Updated strip hit threshold logic#81
Conversation
… these events in revan
| if (g_Verbosity >= c_Warning) { | ||
| cout << m_XmlTag << ": Strip Hit below threshold, deleting SH with Energy " << Energy << " keV " << endl; | ||
| } | ||
| Event->SetStripHitBelowThreshold(true, "with Energy " + to_string(Energy)); |
There was a problem hiding this comment.
if we have multiples this gets overwritten. Do we want that?
Or do we want to keep the smallest value?
There was a problem hiding this comment.
Good catch. Maybe instead we should track how many SH we loose and all of the energies...?
I can edit that.
There was a problem hiding this comment.
Maybe:
AddStripHitBelowThreshold(Energy)
And add a bit of logic into the event to track it?
How many hits, and total loss seems reasonable.
| if (m_SlowThresholdCutMode == MSlowThresholdCutModes::e_Fixed) { //check if user input threshold is enabled (one value applied to all strips) | ||
| Threshold = m_SlowThresholdCutFixedValue; | ||
| } else if (m_SlowThresholdCutMode == MSlowThresholdCutModes::e_File) { //check if threshold file is enabled (unique value applied to each strip) | ||
| double Threshold_map = m_ThresholdMap[R]; // if file enabled, declare value from map |
There was a problem hiding this comment.
Rename to ThresholdFromFile
| if (g_Verbosity >= c_Error) { | ||
| cout << m_XmlTag << ": Error: Threshold not found for read-out element " << R << endl; | ||
| } | ||
| Threshold = 15; // set default threshold if threshold not found |
There was a problem hiding this comment.
It is better to define "magic numbers" explicitly in the beginning:
const double DefaultSlowThreshold = 15.0;
There was a problem hiding this comment.
Think about if this might actually be part of the map file ?
| MString m_StripHitBelowThresholdString; | ||
|
|
||
|
|
||
| // Flags indicating the quality of the event: quality warning, but not to be filtered out |
There was a problem hiding this comment.
Although the names are already long:
...Incomplete always results in a bad flag
... Do you want something similar for the quality flags?
Think of soemthing - or something similar
m_StripHitBelowThresholdQF;
m_StripHitBelowThresholdQualityFlag;
|
I didn't check if it runs, I just reviewed the code. |
|
@zoglauer, I've address all of your comments, if you want to take another look. I'll open up another PR that cleans up the rest of the BD and QA flags. |
Emily-Broadbent
left a comment
Emily-Broadbent
left a comment
There was a problem hiding this comment.
After @zoglauer 's updates, the logic looks good to me
3 participants
Julian reported that the slow threshold cut was dramatically decreasing the number of events in the final ARM plots (using cut of 35 keV).
I confirmed with HP52358-3 gse_20250709T113820.hdf that I recovered only 30% of Compton events in revan when the slow threshold of 35 keV was applied.
The changes implemented:
After these changes we now recover 71% of Compton events in revan with the 35 keV threshold (and 99.5% of the Compton events when using a slow threshold file with realistic thresholds ~10 keV). I confirmed the ARM isn't dramatically altered.