Add data quality-related subgroup definitions to HDF5 data format#2965
Conversation
|
juanpala1997
left a comment
juanpala1997
left a comment
There was a problem hiding this comment.
The proposed changes look okay for me.
I noticed that the structure of both _append_quality_telescope_groups and _append_quality_subarray_groups in hdf5merger.py is very similar, differing only in the monitoring group being traversed. I was wondering if this could be further reduced using only one method with an additional input parameter. But I understand that the current structure probably makes the readability of the code easier, so it seems fine to leave it like this. I approve it.
| if self.telescope_events: | ||
| self._append_monitoring_telescope_groups(other) | ||
| self._append_pixel_statistics(other) | ||
| self._append_quality_telescope_groups(other) |
There was a problem hiding this comment.
I'm not sure we want to tie telescope quality monitoring to telescope events.
Data quality by telescope might still be interesting even if dropping by-telescope DL1 or DL2 reconstructed quantities, right? And data volume is much smaller.
Should we gate this behind a dedicated option? Or just always append if present?
There was a problem hiding this comment.
I was thinking similarly and decided for the moment to follow the suite of the monitoring data. Adding another option would make an already option-heavy config even larger. On the other hand, from the practical point of view, I think at the merging stage we will always have at least camera calibration events, but also muons. illuminator, etc.
maxnoe
changed the title
4 participants
No description provided.