Hello, I am observing some behavior in (D)Athena that I do not understand, and I was hoping someone could explain it to me. I have attached a project file to illustrate my question. The first two groups in the file have been normalized such that mu(E) and norm(E) are identical. I then merged these groups in mu(E) and norm(E), also in the file. Since mu(E) and norm(E) are the same for the groups, I would expect the data merged in mu(E) and norm(E) to also be identical, but it is not. While it is clear that merging in mu(E) is averaging the groups, it seems merging in norm(E) is not. So what is merging in norm(E) doing? Thank you, George
I agree that you have uncovered an edge case where my merging algorithm does something unexpected. Figuring this out and fixing it is nowhere near the top of my list of priorities. I think I would be happier flagging an error when the user sets a zero-length range for the pre- or post-edge lines. Athena is not intended to be a generic data processing tool. Its algorithms are fairly well tested and expected to work when you have data that nominally resembles XAS data -- that is, some kind of step-like function, possibly with wiggles above the step. It has an algorithm for normalization that makes those same assumptions. We all eventually go to the beamline and measure something that needs interpretation, but which is not a step-like function with wiggles. The solution to that problem is not to expect Athena to miraculously do what you want. The solution is to go find the find the right tool for the job. Or, if it doesn't exist, write it yourself. B On Thursday, December 06, 2012 10:59:55 AM George Sterbinsky wrote:
Hello,
I am observing some behavior in (D)Athena that I do not understand, and I was hoping someone could explain it to me. I have attached a project file to illustrate my question. The first two groups in the file have been normalized such that mu(E) and norm(E) are identical. I then merged these groups in mu(E) and norm(E), also in the file. Since mu(E) and norm(E) are the same for the groups, I would expect the data merged in mu(E) and norm(E) to also be identical, but it is not. While it is clear that merging in mu(E) is averaging the groups, it seems merging in norm(E) is not. So what is merging in norm(E) doing?
Thank you, George
-- 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
The unexpected result I demonstrated persists for real step-like XAS data.
The data in the project I attached are actually just monitor counts, and
are only being used here to illustrate the unexpected behavior. It is not
actually data that I am trying to analyze.
George
On Fri, Dec 14, 2012 at 12:12 PM, Bruce Ravel
I agree that you have uncovered an edge case where my merging algorithm does something unexpected. Figuring this out and fixing it is nowhere near the top of my list of priorities. I think I would be happier flagging an error when the user sets a zero-length range for the pre- or post-edge lines.
Athena is not intended to be a generic data processing tool. Its algorithms are fairly well tested and expected to work when you have data that nominally resembles XAS data -- that is, some kind of step-like function, possibly with wiggles above the step. It has an algorithm for normalization that makes those same assumptions.
We all eventually go to the beamline and measure something that needs interpretation, but which is not a step-like function with wiggles. The solution to that problem is not to expect Athena to miraculously do what you want. The solution is to go find the find the right tool for the job. Or, if it doesn't exist, write it yourself.
B
On Thursday, December 06, 2012 10:59:55 AM George Sterbinsky wrote:
Hello,
I am observing some behavior in (D)Athena that I do not understand, and I was hoping someone could explain it to me. I have attached a project file to illustrate my question. The first two groups in the file have been normalized such that mu(E) and norm(E) are identical. I then merged these groups in mu(E) and norm(E), also in the file. Since mu(E) and norm(E) are the same for the groups, I would expect the data merged in mu(E) and norm(E) to also be identical, but it is not. While it is clear that merging in mu(E) is averaging the groups, it seems merging in norm(E) is not. So what is merging in norm(E) doing?
Thank you, George
--
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 _______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
participants (2)
-
Bruce Ravel -
George Sterbinsky