Ticket #11637 (closed: fixed)
IntegrateMDHistoWorkspace, account for number of events
Reported by: | Owen Arnold | Owned by: | Owen Arnold |
---|---|---|---|
Priority: | critical | Milestone: | Release 3.4 |
Component: | Framework | Keywords: | vates |
Cc: | Blocked By: | #11572 | |
Blocking: | Tester: | Andrei Savici |
Description
Thus far, the contribution to each output bin is based on overlap with input bins. The fraction overlap is calculated, and used to assign a weight, which the signal and error_sq of each contributing bin is multiplied through by. I have not accounted for the number of pixels. Clearly the number of events from contributing bins needs to be accounted for.
After discussion with scientists (namely Andrei) on this, it would seem that the best way to handle this is to simply multiply the n_events from each contributing bin by the weight and assign the output to the n_events array in the output workspace. MDHisto workspace already have the n_events array concept, so we merely need to sum (n_events * weight) for each contributing bin.
Change History
comment:2 Changed 5 years ago by Owen Arnold
- Status changed from inprogress to verify
- Resolution set to fixed
This is being verified as pull request #660.
comment:3 Changed 5 years ago by Andrei Savici
- Status changed from verify to verifying
- Tester set to Andrei Savici
comment:4 Changed 5 years ago by Owen Arnold
refs #11637. Fix error in docs
Changeset: 638ddc1635c49a7afd5b53eaacfd9c5ca932aef4
comment:5 Changed 5 years ago by Owen Arnold
refs #11637. Fix casting and warning
Changeset: 8245786a752be9563db7f2e9fe12a2bd1187d26b
comment:6 Changed 5 years ago by Andrei Savici
Merge pull request #660 from mantidproject/11637_use_n_events
refs #11637. n_events weights
Changeset: 854107aa1063eee2fdc10bdae253fdf2087bcc0b
comment:7 Changed 5 years ago by Andrei Savici
- Status changed from verifying to closed
Merge pull request #660 from mantidproject/11637_use_n_events
refs #11637. n_events weights
Full changeset: 854107aa1063eee2fdc10bdae253fdf2087bcc0b
refs #11637. n_events weights
Add unit test for this. Also updated documentation
Changeset: 61eb48f68deef8e600a034450282658a46cceed8