Your message dated Sun, 28 Jul 2013 07:56:41 +0200
with message-id <[email protected]>
and subject line Re: Bug#702461: atlas: dot or eigenvalue rounding issues on
i386
has caused the Debian Bug report #702461,
regarding atlas: dot or eigenvalue rounding issues on i386
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)
--
702461: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=702461
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: atlas
Version: 3.8.4-9
Severity: normal
atlas seems to have pretty severe rounding issues on i386 dot products
or eigenvalues
a scipy test fails with atlas but not with netlib blas or openblas
amd64 is not affected:
update-alternatives --set libblas.so.3
/usr/lib/openblas-base/libopenblas.so.0
python /usr/lib/python2.7/dist-packages/scipy/linalg/tests/test_decomp.py
#works fine except for a known failure
update-alternatives --set libblas.so.3
/usr/lib/atlas-base/atlas/libblas.so.3
python /usr/lib/python2.7/dist-packages/scipy/linalg/tests/test_decomp.py
Arrays are not almost equal to 6 decimals
array([[22, 34, 31, 31, 17],
[45, 45, 42, 19, 29],
[39, 47, 49, 26, 34],
[27, 31, 26, 21, 15],
[38, 44, 44, 24, 30]])
array([[13, 26, 25, 17, 24],
[31, 46, 40, 26, 37],
[26, 40, 19, 25, 25],
[16, 25, 27, 14, 23],
[24, 35, 18, 21, 22]])
(mismatch 25.0%)
x: array([ -3.63672429e-01 +0.00000000e+00j,
-1.72783306e-15 -3.75494248e-08j,
-1.72783306e-15 +3.75494248e-08j, 2.00000000e+00
+0.00000000e+00j])
y: array([ -5.22147673e-01 +0.00000000e+00j,
2.32897931e-16 -2.61904476e-08j,
2.32897931e-16 +2.61904476e-08j, 2.00000000e+00
+0.00000000e+00j])
--- End Message ---
--- Begin Message ---
Version: 3.10.1-2
Le mercredi 06 mars 2013 à 22:08 +0100, Julian Taylor a écrit :
> Package: atlas
> Version: 3.8.4-9
> Severity: normal
>
> atlas seems to have pretty severe rounding issues on i386 dot products
> or eigenvalues
> a scipy test fails with atlas but not with netlib blas or openblas
> amd64 is not affected:
>
> update-alternatives --set libblas.so.3
> /usr/lib/openblas-base/libopenblas.so.0
> python /usr/lib/python2.7/dist-packages/scipy/linalg/tests/test_decomp.py
> #works fine except for a known failure
>
> update-alternatives --set libblas.so.3
> /usr/lib/atlas-base/atlas/libblas.so.3
>
> python /usr/lib/python2.7/dist-packages/scipy/linalg/tests/test_decomp.py
>
>
> Arrays are not almost equal to 6 decimals
>
> array([[22, 34, 31, 31, 17],
> [45, 45, 42, 19, 29],
> [39, 47, 49, 26, 34],
> [27, 31, 26, 21, 15],
> [38, 44, 44, 24, 30]])
> array([[13, 26, 25, 17, 24],
> [31, 46, 40, 26, 37],
> [26, 40, 19, 25, 25],
> [16, 25, 27, 14, 23],
> [24, 35, 18, 21, 22]])
> (mismatch 25.0%)
> x: array([ -3.63672429e-01 +0.00000000e+00j,
> -1.72783306e-15 -3.75494248e-08j,
> -1.72783306e-15 +3.75494248e-08j, 2.00000000e+00
> +0.00000000e+00j])
> y: array([ -5.22147673e-01 +0.00000000e+00j,
> 2.32897931e-16 -2.61904476e-08j,
> 2.32897931e-16 +2.61904476e-08j, 2.00000000e+00
> +0.00000000e+00j])
It seems this is now fixed: I cannot reproduce this issue with ATLAS
3.10. Closing.
--
.''`. Sébastien Villemot
: :' : Debian Developer
`. `' http://www.dynare.org/sebastien
`- GPG Key: 4096R/381A7594
signature.asc
Description: This is a digitally signed message part
--- End Message ---
