Dear list, I have a big collection of data from a catalytic reaction. I tried to export all the spectra (about 200) to a unique file containing the normalized ones. I used the "File>Save marked groups as>norm(E)" option. Unfortunately, Athena (Demeter 0.9.18) is not able to make one unique file (it crashes and I have to initialize it again), instead I had to creat 4 different files of 45 spectra each (the maximum group I could gather on a file). Is there any bug on it or some kind of limitation to creat these kind of files? Best regards, *Danilo OLIVEIRA DE SOUZA* (PhD) Université de Sciences et Technologies de Lille (Université Lille 1) UCCS - Unité de Catalise et Chimie du Solid, Bat. C3, 110. Cité Scientifique, 59655, VIlleneuve d'Ascq CEDEX, France. danilo.olsouza@gmail.com d.oliveira-de-souza@ed.univ-lille1.fr tel: +33 06.63.55.86.12
On 01/22/2014 10:40 AM, Danilo Oliveira de Souza wrote:
Dear list,
I have a big collection of data from a catalytic reaction. I tried to export all the spectra (about 200) to a unique file containing the normalized ones. I used the "File>Save marked groups as>norm(E)" option. Unfortunately, Athena (Demeter 0.9.18) is not able to make one unique file (it crashes and I have to initialize it again), instead I had to creat 4 different files of 45 spectra each (the maximum group I could gather on a file). Is there any bug on it or some kind of limitation to creat these kind of files?
The fact that Ifeffit, written in Fortran, has hard-wired limits on things has been discussed dozens of times on this mailing list. The advice to partition your data into groups of about 50 has been given on this mailing list dozens of times. Y'know, this mailing list has been archived since the beginning and can be searched. A quick skim of the Ifeffit source code (another thing anyone can do) suggests that the hard-wired limit for the maximum number of columns in an output data file is 99. I will try to preempt Matt's response and say that Larch addresses most of these limitations. A version of Athena which can talk to Larch (replacing Ifeffit as the computational backend) will be available soon. B -- Bruce Ravel ------------------------------------ bravel@bnl.gov National Institute of Standards and Technology Synchrotron Science Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage: http://xafs.org/BruceRavel Software: https://github.com/bruceravel
Hi Danilo,
On Wed, Jan 22, 2014 at 9:40 AM, Danilo Oliveira de Souza
Dear list,
I have a big collection of data from a catalytic reaction. I tried to export all the spectra (about 200) to a unique file containing the normalized ones. I used the "File>Save marked groups as>norm(E)" option. Unfortunately, Athena (Demeter 0.9.18) is not able to make one unique file (it crashes and I have to initialize it again), instead I had to creat 4 different files of 45 spectra each (the maximum group I could gather on a file). Is there any bug on it or some kind of limitation to creat these kind of files?
Best regards,
Danilo OLIVEIRA DE SOUZA (PhD) Université de Sciences et Technologies de Lille (Université Lille 1) UCCS - Unité de Catalise et Chimie du Solid, Bat. C3, 110. Cité Scientifique, 59655, VIlleneuve d'Ascq CEDEX, France.
danilo.olsouza@gmail.com d.oliveira-de-souza@ed.univ-lille1.fr tel: +33 06.63.55.86.12
_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
This is sort of "known problem". That is, we know that there will be some threshold where the underlying ifeffit engine will run out of memory, way before any reasonable laptop being used today would actually be short of memory. I can't say that we know that 200 spectra will never work, or that 45 will always work, it depends on the size of the arrays and how much processing you've done on the data. The problem is effectively removed with Larch, the next generation of the Ifeffit engine. I know Bruce is working on (and making progress on!) having Athena use Larch, which will be particularly useful for handling large data sets. For now, I think you're sort of stuck with having to grup the data as you are doing. As always, any help on Larch and Demeter would be greatly welcome. -- --Matt
participants (3)
-
Bruce Ravel -
Danilo Oliveira de Souza -
Matt Newville