Ticket #3021 (closed: fixed)

Opened 9 years ago

Last modified 5 years ago

SNSPowderReduction: Update to use Grouping/Offsets Workspaces and speed up.

Reported by: Janik Zikovsky Owned by: Peter Peterson
Priority: major Milestone: Iteration 29
Component: Mantid Keywords:
Cc: pf9@… Blocked By:
Blocking: Tester: Vickie Lynch

Description

... Speed up Vanadium by focusing to a Workspace2d?

Change History

comment:1 Changed 9 years ago by Janik Zikovsky

  • Status changed from new to accepted

comment:2 Changed 9 years ago by Janik Zikovsky

  • Status changed from accepted to assigned
  • Owner changed from Janik Zikovsky to Peter Peterson

Hi Pete,

This turns out to be more of a refactor than I think I should do since this is your code, but these steps should help:

  • Load the Vanadium run BEFORE the sample run.
    • This looks like it might require you getting the info (_getinfo method) from the sample run beforehand; you could use LoadLogsFromSNSNexus to do this.
  • In DiffractionFocussing of Vanadium, set PreserveEvents=False.
    • You'll have to have rebinned in d-spacing before this step.
    • Then skip CompressEvents, of course.
    • It will avoid running out of memory by NOT having a large sample event workspace in memory and then loading another one.
  • You can refactor to avoid .cal files with:
    • LoadCalFile: into an OffsetsWorkspace and GroupingWorkspace
    • And then use these when Aligning/Focussing.

comment:3 Changed 9 years ago by Peter Peterson

  • Status changed from assigned to accepted

comment:4 Changed 9 years ago by Peter Peterson

(In [12345]) Five minutes just saved 30s (from 100s) on your rietveld reduction. Refs #3021.

comment:5 Changed 9 years ago by Peter Peterson

  • Status changed from accepted to verify
  • Resolution set to fixed

(In [12359]) Vanadium run is focused directly to matrixworkspace. Fixes #3021.

comment:6 Changed 9 years ago by Peter Peterson

  • Status changed from verify to reopened
  • Resolution fixed deleted

There is a bug when the default instrument is not the appropriate one for the data being processed.

comment:7 Changed 9 years ago by Peter Peterson

  • Status changed from reopened to accepted

comment:8 Changed 9 years ago by Peter Peterson

  • Status changed from accepted to verify
  • Resolution set to fixed

(In [12381]) Load calibration workspaces using a loaded data workspace. Fixes #3021.

comment:9 Changed 9 years ago by Janik Zikovsky

  • Status changed from verify to verifying
  • Tester set to Janik Zikovsky

comment:10 Changed 9 years ago by Janik Zikovsky

  • Status changed from verifying to verify
  • Tester Janik Zikovsky deleted

Don't have findnexus_lib so the algorithm fails for me.

comment:11 Changed 9 years ago by Vickie Lynch

  • Status changed from verify to verifying
  • Tester set to Vickie Lynch

comment:12 Changed 9 years ago by Vickie Lynch

  • Status changed from verifying to closed

Tested with PG3 data and works with group, offsets, and mask workspaces. Only took 85 sec.

comment:13 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue ​3868

Note: See TracTickets for help on using tickets.