[DAMD-104] Update docs on converting pfsArm fluxes to counts Created: 19/Dec/20 Updated: 29/Jan/22 Resolved: 08/Jul/21 |
|
| Status: | Won't Fix |
| Project: | Data Model |
| Component/s: | None |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Story | Priority: | Normal |
| Reporter: | rhl | Assignee: | price |
| Resolution: | Won't Fix | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||
| Story Points: | 1 | ||||||||
| Sprint: | 2DDRP-2021 A 2 | ||||||||
| Description |
|
datamodel.txt states:
I believe that this now means pfsFiberProfile (the word appears in the file twice), and this should be updated. In addition, please provide a few more details. I think that it means "multiply by fiberProfiles[fiberId].norm". |
| Comments |
| Comment by rhl [ 20/Dec/20 ] |
|
Having just used this information, I think we should also put a summary statistic into the pfsArm file metadata, probably the medians of the fiberProfiles[i].norm arrays. It's too expensive to process the fiberProfile every time you need to get rough counts for a spectrum (and you can't do it as a numpy array operation, you have to iterate over fibres); yes, you can cache this in user code but that involves an extra global. While updating the datamodel, we should fix the pfsArm; if should read HDU #1 FIBERID Fiber identifier [32-bit INT] NFIBER |
| Comment by price [ 08/Jul/21 ] |
|
I believe this is unnecessary following (And I fixed the FIBERID HDU of pfsArm there.) |