DM-21212: Update existing cp_pipe tasks to pipelineTasks #148
Conversation
czwa
force-pushed
the
tickets/DM-21212
branch
2 times, most recently
from
91529ef
to
158bcd1
Compare
czwa
force-pushed
the
tickets/DM-21212
branch
2 times, most recently
from
47c799f
to
782ad59
Compare
czwa
force-pushed
the
tickets/DM-21212
branch
5 times, most recently
from
20b122d
to
ad16882
Compare
czwa
force-pushed
the
tickets/DM-21212
branch
from
ad16882
to
ae5e5fc
Compare
| doCrosstalk: False | ||
| doDefect: True | ||
| doNanMasking: True | ||
| doInterpolate: True |
There was a problem hiding this comment.
My understanding is that it cannot for complexity reasons (inheritance + config setting can be ambiguous/confusing if they both set config parameters). The real fix is in producing CT/Defects/etc for obs_lsst, which would allow the simply inherited version to be used (like obs_subaru).
pipelines/latiss/cpBias.yaml
Outdated
| connections.inputExps: 'cpBiasProc' | ||
| connections.outputData: 'biasProposal' | ||
| calibrationType: 'bias' | ||
| exposureScaling: "ExposureTime" |
There was a problem hiding this comment.
You look to be scaling by the exposure time, which should be zero...?
There was a problem hiding this comment.
That's a mistake now fixed. It worked because there's a safe guard to prevent scaling by zero, but it's better to ask for the right thing, "None".
| config: | ||
| connections.inputExps: 'cpBiasProc' | ||
| connections.outputData: 'biasProposal' | ||
| calibrationType: 'bias' |
There was a problem hiding this comment.
From the master cp_pipe pipeline definition? No.
From the camera? Yes, which was one of the issues making this change necessary.
pipelines/latiss/cpBias.yaml
Outdated
| contracts: | ||
| - isr.doBias == False | ||
| - cpCombine.calibrationType == "bias" | ||
| - cpCombine.exposureScaling == "ExposureTime" |
There was a problem hiding this comment.
Contracts are not inherited, and are there to ensure command line options don't break what the pipeline is supposed to do.
czwa
force-pushed
the
tickets/DM-21212
branch
2 times, most recently
from
1096901
to
daec367
Compare
czwa
force-pushed
the
tickets/DM-21212
branch
2 times, most recently
from
a9bf4ed
to
ae12075
Compare
|
I've filed a new ticket to handle this issue, https://jira.lsstcorp.org/browse/DM-25329 , in the interest of getting this ancient ticket merged. I'm planning on updating the ci_cpp_latiss code to fully test obs_lsst and LATISS on real data. |
Add LATISS cp_pipe pipelines.