Hi Bruce and Others, I've tried some follow-up troubleshooting. The problem does not occur on another laptop with Windows 7 installed, so it appears to fall under some variation of your "possibility #1." I thought perhaps it might be related to some kind of resource problem (e.g. memory allocation), so I played around with giving my Parallels virtual machine various amounts of memory, processing power, etc., but nothing changed the behavior. I would be interested to know from other people on the list if they have gotten the reconstruction of data from PCA to work under Windows 8 (I don't have another Windows 8 machine readily available to test it on). That would help narrow down whether the issue is a general one with that OS, or whether it's something more quirky having to do with the environment I'm using. --Scott Calvin Sarah Lawrence College On Jun 14, 2013, at 9:18 AM, Bruce Ravel wrote:

On 06/11/2013 09:25 PM, Scott Calvin wrote:

...

Athena is crashing every time I try to "Reconstruct Data" in the PCA dialog. It doesn't seem to matter what data I use, or what space I'm working in (mu or chi), or how many components I use in the reconstruction.

...

I'm using Demeter 0.9.17 with Ifeffit 1.2.11d, under Windows 8 running in Parallels 8 running under OS 10.8.4 on a MacBook Pro.

...

The log file for the latest crash is below. If you'd like a project file too, let me know, but it happens, e.g., with the gold cyanobacteria project file, and with every data set I've tried.

Hi Scott,

I do not see this behavior on any of my computers (which include native WinXp and Win7 machines and linux) and I am having trouble guessing from the log file you attached a possible cause of the problem.

Possibility #1: you are using a configuration that I cannot test. I have never used Parallels and none of my computers are running Windows 8. I'd be surprised if either is the cause of the problem, though. It seems surprising that Athena would get that deeply into its operations before running into a platform-related problem. That said, I'd be interested to know if you see this on a native Win7 or WinXP machine.

Possibility #2: you have figured out some combination of mouse clicks and button presses that I have never tested. The problem you are seeing is that the group considered current (i.e. the one that should be the target of the reconstruction) has become unset. Typically, when you click on an item in the group list, Athena will assign that as the current data set. It will then run a simple test to see if it was among the data groups included in the PCA decomposition. If so, the "reconstruct" button is enabled. If not, the TT button is enabled.

I cannot find a compbination of events that leaves me in a state that unsets this, resulting in your stack trace. So without further instruction, I don't know how to go about solving the problem.

B

Hi all, I seem to be having some trouble with Athena crashing during the reconstruction step for some of my data when doing PCA analysis, and was wondering if anyone might have some insight. I I have attached a project file with the spectra in question (I realize they probably aren't normalized as well as they could be) and the dathena.log file, which is similar to Dr. Calvin's reported error log, aside from the PATH information. However, this is not an issue with Parallels, as I am using a 64 bit version of Windows 7. Here, the issue I have is when reconstructing any of the files starting with TMF08-03 (in the attached "Combined - tailings.prj" project file). I am able to reconstruct all of the TMF08-01 files, and I have also successfully tried PCA/reconstruction on the files from the Au cyanobacteria example, so I do not believe it is an issue with my computer. To the extent that it may be an error with the TMF08-03 files, I have tried to re-import/re-work all the data, but this does not solve the problem, so it does not seems like the data itself is to blame. Perhaps the file names are presenting issues for Athena? (i.e. it might be confused if it is a standard or a sample?) Any help on this issue is greatly appreciated. I apologize in advance for any omitted information or poor error reporting on my part, and thank you for your patience and support with this issue. Cheers! John On Wed, Jun 19, 2013 at 4:17 PM, Erik Farquhar wrote:

Hi Scott, I had no problems doing reconstruction of data on two different machines running Windows 8 and DAthena 0.9.17. I used Bruce's gold cyanobacteria example and followed the sequence of operations shown in the PCA section of the Athena Users' Guide ( http://bruceravel.github.io/demeter/aug/analysis/pca.html). That would seem to point to something quirky about Parallels.

Erik

------------------------------ *From:* Scott Calvin *To:* XAFS Analysis using Ifeffit *Sent:* Wednesday, June 19, 2013 12:07 PM *Subject:* Re: [Ifeffit] DAthena crashes when reconstructing data from PCA

Hi Bruce and Others,

I've tried some follow-up troubleshooting.

The problem does not occur on another laptop with Windows 7 installed, so it appears to fall under some variation of your "possibility #1." I thought perhaps it might be related to some kind of resource problem (e.g. memory allocation), so I played around with giving my Parallels virtual machine various amounts of memory, processing power, etc., but nothing changed the behavior.

I would be interested to know from other people on the list if they have gotten the reconstruction of data from PCA to work under Windows 8 (I don't have another Windows 8 machine readily available to test it on). That would help narrow down whether the issue is a general one with that OS, or whether it's something more quirky having to do with the environment I'm using.

--Scott Calvin Sarah Lawrence College

On Jun 14, 2013, at 9:18 AM, Bruce Ravel wrote:

...

On 06/11/2013 09:25 PM, Scott Calvin wrote:

...

Athena is crashing every time I try to "Reconstruct Data" in the PCA dialog. It doesn't seem to matter what data I use, or what space I'm working in (mu or chi), or how many components I use in the reconstruction.

...

I'm using Demeter 0.9.17 with Ifeffit 1.2.11d, under Windows 8 running in Parallels 8 running under OS 10.8.4 on a MacBook Pro.

...

The log file for the latest crash is below. If you'd like a project file too, let me know, but it happens, e.g., with the gold cyanobacteria project file, and with every data set I've tried.

Hi Scott,

I do not see this behavior on any of my computers (which include native WinXp and Win7 machines and linux) and I am having trouble guessing from the log file you attached a possible cause of the problem.

Possibility #1: you are using a configuration that I cannot test. I have never used Parallels and none of my computers are running Windows 8. I'd be surprised if either is the cause of the problem, though. It seems surprising that Athena would get that deeply into its operations before running into a platform-related problem. That said, I'd be interested to know if you see this on a native Win7 or WinXP machine.

Possibility #2: you have figured out some combination of mouse clicks and button presses that I have never tested. The problem you are seeing is that the group considered current (i.e. the one that should be the target of the reconstruction) has become unset. Typically, when you click on an item in the group list, Athena will assign that as the current data set. It will then run a simple test to see if it was among the data groups included in the PCA decomposition. If so, the "reconstruct" button is enabled. If not, the TT button is enabled.

I cannot find a compbination of events that leaves me in a state that unsets this, resulting in your stack trace. So without further instruction, I don't know how to go about solving the problem.

B

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

-- John Hayes The Grosvenor Group Dept. of Chemistry University of Saskatchewan “Somebody told me it was frightening how much topsoil we are losing each year, but I told that story around the campfire and nobody got scared.”

John, I now understand the nature of the problem you observed. In fact, I suspect I can now explain the problem that Scott reported a couple months ago. One of the things about doing PCA on a subset of the data in a project file is that I have to somehow keep track of which set of data groups have been included in the PCA so that I can know which data groups should be reconstructed and which could be objects of a target transform. I did a very stupid thing that resulted in easily loosing track of which data groups were in the PCA. I think I will have this fixed in the next release. Until then, I can suggest a work-around. In the case of the project file you sent me, simply remove all the reference spectra from the project, then rerun the PCA. In general, if you reorganize the group list (http://bruceravel.github.io/demeter/aug/ui/glist.html#reorganizingthegroupli...) so that the groups in the PCA are all at the top of the group list, you will not trigger my silly bug. As always, thanks for reporting this problem. As I often say, whatever quality the software has is thanks to the generous participation of its users. B On 07/25/2013 10:56 AM, Ravel, Bruce wrote:

Hi John,

Firstly, I should apologize for a very slow response. Unfortunately, you posted your problem with PCA right before I went on travel for a while. Back home, I have finally worked my way to this item on my to do list.

I am seeing the same problem you report on my own computer. Don't know what the problem is yet, but I'll work on it. I'll keep you posted and hopefully it will get resolved for the next release.

Cheers, B

Bruce Ravel ------------------------------------ bravel@bnl.gov

National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973

Homepage: http://xafs.org/BruceRavel Software: https://github.com/bruceravel

________________________________________ From: ifeffit-bounces@millenia.cars.aps.anl.gov [ifeffit-bounces@millenia.cars.aps.anl.gov] on behalf of John Hayes [ginger.jesus@usask.ca] Sent: Wednesday, July 03, 2013 6:43 PM To: XAFS Analysis using Ifeffit Subject: Re: [Ifeffit] DAthena crashes when reconstructing data from PCA

Hi all,

I seem to be having some trouble with Athena crashing during the reconstruction step for some of my data when doing PCA analysis, and was wondering if anyone might have some insight. I I have attached a project file with the spectra in question (I realize they probably aren't normalized as well as they could be) and the dathena.log file, which is similar to Dr. Calvin's reported error log, aside from the PATH information. However, this is not an issue with Parallels, as I am using a 64 bit version of Windows 7.

Here, the issue I have is when reconstructing any of the files starting with TMF08-03 (in the attached "Combined - tailings.prj" project file). I am able to reconstruct all of the TMF08-01 files, and I have also successfully tried PCA/reconstruction on the files from the Au cyanobacteria example, so I do not believe it is an issue with my computer. To the extent that it may be an error with the TMF08-03 files, I have tried to re-import/re-work all the data, but this does not solve the problem, so it does not seems like the data itself is to blame. Perhaps the file names are presenting issues for Athena? (i.e. it might be confused if it is a standard or a sample?)

Any help on this issue is greatly appreciated. I apologize in advance for any omitted information or poor error reporting on my part, and thank you for your patience and support with this issue.

Cheers!

John

-- John Hayes

The Grosvenor Group Dept. of Chemistry University of Saskatchewan

“Somebody told me it was frightening how much topsoil we are losing each year, but I told that story around the campfire and nobody got scared.”

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

-- 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

On 07/29/2013 01:35 PM, John Hayes wrote:

I have noticed that when I changed the order of the data groups the calculated eigenvalues change when the first group on the list changes. The changes are very small and not significant, but based on my (very limited) knowledge of PCA, it would seem that the order of the data would not matter.

John, I agree that order should not matter in terms of the useful output of a PCA. However, it seems plausible that the order of input vectors could result in small numerical differences in the eigenvectors. It even seems plausible that input vectors in a different order could result in eigenvectors that are rotated by some non-negligible angle. In truth, I have neither thought about this nor investigated it before. I'll put it on my to do list, but I suspect that the bottom line is one of the two things I suggested. BTW, the specific problem you reported will certainly be fixed in the next release. I got that working this morning. 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