-14
1.0794e-04 1.0794e-04 3.9425e-14
Press <Enter> to proceed to the next test
Eberhardt, K., NIST
parameter certified value rel. error
ans =
2.07438 2.07438 0.00000
0.00000 0.00000 NaN
p-error certified value rel. error
ans =
0.016529 0.016529 0.000000
0.000000 0.000000 NaN
Press <Enter> to proceed to the next test
Wampler1
parameter certified value rel. error
ans =
1.0000e+00 1.0000e+00 1.6653e-14
1.0000e+00 1.0000e+00 8.6264e-13
1.0000e+00 1.0000e+00 1.6126e-11
1.0000e+00 1.0000e+00 1.2987e-10
1.0000e+00 1.0000e+00 4.0824e-10
1.0000e+00 1.0000e+00 3.0358e-10
p-error certified value rel. error
ans =
4.6842e-15 0.0000e+00 Inf
2.3543e-13 0.0000e+00 Inf
4.2317e-12 0.0000e+00 Inf
3.2500e-11 0.0000e+00 Inf
9.8565e-11 0.0000e+00 Inf
8.9756e-11 0.0000e+00 Inf
Press <Enter> to proceed to the next test
Wampler2
parameter certified value rel. error
ans =
1.0000e-05 1.0000e-05 3.5237e-14
1.0000e-04 1.0000e-04 1.8038e-13
1.0000e-03 1.0000e-03 3.2548e-13
1.0000e-02 1.0000e-02 2.4702e-13
1.0000e-01 1.0000e-01 6.9389e-14
1.0000e+00 1.0000e+00 3.3307e-15
p-error certified value rel. error
ans =
1.4802e-19 0.0000e+00 Inf
7.4399e-18 0.0000e+00 Inf
1.3372e-16 0.0000e+00 Inf
1.0270e-15 0.0000e+00 Inf
3.1147e-15 0.0000e+00 Inf
2.8363e-15 0.0000e+00 Inf
Press <Enter> to proceed to the next test
Wampler3
parameter certified value rel. error
ans =
1.0000e+00 1.0000e+00 1.4433e-14
1.0000e+00 1.0000e+00 7.5340e-13
1.0000e+00 1.0000e+00 1.4138e-11
1.0000e+00 1.0000e+00 1.1475e-10
1.0000e+00 1.0000e+00 3.6597e-10
1.0000e+00 1.0000e+00 2.8173e-10
p-error certified value rel. error
ans =
1.1232e-01 1.1232e-01 2.5451e-14
5.6457e+00 5.6457e+00 2.6587e-14
1.0148e+02 1.0148e+02 2.6048e-14
7.7934e+02 7.7934e+02 3.3405e-14
2.3636e+03 2.3636e+03 3.6748e-14
2.1523e+03 2.1523e+03 2.9579e-14
Press <Enter> to proceed to the next test
Wampler4
parameter certified value rel. error
ans =
1.0000e+00 1.0000e+00 1.9607e-13
1.0000e+00 1.0000e+00 1.0056e-11
1.0000e+00 1.0000e+00 1.8266e-10
1.0000e+00 1.0000e+00 1.3825e-09
1.0000e+00 1.0000e+00 3.8201e-09
1.0000e+00 1.0000e+00 1.8820e-09
p-error certified value rel. error
ans =
1.1232e+01 1.1232e+01 2.1350e-14
5.6457e+02 5.6457e+02 2.2151e-14
1.0148e+04 1.0148e+04 2.1869e-14
7.7934e+04 7.7934e+04 2.9128e-14
2.3636e+05 2.3636e+05 3.2631e-14
2.1523e+05 2.1523e+05 2.5421e-14
Press <Enter> to proceed to the next test
Wampler5
parameter certified value rel. error
ans =
1.0000e+00 1.0000e+00 2.1246e-11
1.0000e+00 1.0000e+00 1.0910e-09
1.0000e+00 1.0000e+00 1.9863e-08
1.0000e+00 1.0000e+00 1.5111e-07
1.0000e+00 1.0000e+00 4.2242e-07
1.0000e+00 1.0000e+00 2.1826e-07
p-error certified value rel. error
ans =
1.1232e+03 1.1232e+03 2.1255e-14
5.6457e+04 5.6457e+04 2.2296e-14
1.0148e+06 1.0148e+06 2.1912e-14
7.7934e+06 7.7934e+06 2.9278e-14
2.3636e+07 2.3636e+07 3.2626e-14
2.1523e+07 2.1523e+07 2.5443e-14
Press <Enter> to proceed to the next test
octave:69>
octave:69>
octave:69>
octave:69>
octave:69> diary off
------------------------------------------------------------------------------
_______________________________________________
Octave-dev mailing list
Octave-dev@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/octave-dev
