Ticket #4337 (closed: fixed)
MUON: Improve fitting in CalMuonDeadTime when bad data
Reported by: | Anders Markvardsen | Owned by: | Anders Markvardsen |
---|---|---|---|
Priority: | critical | Milestone: | Release 2.0 |
Component: | Mantid | Keywords: | |
Cc: | robert.whitley@… | Blocked By: | |
Blocking: | Tester: | Michael Reuter |
Description
From discussion today Sean/Jon/Rob/Anders
When fitting a line to rubbish data currently struggle with obtaining reasonable fit.
Possible options to improve this
- Add a new minimizer
- Play around with how the standard deviations are calculated
- bin the DataFitted WS before doing the fit
Attachments
Change History
comment:2 Changed 9 years ago by Anders Markvardsen
Small adjustment to CalMuonDeadTiem. Re #4337
Changeset: 0a96987719b16a951671ddb893b33b7e8f5e9bd8
Changed 9 years ago by Anders Markvardsen
- Attachment EMU00019023.nxs added
Steve deadtime example file
comment:4 Changed 9 years ago by Anders Markvardsen
- Status changed from accepted to verify
- Resolution set to fixed
To test e.g.
1) load added EMU file (note sometimes attached .nxs files appears to be corrupted, if yes contact me and I will email file)
2) run CalMuonDeadTime which FirstGoodData = 0.2 and LastGoodData = 8.0
3) view the TableWorkspace outputted by this algorithms
4) chance the number LastGoodData to a larger value and hopefully you should find the numbers in the TableWorkspace changes little as a consequence of this
comment:5 Changed 9 years ago by Michael Reuter
- Status changed from verify to verifying
- Tester set to Michael Reuter
Fix to CalMuonDeadTime. Re #4337
Turned out that muon data with count equal to zero have zero errors (at least an EMU dataset). Zero errors are no good when fitting.