bugrelax - Bugs: bug #21079, RelaxError failure of the...

 
 
Show feedback again

bug #21079: RelaxError failure of the dauvergne_protocol auto-analysis when the "local tm" global model is selected.

Submitted by:  Edward d Auvergne <bugman>
Submitted on:  Thu 29 Aug 2013 08:10:57 AM UTC  
 
Category: relax's source codeSpecific analysis category: Model-free analysis
Priority: 3 - LowSeverity: 2 - Minor
Status: FixedAssigned to: Edward d Auvergne <bugman>
Originator Name: Open/Closed: Closed
Release: Repository: trunkOperating System: All systems

Add a New Comment (Rich MarkupRich Markup):
   

You are not logged in

Please log in, so followups can be emailed to you.

 

(Jump to the original submission Jump to the original submission)

Mon 02 Sep 2013 01:02:57 PM UTC, SVN revision 20747:

Merged revisions 20743-20744 via svnmerge from
svn+ssh://bugman@svn.gna.org/svn/relax/trunk

........
r20743 | bugman | 2013-09-02 14:45:59 +0200 (Mon, 02 Sep 2013) | 5 lines

Modified the create_mc_data() method to partly fix bug #21079 (https://gna.org/bugs/?21079).

Some spins with local tm models remain selected despite not containing any data. These are handled
explicitly. Instead of a RelaxNoModelError being raised, the method returns None to indicate that
something went wrong.
........
r20744 | bugman | 2013-09-02 14:46:00 +0200 (Mon, 02 Sep 2013) | 5 lines

Final fix for bug #21079 (https://gna.org/bugs/?21079).

This is the failure of the dauvergne_protocol auto-analysis when the "local tm" global model is
selected. The Monte Carlo create_data() method not skips data from the base_data_loop() if the
create_mc_data() method returns None.
........

(Browse SVN revision 20747)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Mon 02 Sep 2013 12:51:36 PM UTC, SVN revision 20745:

Merged revisions 20741 via svnmerge from
svn+ssh://bugman@svn.gna.org/svn/relax/trunk

........
r20741 | bugman | 2013-09-02 14:12:14 +0200 (Mon, 02 Sep 2013) | 5 lines

Created the Mf.test_bug_21079_local_tm_global_selection system test.

This is to catch bug #21079 (https://gna.org/bugs/?21079).
........

(Browse SVN revision 20745)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Mon 02 Sep 2013 12:46:00 PM UTC, SVN revision 20744:

Final fix for bug #21079 (https://gna.org/bugs/?21079).

This is the failure of the dauvergne_protocol auto-analysis when the "local tm" global model is
selected. The Monte Carlo create_data() method not skips data from the base_data_loop() if the
create_mc_data() method returns None.

(Browse SVN revision 20744)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Mon 02 Sep 2013 12:45:59 PM UTC, SVN revision 20743:

Modified the create_mc_data() method to partly fix bug #21079 (https://gna.org/bugs/?21079).

Some spins with local tm models remain selected despite not containing any data. These are handled
explicitly. Instead of a RelaxNoModelError being raised, the method returns None to indicate that
something went wrong.

(Browse SVN revision 20743)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Mon 02 Sep 2013 12:12:15 PM UTC, SVN revision 20741:

Created the Mf.test_bug_21079_local_tm_global_selection system test.

This is to catch bug #21079 (https://gna.org/bugs/?21079).

(Browse SVN revision 20741)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Mon 02 Sep 2013 09:57:50 AM UTC, comment #3:

This bug triggers at exactly the same position as bug #14941 (https://gna.org/bugs/?14941), though it is probably not related.

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Mon 02 Sep 2013 09:49:57 AM UTC, comment #2:

The attached script can be used to replicate the bug. First unpack the 'results_data.rar' and 'local_tm.rar' files, then place this script in the same directory. Running the script mimics the last steps of the model-free auto-analysis (the dauvergne_protocol), allowing the bug to be triggered.

(file #18826)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Thu 29 Aug 2013 09:06:35 AM UTC, comment #1:

The error is not a simple case of just the "local tm" global model being selected. The attached state file and script demonstrate that normal selection of this global model works fine. The problem lies elsewhere.

(file #18788, file #18789)

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.
Thu 29 Aug 2013 08:10:57 AM UTC, original submission:

This bug has been mentioned twice in the relax-users mailing list:

http://thread.gmane.org/gmane.science.nmr.relax.user/1295
http://thread.gmane.org/gmane.science.nmr.relax.user/1518

However no bug report has been submitted. Hence the problem could never be fixed due to a severe lack of information. This is the bug report for both instances.

Running relax with the debug flag 'relax -d' would result in a much better indication to what the problem really is. And attaching a relax state file saved just before the error message - or the results files for each global diffusion model loaded prior to the error - would be extremely helpful.

The last messages from Nicolas Doucet are:

relax> model_selection(method='AIC', modsel_pipe='final - mf (Sun Sep 16 17:10:51 2012)', bundle='mf (Sun Sep 16 17:10:51 2012)', pipes=['local_tm - mf (Sun Sep 16 17:10:51 2012)', 'sphere - mf (Sun Sep 16 17:10:51 2012)', 'prolate - mf (Sun Sep 16 17:10:51 2012)', 'oblate - mf (Sun Sep 16 17:10:51 2012)', 'ellipsoid - mf (Sun Sep 16 17:10:51 2012)'])
AIC model selection.

Global model - all diffusion tensor parameters and spin specific model-free parameters.
Data pipe Num_params_(k) Num_data_sets_(n) Chi2 Criterion
local_tm - mf (Sun Sep 16 17:10:51 2012) 240 696 375395.80880 375875.80880
sphere - mf (Sun Sep 16 17:10:51 2012) 160 696 380497.43227 380817.43227
prolate - mf (Sun Sep 16 17:10:51 2012) 164 696 380806.43290 381134.43290
oblate - mf (Sun Sep 16 17:10:51 2012) 168 696 380546.69280 380882.69280
ellipsoid - mf (Sun Sep 16 17:10:51 2012) 171 696 380696.67633 381038.67633
The model from the data pipe 'local_tm - mf (Sun Sep 16 17:10:51 2012)' has been selected.

relax> monte_carlo.setup(number=500)

relax> monte_carlo.create_data(method='back_calc')
RelaxError: The specific model has not been selected or set up.

Edward d Auvergne <bugman>
Project AdministratorIn charge of this item.

 

(Note: upload size limit is set to 1024 kB, after insertion of the required escape characters.)

Attach File(s):
   
   
Comment:
   

Attached Files
file #18826:  replicate_bug_21079.py added by bugman (592B - text/x-python - Script to use together with the 'results_data.rar' and 'local_tm.rar' files to replicate the bug.)
file #18809:  results_data.rar added by None (326kB - application/octet-stream - results file that produce the error)
file #18808:  local_tm.rar added by None (967kB - application/octet-stream - Here is the results files that produce the error.)
file #18788:  pre_error_state.bz2 added by bugman (7kB - application/x-bzip2 - Script and relax saved state in an attempt to replicate the bug.)
file #18789:  error_replicate_script.py added by bugman (435B - text/x-python - Script and relax saved state in an attempt to replicate the bug.)

 

Depends on the following items: None found

Items that depend on this one: None found

 

Carbon-Copy List
  • -unavailable- added by bugman (Submitted the item)
  • -unavailable- added by bugman
  • -unavailable- added by bugman
  •  

    Do you think this task is very important?
    If so, you can click here to add your encouragement to it.
    This task has 0 encouragements so far.

    Only logged-in users can vote.

     

    Please enter the title of George Orwell's famous dystopian book (it's a date):

     

     

    Follow 11 latest changes.

    Date Changed By Updated Field Previous Value => Replaced By
    Tue 04 Feb 2014 05:28:33 PM UTCbugmanSpecific analysis categoryNone=>Model-free analysis
    Mon 02 Sep 2013 12:50:24 PM UTCbugmanStatusConfirmed=>Fixed
      Open/ClosedOpen=>Closed
    Mon 02 Sep 2013 09:49:57 AM UTCbugmanAttached File-=>Added replicate_bug_21079.py, #18826
      StatusIn Progress=>Confirmed
    Sat 31 Aug 2013 05:54:22 PM UTCNoneAttached File-=>Added results_data.rar, #18809
    Sat 31 Aug 2013 05:47:51 PM UTCNoneAttached File-=>Added local_tm.rar, #18808
    Thu 29 Aug 2013 09:06:35 AM UTCbugmanAttached File-=>Added pre_error_state.bz2, #18788
      Attached File-=>Added error_replicate_script.py, #18789
    Thu 29 Aug 2013 08:10:57 AM UTCbugmanCarbon-Copy-=>Added -unavailable-
      Carbon-Copy-=>Added -unavailable-
    Show feedback again

    Back to the top


    Powered by Savane 3.1-cleanup