Update test data and tests with rotated DonutStamps.

Quick fix to get WEP pipeline running on macOS.
Update docs.

doc/versionHistory.rst

@@ -6,6 +6,14 @@
 Version History
 ##################
 
+.. _lsst.ts.wep-7.1.0:
+
+-------------
+7.1.0
+-------------
+
+* Save all DonutStamps with images aligned with focal plane science sensors.
+
 .. _lsst.ts.wep-7.0.0:
 
 -------------
@@ -20,7 +28,7 @@ Version History
 6.4.12
 -------------
 
-* Update  ts_pre_commit_config with ruff.
+* Update ts_pre_commit_config with ruff.
 
 .. _lsst.ts.wep-6.4.11:
 

python/lsst/ts/wep/task/cutOutDonutsBase.py

@@ -28,7 +28,6 @@
 import lsst.afw.cameraGeom
 import lsst.pex.config as pexConfig
 import lsst.pipe.base as pipeBase
-import lsst.obs.lsst as obs_lsst
 import numpy as np
 from lsst.daf.base import PropertyList
 from lsst.fgcmcal.utilities import lookupStaticCalibrations
@@ -40,10 +39,10 @@
 from lsst.ts.wep.utils import (
     DonutTemplateType,
     createInstDictFromConfig,
-    getCamTypeFromButlerName,
     getCameraFromButlerName,
+    getCamTypeFromButlerName,
 )
-from scipy.ndimage import binary_dilation, shift, rotate
+from scipy.ndimage import binary_dilation, rotate, shift
 from scipy.signal import correlate
 
 
@@ -156,6 +155,7 @@ class CutOutDonutsBaseTaskConfig(
         optional=True,
     )
 
+
 class CutOutDonutsBaseTask(pipeBase.PipelineTask):
     """
     Base class for CutOutDonuts tasks.
@@ -533,17 +533,12 @@ def cutOutStamps(self, exposure, donutCatalog, defocalType, cameraName):
             camera = getCameraFromButlerName(cameraName)
             detectorInfo = camera.get(detectorName)
 
-            # Rotate any sensors that are not lined up with the focal plane.
-            # Mostly just for the corner wavefront sensors. The negative sign
-            # creates the correct rotation based upon closed loop tests
-            # with R04 and R40 corner sensors.
+            # Rotate sensors to line up with the science sensors in the
+            # focal plane.
             eulerZ = -detectorInfo.getOrientation().getYaw().asDegrees()
-            donutStamp.stamp_im.image.array = rotate(donutStamp.stamp_im.image.array, eulerZ)
-
-            # NOTE: TS_WEP expects these images to be transposed
-            # TODO: Look into this
-            # if self.transposeImages:
-            #     finalStamp.image.array = finalStamp.image.array.T
+            donutStamp.stamp_im.image.array = rotate(
+                donutStamp.stamp_im.image.array, eulerZ
+            )
 
             donutStamp.stamp_im.setMask(donutStamp.mask_comp)
 

python/lsst/ts/wep/task/donutStamp.py

@@ -27,12 +27,16 @@
 import lsst.afw.image as afwImage
 import lsst.afw.table as afwTable
 import lsst.geom
-import lsst.obs.lsst as obs_lsst
 import numpy as np
 from lsst.afw.cameraGeom import FIELD_ANGLE, PIXELS
 from lsst.meas.algorithms.stamps import AbstractStamp
 from lsst.ts.wep.cwfs.compensableImage import CompensableImage
-from lsst.ts.wep.utils import DefocalType, FilterType, getFilterTypeFromBandLabel, getCameraFromButlerName
+from lsst.ts.wep.utils import (
+    DefocalType,
+    FilterType,
+    getCameraFromButlerName,
+    getFilterTypeFromBandLabel,
+)
 
 
 @dataclass

tests/task/test_calcZernikesTaskCwfs.py

@@ -23,7 +23,6 @@
 
 import lsst.utils.tests
 import numpy as np
-from scipy.ndimage import rotate
 from lsst.daf import butler as dafButler
 from lsst.ts.wep.task.calcZernikesTask import CalcZernikesTask, CalcZernikesTaskConfig
 from lsst.ts.wep.task.combineZernikesMeanTask import CombineZernikesMeanTask
@@ -150,38 +149,34 @@ def testEstimateCornerZernikes(self):
         donutStampsExtra = DonutStamps.readFits(
             os.path.join(donutStampDir, "R04_SW0_donutStamps.fits")
         )
-        for stamp in donutStampsExtra:
-            stamp.stamp_im.image.array = rotate(stamp.stamp_im.image.array, 90).T
         donutStampsIntra = DonutStamps.readFits(
             os.path.join(donutStampDir, "R04_SW1_donutStamps.fits")
         )
-        for stamp in donutStampsIntra:
-            stamp.stamp_im.image.array = rotate(stamp.stamp_im.image.array, -90).T
         zernCoeffAllR04 = self.task.estimateZernikes(donutStampsExtra, donutStampsIntra)
         zernCoeffAvgR04 = self.task.combineZernikes.run(
             zernCoeffAllR04
         ).combinedZernikes
         trueZernCoeffR04 = np.array(
             [
-                -0.71201408,
-                1.12248525,
-                0.77794367,
-                -0.04085477,
-                -0.05272933,
-                0.16054277,
-                0.081405,
-                -0.04382461,
-                -0.04830676,
-                -0.06218882,
-                0.10246469,
-                0.0197683,
-                0.007953,
-                0.00668697,
-                -0.03570788,
-                -0.03020376,
-                0.0039522,
-                0.04793133,
-                -0.00804605,
+                -0.35353452,
+                -0.07365128,
+                0.62222451,
+                -0.06206281,
+                -0.09065757,
+                0.21722746,
+                -0.20491936,
+                0.00849322,
+                -0.01150489,
+                0.02599147,
+                -0.00150702,
+                -0.14100845,
+                0.02294787,
+                0.02284791,
+                0.02116483,
+                -0.02537743,
+                0.01866772,
+                0.01653037,
+                -0.00552862,
             ]
         )
         # Make sure the total rms error is less than 0.35 microns off
@@ -194,38 +189,34 @@ def testEstimateCornerZernikes(self):
         donutStampsExtra = DonutStamps.readFits(
             os.path.join(donutStampDir, "R40_SW0_donutStamps.fits")
         )
-        for stamp in donutStampsExtra:
-            stamp.stamp_im.image.array = rotate(stamp.stamp_im.image.array, -90).T
         donutStampsIntra = DonutStamps.readFits(
             os.path.join(donutStampDir, "R40_SW1_donutStamps.fits")
         )
-        for stamp in donutStampsIntra:
-            stamp.stamp_im.image.array = rotate(stamp.stamp_im.image.array, 90).T
         zernCoeffAllR40 = self.task.estimateZernikes(donutStampsExtra, donutStampsIntra)
         zernCoeffAvgR40 = self.task.combineZernikes.run(
             zernCoeffAllR40
         ).combinedZernikes
         trueZernCoeffR40 = np.array(
             [
-                -0.6535694,
-                1.00838499,
-                0.55968811,
-                -0.08899825,
-                0.00173607,
-                0.04133107,
-                -0.10913093,
-                -0.04363778,
-                -0.03149601,
-                -0.04941225,
-                0.09980538,
-                0.03704486,
-                -0.00210766,
-                0.01737253,
-                0.01727539,
-                0.01278011,
-                0.01212878,
-                0.03876888,
-                -0.00559142,
+                -3.83610201e-01,
+                -2.06528254e-01,
+                5.42893431e-01,
+                7.74255848e-02,
+                3.40529812e-02,
+                5.45565149e-02,
+                8.65849308e-02,
+                1.75029212e-02,
+                -1.40149246e-04,
+                4.11223127e-02,
+                -2.42644902e-03,
+                -1.52392233e-01,
+                -1.24547354e-02,
+                -2.33075716e-02,
+                7.35477674e-04,
+                1.93518814e-02,
+                -3.65768735e-03,
+                4.12718699e-02,
+                -6.93386734e-03,
             ]
         )
         # Make sure the total rms error is less than 0.35 microns off

tests/task/test_cutOutDonutsBase.py

@@ -265,7 +265,9 @@ def testCutOutStamps(self):
             lsst.geom.Extent2I(160),
         )
         expCutOut = exposure[stampBBox].image.array
-        np.testing.assert_array_equal(donutStamps[0].stamp_im.image.array, expCutOut)
+        np.testing.assert_array_almost_equal(
+            donutStamps[0].stamp_im.image.array, expCutOut
+        )
 
     def testCutOutStampsBlended(self):
         exposure = self.butler.get(