Ticket #5021 (closed: fixed)

Opened 9 years ago

Last modified 5 years ago

ConvertToDiffractionMDWorkspace: convert one MDEvent per histogram bin

Reported by: Janik Zikovsky Owned by: Janik Zikovsky
Priority: critical Milestone: Release 2.1
Component: Mantid Keywords:
Cc: Blocked By:
Blocking: Tester: Vickie Lynch

Description

  • Take a MatrixWorkspace input.
  • Instead of using individual events, use the binning. Add an option for this.
  • Convert each spectrum on the fly to an EventList as needed.
  • Convert to MD with the existing code.

Change History

comment:1 Changed 9 years ago by Janik Zikovsky

Refs #5021: method in EventList to create it from a histogram

for code reuse from ConvertToEventWorkspace

Changeset: e3edcdf079cf1767c8be91ec7d405e7bdee292fc

comment:2 Changed 9 years ago by Janik Zikovsky

Refs #5021: ConvertToDiffractionMDWorkspace can make 1 MDEvent per bin

Changeset: 0ceb47c39519e468847fce7a0715668c4d2f8115

comment:3 Changed 9 years ago by Janik Zikovsky

  • Status changed from new to accepted

comment:4 Changed 9 years ago by Janik Zikovsky

Refs #5021: a test fails on Ubuntu, sometimes!

disabling it. Also wiki docs

Changeset: e5decb4665cf61fa7c74c51cc7d106c86d0c603b

comment:5 Changed 9 years ago by Russell Taylor

Re #5021. Fix system tests.

Changeset: 7d670fdcaa952dfb36703c1d1bea9dc30a6c260d

comment:6 Changed 9 years ago by Janik Zikovsky

Refs #5021 fix thread-safety issue in ConvertToDiffractionMDWorkspace

Changeset: 7c75c5174cac83580926615c4ed5c5b238930d58

comment:7 Changed 9 years ago by Janik Zikovsky

Refs #5021: check units sooner rather than later

Changeset: 3ddb5d1f5e1cefd254babd6bfe3bfb50dca943b3

comment:8 Changed 9 years ago by Janik Zikovsky

Refs #5021 performance improvements in ConvertToDiffractionMDWorkspace

also fix for TestViewer

Changeset: 459e9f1ee24d7b4c0918a81540920178bfe19a4e

comment:9 Changed 9 years ago by Janik Zikovsky

Refs #5021: logging instead of stdout

Changeset: b249e36635872c66a19ff89fa6554b2c1a85e7df

comment:10 Changed 9 years ago by Janik Zikovsky

Refs #5021: save on memory by converting to events on the fly

in the case of Workspace2D input. Also more tests

Changeset: fe8316b6988b08548a47e3dfbe4ba69cade9350f

comment:11 Changed 8 years ago by Janik Zikovsky

Refs #5021 ConvertToMD sped up 109 sec->62 sec

for a large-ish event file, by splitting more frequently. Also refs #4800 parallel binnining in SliceViewer dynamic rebinning

Changeset: 26b89c042537d7e115ed9e9935b18b9658392aae

comment:12 Changed 8 years ago by Janik Zikovsky

Refs #5021: FIFO scheduler takes time down to 52 sec

for TOPAZ_5625 (down from 109 seconds before split change). FIFO is a lot more efficient than the largest cost scheduler for lots of small tasks

Changeset: 7dc1271a177ac1b6fac9716b9b0386b271788503

comment:13 Changed 8 years ago by Janik Zikovsky

Refs #5021: attempt to split MDBoxes on addEvent()

segfaults sometimes...

Changeset: c8a09efa2173103a839e2e316308166416a1bddd

comment:14 Changed 8 years ago by Janik Zikovsky

Revert "Refs #5021: attempt to split MDBoxes on addEvent()"

This reverts commit 5196f531ee42e0ab7cff184d80e45fa1addda52d.

Changeset: d52ce1de13bebf7879378d582022563407a3fcc0

comment:15 Changed 8 years ago by Janik Zikovsky

refs #5021: MinRecursionDepth option for ConvertToDiffractionMD

Changeset: f0877ae5b60f8ae3379645e33dd85815b4e517d4

comment:16 Changed 8 years ago by Janik Zikovsky

refs #5021: MinRecursionDepth option for ConvertToDiffractionMD

Changeset: f0877ae5b60f8ae3379645e33dd85815b4e517d4

comment:17 Changed 8 years ago by Janik Zikovsky

Refs #5021: empirical optimization of box splitting time

to get reasonably fast results both on small and large workspaces

Changeset: adff7ad569192c1d9f9159fe22431574525f4525

comment:18 Changed 8 years ago by Janik Zikovsky

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

Note to testers: this is the OneEventPerBin option. Should work on EventW orkspaces and Workspace2Ds. Details in the wiki page http://www.mantidproject.org/ConvertToDiffractionMDWorkspace#OneEventPerBin

comment:19 Changed 8 years ago by Janik Zikovsky

Refs #5021: empirical optimization of box splitting time

to get reasonably fast results both on small and large workspaces

Changeset: adff7ad569192c1d9f9159fe22431574525f4525

comment:20 Changed 8 years ago by Vickie Lynch

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

comment:21 Changed 8 years ago by Vickie Lynch

  • Status changed from verifying to closed

OneEventPerBin option works on EventWorkspace that is binned and Workspace2D

comment:22 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue 5867

Note: See TracTickets for help on using tickets.