Ticket #6329 (closed: fixed)

Opened 8 years ago

Last modified 5 years ago

Modify ConvertToMD to care about masked spectra/detectors

Reported by: Alex Buts Owned by: Alex Buts
Priority: major Milestone: Release 2.4
Component: Mantid Keywords:
Cc: Blocked By:
Blocking: Tester: Roman Tolchenov

Description

Currently all ConvertToMD plugins ignore isMasked property of the detector and expect masks to be applied as NaN on the workspace data

Mantid at the moment and in the nearest future nullifies the masked spectra.

As the result, the masked spectra are transfered to MDWorkspace and accounted for as legitimate zeros.

This is wrong and should be fixed

Change History

comment:1 Changed 8 years ago by Alex Buts

  • Status changed from new to accepted

comment:2 Changed 8 years ago by Alex Buts

refs #6329 Preliminary steps to return detector's isMasked status

Changeset: bea170b0f6e8a5e58feef05cb9ef96f2e3432df5

comment:3 Changed 8 years ago by Alex Buts

refs #6329 PreprocessDetectorsToMD builds masked spectra list

Changeset: a3e9893b23ed83597a9c562882073a3880cf1284

comment:4 Changed 8 years ago by Alex Buts

refs #6329 This should fix it

ConvertToMD considers the state of the detectors (rather spectra) masks. Ideally this functionality should be removed and spectra be masked by setting NaN to the signal .

Changeset: 53eac3129fe49641c9ac52eaa9238154577efa7c

comment:5 Changed 8 years ago by Alex Buts

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

Tester should convert to MD some physical workspace and then mask half of this workspace. (or mask e.g. peaks or the part, which correspond to some recognizable part of the reciprocal space)

Then, ConvertToMD algorithm with "UpdateMasks" parameter disabled should place 0-s in appropriate places of the reciprocal space and with "UpdateMasks" enabled --completely eliminate the correspondent data from the result.

Keep this test for the future, as when Mantid will masks missing spectra through NaN-s thee results should be the same as with UpdateMasks enabled.

comment:6 Changed 8 years ago by Alex Buts

refs #6329 Preliminary steps to return detector's isMasked status

Changeset: bea170b0f6e8a5e58feef05cb9ef96f2e3432df5

comment:7 Changed 8 years ago by Alex Buts

refs #6329 PreprocessDetectorsToMD builds masked spectra list

Changeset: a3e9893b23ed83597a9c562882073a3880cf1284

comment:8 Changed 8 years ago by Alex Buts

refs #6329 This should fix it

ConvertToMD considers the state of the detectors (rather spectra) masks. Ideally this functionality should be removed and spectra be masked by setting NaN to the signal .

Changeset: 53eac3129fe49641c9ac52eaa9238154577efa7c

comment:9 Changed 8 years ago by Roman Tolchenov

  • Status changed from verify to verifying
  • Tester set to Roman Tolchenov

comment:10 Changed 8 years ago by Roman Tolchenov

  • Status changed from verifying to closed

An MD workspace created form a half-masked matrix workspace looks different from the one created without masking. The masked MD workspace looks less dense. This confirms that masking isn't ignored in ConvertToMD.

comment:11 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue ​7175

Note: See TracTickets for help on using tickets.