You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
ORNL scientists would like to use a single workflow for both powder and single crystal data analysis. Currently, the single crystal workflow uses SHIVER and produces an MDHistoWorkspace as an .nxs file. They would like to use MSlice to produce such a file for the powder data workflow too. This would require both rebinned 1D and 2D data to be saveable as MDHisto currently the "NeXus" save options produce Workspace2D types.
Secondary to this is that some scientists want to process the data rebinned in MSlice further within Mantid. However, MSlice intentionally hides its workspace from the Mantid Analysis Data Service (ADS) because it generates a lot of internal workspace and this was thought to confuse users. In order to allow further processing, an option should be enabled to "unhide" and/or convert some of the rebinned 1D and 2D data as MDHistoWorkspaces.
Detailed description
One option to accomplish this is to add an extra option to the "Save" (disk icon) option for 1D and 2D plots (code here).
Alternatively, a button could be added to the "Slice" and "Cut" tabs to save to disk or to "export" to the ADS.
One complication is with the internal type of the workspaces. There are two workflows in MSlice depending on how the original data was reduced:
If the data was reduced using a "1-to-1" or similar grouping, it is converted to an MDEventWorkspace and BinMD is used for rebinning. In this case both the 1D and 2D rebinned data are MDHistoWorkspaces
If the data used a "powder grouping" then it is used directly as a Workspace2D and the SofQWNormalisedPolygon algorithm is used for rebinning. This creates Workspace2D output workspaces. In the 1D case (which is exposed to users), the Workspace2D is converted to an MDHistoWorkspace using CreateMDHistoWorkspace (code here). However, the 2D case is not exposed to users so will need a conversion to MDHistoWorkspace. Note that this conversion loses some instrument information which may make it unsuitable for further processing - specific use cases will need to be investigated.
The text was updated successfully, but these errors were encountered:
Summary
ORNL scientists would like to use a single workflow for both powder and single crystal data analysis. Currently, the single crystal workflow uses SHIVER and produces an
MDHistoWorkspace
as an.nxs
file. They would like to use MSlice to produce such a file for the powder data workflow too. This would require both rebinned 1D and 2D data to be saveable asMDHisto
currently the "NeXus" save options produceWorkspace2D
types.Secondary to this is that some scientists want to process the data rebinned in MSlice further within Mantid. However, MSlice intentionally hides its workspace from the Mantid Analysis Data Service (ADS) because it generates a lot of internal workspace and this was thought to confuse users. In order to allow further processing, an option should be enabled to "unhide" and/or convert some of the rebinned 1D and 2D data as
MDHistoWorkspace
s.Detailed description
One option to accomplish this is to add an extra option to the "Save" (disk icon) option for 1D and 2D plots (code here).
Alternatively, a button could be added to the "Slice" and "Cut" tabs to save to disk or to "export" to the ADS.
One complication is with the internal type of the workspaces. There are two workflows in MSlice depending on how the original data was reduced:
MDEventWorkspace
andBinMD
is used for rebinning. In this case both the 1D and 2D rebinned data areMDHistoWorkspace
sWorkspace2D
and theSofQWNormalisedPolygon
algorithm is used for rebinning. This createsWorkspace2D
output workspaces. In the 1D case (which is exposed to users), theWorkspace2D
is converted to anMDHistoWorkspace
usingCreateMDHistoWorkspace
(code here). However, the 2D case is not exposed to users so will need a conversion toMDHistoWorkspace
. Note that this conversion loses some instrument information which may make it unsuitable for further processing - specific use cases will need to be investigated.The text was updated successfully, but these errors were encountered: