|
»
»
result of "0^ones(n)" shows obscure dependency on dimension n
|
View:
New views
6 Messages
—
Rating Filter:
Alert me
|
 |
result of "0^ones(n)" shows obscure dependency on dimension n
by Rolf Fabian
::
Rate this Message:
octave-3.0.0.exe:> 0^ones(1)
ans = 0 octave-3.0.0.exe:> 0^ones(2) ans = 0.50000 -0.50000 -0.50000 0.50000 octave-3.0.0.exe:> 0^ones(3) ans = NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi octave-3.0.0.exe:> 0^ones(4) ans = NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi octave-3.0.0.exe:> 0^ones(5) ans = 0.80000 -0.20000 -0.20000 -0.20000 -0.20000 -0.20000 0.80000 -0.20000 -0.20000 -0.20000 -0.20000 -0.20000 0.80000 -0.20000 -0.20000 -0.20000 -0.20000 -0.20000 0.80000 -0.20000 -0.20000 -0.20000 -0.20000 -0.20000 0.80000 Is this a feature ? I bet it isn't, even if I don't have access to MatLab in order to check its output for thoses cases against above Octave's results. Rolf Fabian < r dot fabian at jacobs-university dot de> Rolf Fabian
<r dot fabian at jacobs-university dot de> |
|
|
Re: result of "0^ones(n)" shows obscure dependency on dimension n
by Ben Abbott
::
Rate this Message:
On Monday, January 14, 2008, at 10:19AM, "Rolf Fabian" <Rolf.Fabian@...> wrote:
> >octave-3.0.0.exe:> 0^ones(1) >ans = 0 >octave-3.0.0.exe:> 0^ones(2) >ans = > 0.50000 -0.50000 > -0.50000 0.50000 > >octave-3.0.0.exe:> 0^ones(3) >ans = > NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi > >octave-3.0.0.exe:> 0^ones(4) >ans = > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > >octave-3.0.0.exe:> 0^ones(5) >ans = > 0.80000 -0.20000 -0.20000 -0.20000 -0.20000 > -0.20000 0.80000 -0.20000 -0.20000 -0.20000 > -0.20000 -0.20000 0.80000 -0.20000 -0.20000 > -0.20000 -0.20000 -0.20000 0.80000 -0.20000 > -0.20000 -0.20000 -0.20000 -0.20000 0.80000 > >Is this a feature ? > >I bet it isn't, even if I don't have access to MatLab in order >to check its output for thoses cases against above Octave's >results. > >Rolf Fabian > >< r dot fabian at jacobs-university dot de> I compared to Matlab 2007b ... it appears to have a similar "feature" as well. >> for n=1:10 x=0^ones(n,n); disp(x) end 0 0.500000000000000 -0.500000000000000 -0.500000000000000 0.500000000000000 NaN Inf -Inf NaN Inf -Inf NaN -Inf Inf NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN Inf -Inf NaN NaN NaN NaN Inf -Inf NaN NaN NaN NaN Inf -Inf NaN NaN NaN NaN Inf -Inf NaN NaN NaN NaN Inf -Inf NaN NaN NaN NaN -Inf Inf NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN -Inf NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN Inf NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN Inf NaN NaN NaN NaN NaN NaN NaN NaN Inf NaN NaN NaN NaN NaN NaN NaN NaN Inf NaN NaN NaN NaN NaN NaN NaN NaN Inf NaN NaN NaN NaN NaN NaN NaN NaN Inf NaN NaN NaN NaN NaN NaN NaN NaN Inf NaN NaN NaN NaN NaN NaN NaN NaN -Inf NaN NaN NaN NaN NaN NaN NaN NaN -Inf NaN NaN NaN NaN NaN NaN NaN NaN Inf Columns 1 through 5 0.888888888888889 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 0.888888888888889 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 0.888888888888889 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 0.888888888888889 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 0.888888888888889 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 0 0 0 0 0 Columns 6 through 10 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 0.000000000000000 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.000000000000000 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.000000000000000 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.000000000000000 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.000000000000000 0.888888888888889 -0.111111111111111 -0.111111111111111 -0.111111111111111 -0.000000000000000 -0.111111111111111 0.888888888888889 -0.111111111111111 -0.111111111111111 -0.000000000000000 -0.111111111111111 -0.111111111111111 0.888888888888889 -0.111111111111111 -0.000000000000000 -0.111111111111111 -0.111111111111111 -0.111111111111111 0.888888888888889 0.000000000000000 0 0 0 0 0 >> Ben _______________________________________________ Bug-octave mailing list Bug-octave@... https://www.cae.wisc.edu/mailman/listinfo/bug-octave |
|
|
Re: result of "0^ones(n)" shows obscure dependency on dimension n
by Rolf Fabian
::
Rate this Message:
OK, lost bet.
But I can live with that lost bet I noticed also big differences between this 'feature' in Octave and the 'similar feature' in MatLab. Now I'm really curious about what will happen. Will Octave clone the 'similar feature' of MatLab into version 3.1 ? In my eyes it's actually a bug for both programs. I don't consider to report it to mathworks because I'm not using MatLab at all. Thanks Ben for checking this stuff with matlab. Rolf Fabian
Rolf Fabian
<r dot fabian at jacobs-university dot de> |
|
|
Re: result of "0^ones(n)" shows obscure dependency on dimension n
by Sergei Steshenko-2
::
Rate this Message:
--- Rolf Fabian <Rolf.Fabian@...> wrote: > > octave-3.0.0.exe:> 0^ones(1) > ans = 0 > octave-3.0.0.exe:> 0^ones(2) > ans = > 0.50000 -0.50000 > -0.50000 0.50000 > > octave-3.0.0.exe:> 0^ones(3) > ans = > NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi > > octave-3.0.0.exe:> 0^ones(4) > ans = > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > > octave-3.0.0.exe:> 0^ones(5) > ans = > 0.80000 -0.20000 -0.20000 -0.20000 -0.20000 > -0.20000 0.80000 -0.20000 -0.20000 -0.20000 > -0.20000 -0.20000 0.80000 -0.20000 -0.20000 > -0.20000 -0.20000 -0.20000 0.80000 -0.20000 > -0.20000 -0.20000 -0.20000 -0.20000 0.80000 > > Is this a feature ? > > I bet it isn't, even if I don't have access to MatLab in order > to check its output for thoses cases against above Octave's > results. > > Rolf Fabian > > < r dot fabian at jacobs-university dot de> > > ----- > Rolf Fabian > <r dot fabian at jacobs-university dot de> > > -- > View this message in context: > > Sent from the Octave - Bugs mailing list archive at Nabble.com. > > _______________________________________________ > Bug-octave mailing list > Bug-octave@... > https://www.cae.wisc.edu/mailman/listinfo/bug-octave > Interestingly enough, my results (I'm on Linux, octave-3.0.0 built by myself from sources from scratch) differ from yours - see the "0^ones(5)" ones: " octave:1> 0^ones(1) ans = 0 octave:2> 0^ones(2) ans = 0.50000 -0.50000 -0.50000 0.50000 octave:3> 0^ones(3) ans = NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi octave:4> 0^ones(4) ans = NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi octave:5> 0^ones(5) ans = NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi octave:6> ". It remains "NaN - NaNi" until at least ones(10) - haven't checked further. Regards, Sergei. Applications From Scratch: http://appsfromscratch.berlios.de/ ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs _______________________________________________ Bug-octave mailing list Bug-octave@... https://www.cae.wisc.edu/mailman/listinfo/bug-octave |
|
|
Re: result of "0^ones(n)" shows obscure dependency on dimension n
by Sergei Steshenko-2
::
Rate this Message:
--- Sergei Steshenko <sergstesh@...> wrote: > > --- Rolf Fabian <Rolf.Fabian@...> wrote: > > > > > octave-3.0.0.exe:> 0^ones(1) > > ans = 0 > > octave-3.0.0.exe:> 0^ones(2) > > ans = > > 0.50000 -0.50000 > > -0.50000 0.50000 > > > > octave-3.0.0.exe:> 0^ones(3) > > ans = > > NaN - NaNi NaN - NaNi NaN - NaNi > > NaN - NaNi NaN - NaNi NaN - NaNi > > NaN - NaNi NaN - NaNi NaN - NaNi > > > > octave-3.0.0.exe:> 0^ones(4) > > ans = > > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > > > > octave-3.0.0.exe:> 0^ones(5) > > ans = > > 0.80000 -0.20000 -0.20000 -0.20000 -0.20000 > > -0.20000 0.80000 -0.20000 -0.20000 -0.20000 > > -0.20000 -0.20000 0.80000 -0.20000 -0.20000 > > -0.20000 -0.20000 -0.20000 0.80000 -0.20000 > > -0.20000 -0.20000 -0.20000 -0.20000 0.80000 > > > > Is this a feature ? > > > > I bet it isn't, even if I don't have access to MatLab in order > > to check its output for thoses cases against above Octave's > > results. > > > > Rolf Fabian > > > > < r dot fabian at jacobs-university dot de> > > > > ----- > > Rolf Fabian > > <r dot fabian at jacobs-university dot de> > > > > -- > > View this message in context: > > > > > Sent from the Octave - Bugs mailing list archive at Nabble.com. > > > > _______________________________________________ > > Bug-octave mailing list > > Bug-octave@... > > https://www.cae.wisc.edu/mailman/listinfo/bug-octave > > > > Interestingly enough, my results (I'm on Linux, octave-3.0.0 built by myself from > sources from scratch) differ from yours - see the "0^ones(5)" ones: > > " > octave:1> 0^ones(1) > ans = 0 > octave:2> 0^ones(2) > ans = > > 0.50000 -0.50000 > -0.50000 0.50000 > > octave:3> 0^ones(3) > ans = > > NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi > > octave:4> 0^ones(4) > ans = > > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > > octave:5> 0^ones(5) > ans = > > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi NaN - NaNi > > octave:6> > ". > > It remains "NaN - NaNi" until at least ones(10) - haven't checked further. > > Regards, > Sergei. > > Applications From Scratch: http://appsfromscratch.berlios.de/ > > > ____________________________________________________________________________________ > Never miss a thing. Make Yahoo your home page. > http://www.yahoo.com/r/hs > _______________________________________________ > Bug-octave mailing list > Bug-octave@... > https://www.cae.wisc.edu/mailman/listinfo/bug-octave > It's even more severe: " octave:5> 0.00000000000000000001^ones(10) ans = 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 -0.100000 0.900000 octave:6> 0.00000000000000000001^ones(2) ans = 0.50000 -0.50000 -0.50000 0.50000 octave:7> (0.00000000000000000001 * ones(2)).^ones(2) ans = 1.0000e-20 1.0000e-20 1.0000e-20 1.0000e-20 ". - the last result is what I'd expect. I'm trying to say that the problem is not specific to exponentiation of 0. Regards, Sergei. Applications From Scratch: http://appsfromscratch.berlios.de/ ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping _______________________________________________ Bug-octave mailing list Bug-octave@... https://www.cae.wisc.edu/mailman/listinfo/bug-octave |
|
|
scalar1 ^ (scalar2 * ones(n)) problem
by Sergei Steshenko-2
::
Rate this Message:
Hello All,
this is a continuation of result of "0^ones(n)" shows obscure dependency on dimension n thread - maybe it should be in that thread. Here are some quick example - self-built octave-3.0.0 on "Linux amdam2 2.6.18.8-0.7-default #1 SMP Tue Oct 2 17:21:08 UTC 2007 i686 athlon i386 GNU/Linux" : " octave:2> 1^ones(1) ans = 1 octave:3> 1^ones(2) ans = 1 0 0 1 octave:4> 1^ones(3) ans = 1.0000e-00 -3.5453e-17 -1.4707e-16 8.4568e-18 1.0000e+00 -1.5233e-17 -1.0072e-16 -1.7672e-17 1.0000e+00 octave:5> 1^ones(4) ans = 1.0000e+00 5.2855e-17 -3.3515e-17 -3.3854e-17 -8.5977e-17 1.0000e+00 -7.9567e-17 -7.9879e-17 -2.4462e-17 -1.4867e-17 1.0000e+00 -2.5275e-17 -9.8527e-17 -8.8850e-17 -2.1142e-17 1.0000e+00 octave:6> ones(3) .^ ones(3) ans = 1 1 1 1 1 1 1 1 1 octave:7> (2 * ones(3)) .^ (0.5 * ones(3)) ans = 1.4142 1.4142 1.4142 1.4142 1.4142 1.4142 1.4142 1.4142 1.4142 octave:8> 1 ^ (0.5 * ones(3)) ans = 1.0000e-00 -3.5453e-17 -1.4707e-16 8.4568e-18 1.0000e+00 -1.5233e-17 -1.0072e-16 -1.7672e-17 1.0000e+00 octave:9> 2 ^ (0.5 * ones(3)) ans = 1.60948 0.60948 0.60948 0.60948 1.60948 0.60948 0.60948 0.60948 1.60948 " - I expecting results along the lines of " octave:6> ones(3) .^ ones(3) ans = 1 1 1 1 1 1 1 1 1 octave:7> (2 * ones(3)) .^ (0.5 * ones(3)) ans = 1.4142 1.4142 1.4142 1.4142 1.4142 1.4142 1.4142 1.4142 1.4142 ". Thanks, Sergei. Applications From Scratch: http://appsfromscratch.berlios.de/ ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping _______________________________________________ Bug-octave mailing list Bug-octave@... https://www.cae.wisc.edu/mailman/listinfo/bug-octave |
| Free embeddable forum powered by Nabble | Forum Help |
