AL-479: Move ramp fit code from JWST to STCAL.

[kmacdonald-stsci]

Resolves AL-479

Description

Moved ramp fitting code from JWST to STCAL. All unit tests and regression tests are passing, using ramp fitting path to STCAL.

Checklist

• Tests
• Documentation
• Change log
• Milestone
• Label(s)

[jdavies-st]

Looks good. Regression tests on this branch and the one on stcal look good:

https://plwishmaster.stsci.edu:8081/job/RT/job/jdavies-dev/241/

This PR depends on spacetelescope/stcal#6 being merged and a release of stcal happening. This PR should then be updated to specify that version or greater of stcal in setup.cfg. Then it can be merged.

🚀

[codecov]

Codecov Report

Merging #6023 (d0093b0) into master (aed4cd4) will increase coverage by 0.01%.
The diff coverage is 100.00%.

❗ Current head d0093b0 differs from pull request most recent head 523ef17. Consider uploading reports for the commit 523ef17 to get more accurate results

@@            Coverage Diff             @@
##           master    #6023      +/-   ##
==========================================
+ Coverage   77.53%   77.55%   +0.01%     
==========================================
  Files         405      405              
  Lines       35174    35220      +46     
==========================================
+ Hits        27272    27314      +42     
- Misses       7902     7906       +4     

Flag	Coverage Δ		*Carryforward flag
nightly	77.53% <100.00%> (ø)		Carriedforward from aed4cd4
unit	55.35% <100.00%> (-1.31%)	⬇️

*This pull request uses carry forward flags. Click here to find out more.

Impacted Files	Coverage Δ
jwst/ramp_fitting/ramp_fit_step.py	91.79% <100.00%> (-0.26%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 221273a...523ef17. Read the comment docs.

[nden]

spacetelescope/stcal#6 was merged and I pointed this branch to stcal/main. Once tests pass we need an stcal release and can merge this one.

[nden]

lgtm

[nden]

stcal 0.2.0 was released on PyPi

[hbushouse]

So the various JWST instrument-specific hooks for special handling of a few things like MIRI first and last frames, etc. are now all in the supposedly "common" code over in stcal?

[nden]

They are because the common code works on data models.

[kmacdonald-stsci]

I am happy to remove it from STCAL and put it into JWST. There is only one specific MIRI reference in the STCAL code, so moving it should be easy.

Are there other parts of the ramp fit code that are JWST specific that should be removed from the ramp fitting code in STCAL?

[nden]

This now points to the stcal release and can be merged.

[hbushouse]

I am happy to remove it from STCAL and put it into JWST. There is only one specific MIRI reference in the STCAL code, so moving it should be easy.

Are there other parts of the ramp fit code that are JWST specific that should be removed from the ramp fitting code in STCAL?

No need to unmove any of the code. I was just curious. I guess the assumption is that Roman will use the same RampModel and ImageModel models?

[kmacdonald-stsci]

The assumption is that both pipelines will only use a JWST-like input RampModel containing expected values and kewords. The model input variable to ramp_fit is the only variable that depends is a data model. I've removed usage of data models everywhere else in the code. All returned values from the ramp fitting code in STCAL returns tuples of computed arrays the pipeline step code is then responsible for putting into their respective data models, like an ImageModel.

[dmggh]

I think that this may have already been discussed, but as this is a change in the design and code structure (and the first of these changes among several calibration steps) this should be described in the docs for ramp fitting. The restructuring to use common code should probably also be described in other, more general sections of the docs.

[nden]

@dmggh good point and agreed!