Coverage for python/lsst/fgcmcal/fgcmBuildStarsBase.py : 19%
Hot-keys on this page
r m x p toggle line displays
j k next/prev highlighted chunk
0 (zero) top of page
1 (one) first highlighted chunk
1# This file is part of fgcmcal.
2#
3# Developed for the LSST Data Management System.
4# This product includes software developed by the LSST Project
5# (https://www.lsst.org).
6# See the COPYRIGHT file at the top-level directory of this distribution
7# for details of code ownership.
8#
9# This program is free software: you can redistribute it and/or modify
10# it under the terms of the GNU General Public License as published by
11# the Free Software Foundation, either version 3 of the License, or
12# (at your option) any later version.
13#
14# This program is distributed in the hope that it will be useful,
15# but WITHOUT ANY WARRANTY; without even the implied warranty of
16# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17# GNU General Public License for more details.
18#
19# You should have received a copy of the GNU General Public License
20# along with this program. If not, see <https://www.gnu.org/licenses/>.
21"""Base class for BuildStars using src tables or sourceTable_visit tables.
22"""
24import os
25import sys
26import traceback
27import abc
29import numpy as np
31import lsst.pex.config as pexConfig
32import lsst.pipe.base as pipeBase
33import lsst.afw.table as afwTable
34import lsst.geom as geom
35from lsst.daf.base import PropertyList
36from lsst.daf.base.dateTime import DateTime
37from lsst.meas.algorithms.sourceSelector import sourceSelectorRegistry
39from .utilities import computeApertureRadiusFromDataRef
40from .fgcmLoadReferenceCatalog import FgcmLoadReferenceCatalogTask
42import fgcm
44REFSTARS_FORMAT_VERSION = 1
46__all__ = ['FgcmBuildStarsConfigBase', 'FgcmBuildStarsRunner', 'FgcmBuildStarsBaseTask']
49class FgcmBuildStarsConfigBase(pexConfig.Config):
50 """Base config for FgcmBuildStars tasks"""
52 instFluxField = pexConfig.Field(
53 doc=("Faull name of the source instFlux field to use, including 'instFlux'. "
54 "The associated flag will be implicitly included in badFlags"),
55 dtype=str,
56 default='slot_CalibFlux_instFlux',
57 )
58 minPerBand = pexConfig.Field(
59 doc="Minimum observations per band",
60 dtype=int,
61 default=2,
62 )
63 matchRadius = pexConfig.Field(
64 doc="Match radius (arcseconds)",
65 dtype=float,
66 default=1.0,
67 )
68 isolationRadius = pexConfig.Field(
69 doc="Isolation radius (arcseconds)",
70 dtype=float,
71 default=2.0,
72 )
73 densityCutNside = pexConfig.Field(
74 doc="Density cut healpix nside",
75 dtype=int,
76 default=128,
77 )
78 densityCutMaxPerPixel = pexConfig.Field(
79 doc="Density cut number of stars per pixel",
80 dtype=int,
81 default=1000,
82 )
83 matchNside = pexConfig.Field(
84 doc="Healpix Nside for matching",
85 dtype=int,
86 default=4096,
87 )
88 coarseNside = pexConfig.Field(
89 doc="Healpix coarse Nside for partitioning matches",
90 dtype=int,
91 default=8,
92 )
93 filterMap = pexConfig.DictField(
94 doc="Mapping from 'filterName' to band.",
95 keytype=str,
96 itemtype=str,
97 default={},
98 )
99 requiredBands = pexConfig.ListField(
100 doc="Bands required for each star",
101 dtype=str,
102 default=(),
103 )
104 primaryBands = pexConfig.ListField(
105 doc=("Bands for 'primary' star matches. "
106 "A star must be observed in one of these bands to be considered "
107 "as a calibration star."),
108 dtype=str,
109 default=None
110 )
111 visitDataRefName = pexConfig.Field(
112 doc="dataRef name for the 'visit' field, usually 'visit'.",
113 dtype=str,
114 default="visit"
115 )
116 ccdDataRefName = pexConfig.Field(
117 doc="dataRef name for the 'ccd' field, usually 'ccd' or 'detector'.",
118 dtype=str,
119 default="ccd"
120 )
121 doApplyWcsJacobian = pexConfig.Field(
122 doc="Apply the jacobian of the WCS to the star observations prior to fit?",
123 dtype=bool,
124 default=True
125 )
126 psfCandidateName = pexConfig.Field(
127 doc="Name of field with psf candidate flag for propagation",
128 dtype=str,
129 default="calib_psf_candidate"
130 )
131 doSubtractLocalBackground = pexConfig.Field(
132 doc=("Subtract the local background before performing calibration? "
133 "This is only supported for circular aperture calibration fluxes."),
134 dtype=bool,
135 default=False
136 )
137 localBackgroundFluxField = pexConfig.Field(
138 doc="Full name of the local background instFlux field to use.",
139 dtype=str,
140 default='base_LocalBackground_instFlux'
141 )
142 sourceSelector = sourceSelectorRegistry.makeField(
143 doc="How to select sources",
144 default="science"
145 )
146 apertureInnerInstFluxField = pexConfig.Field(
147 doc=("Full name of instFlux field that contains inner aperture "
148 "flux for aperture correction proxy"),
149 dtype=str,
150 default='base_CircularApertureFlux_12_0_instFlux'
151 )
152 apertureOuterInstFluxField = pexConfig.Field(
153 doc=("Full name of instFlux field that contains outer aperture "
154 "flux for aperture correction proxy"),
155 dtype=str,
156 default='base_CircularApertureFlux_17_0_instFlux'
157 )
158 doReferenceMatches = pexConfig.Field(
159 doc="Match reference catalog as additional constraint on calibration",
160 dtype=bool,
161 default=True,
162 )
163 fgcmLoadReferenceCatalog = pexConfig.ConfigurableField(
164 target=FgcmLoadReferenceCatalogTask,
165 doc="FGCM reference object loader",
166 )
167 nVisitsPerCheckpoint = pexConfig.Field(
168 doc="Number of visits read between checkpoints",
169 dtype=int,
170 default=500,
171 )
173 def setDefaults(self):
174 sourceSelector = self.sourceSelector["science"]
175 sourceSelector.setDefaults()
177 sourceSelector.doFlags = True
178 sourceSelector.doUnresolved = True
179 sourceSelector.doSignalToNoise = True
180 sourceSelector.doIsolated = True
182 sourceSelector.signalToNoise.minimum = 10.0
183 sourceSelector.signalToNoise.maximum = 1000.0
185 # FGCM operates on unresolved sources, and this setting is
186 # appropriate for the current base_ClassificationExtendedness
187 sourceSelector.unresolved.maximum = 0.5
190class FgcmBuildStarsRunner(pipeBase.ButlerInitializedTaskRunner):
191 """Subclass of TaskRunner for FgcmBuildStars tasks
193 fgcmBuildStarsTask.run() and fgcmBuildStarsTableTask.run() take a number of
194 arguments, one of which is the butler (for persistence and mapper data),
195 and a list of dataRefs extracted from the command line. Note that FGCM
196 runs on a large set of dataRefs, and not on single dataRef/tract/patch.
197 This class transforms the process arguments generated by the ArgumentParser
198 into the arguments expected by FgcmBuildStarsTask.run(). This runner does
199 not use any parallelization.
200 """
201 @staticmethod
202 def getTargetList(parsedCmd):
203 """
204 Return a list with one element: a tuple with the butler and
205 list of dataRefs
206 """
207 # we want to combine the butler with any (or no!) dataRefs
208 return [(parsedCmd.butler, parsedCmd.id.refList)]
210 def __call__(self, args):
211 """
212 Parameters
213 ----------
214 args: `tuple` with (butler, dataRefList)
216 Returns
217 -------
218 exitStatus: `list` with `lsst.pipe.base.Struct`
219 exitStatus (0: success; 1: failure)
220 """
221 butler, dataRefList = args
223 task = self.TaskClass(config=self.config, log=self.log)
225 exitStatus = 0
226 if self.doRaise:
227 task.runDataRef(butler, dataRefList)
228 else:
229 try:
230 task.runDataRef(butler, dataRefList)
231 except Exception as e:
232 exitStatus = 1
233 task.log.fatal("Failed: %s" % e)
234 if not isinstance(e, pipeBase.TaskError):
235 traceback.print_exc(file=sys.stderr)
237 task.writeMetadata(butler)
239 # The task does not return any results:
240 return [pipeBase.Struct(exitStatus=exitStatus)]
242 def run(self, parsedCmd):
243 """
244 Run the task, with no multiprocessing
246 Parameters
247 ----------
248 parsedCmd: `lsst.pipe.base.ArgumentParser` parsed command line
249 """
251 resultList = []
253 if self.precall(parsedCmd):
254 targetList = self.getTargetList(parsedCmd)
255 resultList = self(targetList[0])
257 return resultList
260class FgcmBuildStarsBaseTask(pipeBase.CmdLineTask, abc.ABC):
261 """
262 Base task to build stars for FGCM global calibration
264 Parameters
265 ----------
266 butler : `lsst.daf.persistence.Butler`
267 """
268 def __init__(self, butler=None, **kwargs):
269 pipeBase.CmdLineTask.__init__(self, **kwargs)
270 self.makeSubtask("sourceSelector")
271 # Only log warning and fatal errors from the sourceSelector
272 self.sourceSelector.log.setLevel(self.sourceSelector.log.WARN)
274 # no saving of metadata for now
275 def _getMetadataName(self):
276 return None
278 @pipeBase.timeMethod
279 def runDataRef(self, butler, dataRefs):
280 """
281 Cross-match and make star list for FGCM Input
283 Parameters
284 ----------
285 butler: `lsst.daf.persistence.Butler`
286 dataRefs: `list` of `lsst.daf.persistence.ButlerDataRef`
287 Source data references for the input visits.
289 Raises
290 ------
291 RuntimeErrror: Raised if `config.doReferenceMatches` is set and
292 an fgcmLookUpTable is not available, or if computeFluxApertureRadius()
293 fails if the calibFlux is not a CircularAperture flux.
294 """
295 datasetType = dataRefs[0].butlerSubset.datasetType
296 self.log.info("Running with %d %s dataRefs", len(dataRefs), datasetType)
298 if self.config.doReferenceMatches:
299 # Ensure that we have a LUT
300 if not butler.datasetExists('fgcmLookUpTable'):
301 raise RuntimeError("Must have fgcmLookUpTable if using config.doReferenceMatches")
302 # Compute aperture radius if necessary. This is useful to do now before
303 # any heavy lifting has happened (fail early).
304 calibFluxApertureRadius = None
305 if self.config.doSubtractLocalBackground:
306 try:
307 calibFluxApertureRadius = computeApertureRadiusFromDataRef(dataRefs[0],
308 self.config.instFluxField)
309 except RuntimeError as e:
310 raise RuntimeError("Could not determine aperture radius from %s. "
311 "Cannot use doSubtractLocalBackground." %
312 (self.config.instFluxField)) from e
314 groupedDataRefs = self.findAndGroupDataRefs(butler, dataRefs)
316 camera = butler.get('camera')
318 # Make the visit catalog if necessary
319 # First check if the visit catalog is in the _current_ path
320 # We cannot use Gen2 datasetExists() because that checks all parent
321 # directories as well, which would make recovering from faults
322 # and fgcmcal reruns impossible.
323 visitCatDataRef = butler.dataRef('fgcmVisitCatalog')
324 filename = visitCatDataRef.get('fgcmVisitCatalog_filename')[0]
325 if os.path.exists(filename):
326 # This file exists and we should continue processing
327 inVisitCat = visitCatDataRef.get()
328 if len(inVisitCat) != len(groupedDataRefs):
329 raise RuntimeError("Existing visitCatalog found, but has an inconsistent "
330 "number of visits. Cannot continue.")
331 else:
332 inVisitCat = None
334 visitCat = self.fgcmMakeVisitCatalog(camera, groupedDataRefs,
335 visitCatDataRef=visitCatDataRef,
336 inVisitCat=inVisitCat)
338 # Persist the visitCat as a checkpoint file.
339 visitCatDataRef.put(visitCat)
341 starObsDataRef = butler.dataRef('fgcmStarObservations')
342 filename = starObsDataRef.get('fgcmStarObservations_filename')[0]
343 if os.path.exists(filename):
344 inStarObsCat = starObsDataRef.get()
345 else:
346 inStarObsCat = None
348 rad = calibFluxApertureRadius
349 fgcmStarObservationCat = self.fgcmMakeAllStarObservations(groupedDataRefs,
350 visitCat,
351 calibFluxApertureRadius=rad,
352 starObsDataRef=starObsDataRef,
353 visitCatDataRef=visitCatDataRef,
354 inStarObsCat=inStarObsCat)
355 visitCatDataRef.put(visitCat)
356 starObsDataRef.put(fgcmStarObservationCat)
358 # Always do the matching.
359 fgcmStarIdCat, fgcmStarIndicesCat, fgcmRefCat = self.fgcmMatchStars(butler,
360 visitCat,
361 fgcmStarObservationCat)
363 # Persist catalogs via the butler
364 butler.put(fgcmStarIdCat, 'fgcmStarIds')
365 butler.put(fgcmStarIndicesCat, 'fgcmStarIndices')
366 if fgcmRefCat is not None:
367 butler.put(fgcmRefCat, 'fgcmReferenceStars')
369 @abc.abstractmethod
370 def findAndGroupDataRefs(self, butler, dataRefs):
371 """
372 Find and group dataRefs (by visit).
374 Parameters
375 ----------
376 butler: `lsst.daf.persistence.Butler`
377 dataRefs: `list` of `lsst.daf.persistence.ButlerDataRef`
378 Data references for the input visits.
380 Returns
381 -------
382 groupedDataRefs: `dict` [`int`, `list`]
383 Dictionary with visit keys, and `list`s of `lsst.daf.persistence.ButlerDataRef`
384 """
385 raise NotImplementedError("findAndGroupDataRefs not implemented.")
387 @abc.abstractmethod
388 def fgcmMakeAllStarObservations(self, groupedDataRefs, visitCat,
389 calibFluxApertureRadius=None,
390 visitCatDataRef=None,
391 starObsDataRef=None,
392 inStarObsCat=None):
393 """
394 Compile all good star observations from visits in visitCat. Checkpoint files
395 will be stored if both visitCatDataRef and starObsDataRef are not None.
397 Parameters
398 ----------
399 groupedDataRefs: `dict` of `list`s
400 Lists of `lsst.daf.persistence.ButlerDataRef`, grouped by visit.
401 visitCat: `afw.table.BaseCatalog`
402 Catalog with visit data for FGCM
403 calibFluxApertureRadius: `float`, optional
404 Aperture radius for calibration flux.
405 visitCatDataRef: `lsst.daf.persistence.ButlerDataRef`, optional
406 Dataref to write visitCat for checkpoints
407 starObsDataRef: `lsst.daf.persistence.ButlerDataRef`, optional
408 Dataref to write the star observation catalog for checkpoints.
409 inStarObsCat: `afw.table.BaseCatalog`
410 Input observation catalog. If this is incomplete, observations
411 will be appended from when it was cut off.
413 Returns
414 -------
415 fgcmStarObservations: `afw.table.BaseCatalog`
416 Full catalog of good observations.
418 Raises
419 ------
420 RuntimeError: Raised if doSubtractLocalBackground is True and
421 calibFluxApertureRadius is not set.
422 """
423 raise NotImplementedError("fgcmMakeAllStarObservations not implemented.")
425 def fgcmMakeVisitCatalog(self, camera, groupedDataRefs,
426 visitCatDataRef=None, inVisitCat=None):
427 """
428 Make a visit catalog with all the keys from each visit
430 Parameters
431 ----------
432 camera: `lsst.afw.cameraGeom.Camera`
433 Camera from the butler
434 groupedDataRefs: `dict`
435 Dictionary with visit keys, and `list`s of
436 `lsst.daf.persistence.ButlerDataRef`
437 visitCatDataRef: `lsst.daf.persistence.ButlerDataRef`, optional
438 Dataref to write visitCat for checkpoints
439 inVisitCat: `afw.table.BaseCatalog`
440 Input (possibly incomplete) visit catalog
442 Returns
443 -------
444 visitCat: `afw.table.BaseCatalog`
445 """
447 self.log.info("Assembling visitCatalog from %d %ss" %
448 (len(groupedDataRefs), self.config.visitDataRefName))
450 nCcd = len(camera)
452 if inVisitCat is None:
453 schema = self._makeFgcmVisitSchema(nCcd)
455 visitCat = afwTable.BaseCatalog(schema)
456 visitCat.reserve(len(groupedDataRefs))
458 for i, visit in enumerate(sorted(groupedDataRefs)):
459 rec = visitCat.addNew()
460 rec['visit'] = visit
461 rec['used'] = 0
462 rec['sources_read'] = 0
463 else:
464 visitCat = inVisitCat
466 # No matter what, fill the catalog. This will check if it was
467 # already read.
468 self._fillVisitCatalog(visitCat, groupedDataRefs,
469 visitCatDataRef=visitCatDataRef)
471 return visitCat
473 def _fillVisitCatalog(self, visitCat, groupedDataRefs,
474 visitCatDataRef=None):
475 """
476 Fill the visit catalog with visit metadata
478 Parameters
479 ----------
480 visitCat: `afw.table.BaseCatalog`
481 Catalog with schema from _makeFgcmVisitSchema()
482 groupedDataRefs: `dict`
483 Dictionary with visit keys, and `list`s of
484 `lsst.daf.persistence.ButlerDataRef
485 visitCatDataRef: `lsst.daf.persistence.ButlerDataRef`, optional
486 Dataref to write visitCat for checkpoints
487 """
489 bbox = geom.BoxI(geom.PointI(0, 0), geom.PointI(1, 1))
491 for i, visit in enumerate(sorted(groupedDataRefs)):
492 # We don't use the bypasses since we need the psf info which does
493 # not have a bypass
494 # TODO: When DM-15500 is implemented in the Gen3 Butler, this
495 # can be fixed
497 # Do not read those that have already been read
498 if visitCat['used'][i]:
499 continue
501 if (i % self.config.nVisitsPerCheckpoint) == 0:
502 self.log.info("Retrieving metadata for %s %d (%d/%d)" %
503 (self.config.visitDataRefName, visit, i, len(groupedDataRefs)))
504 # Save checkpoint if desired
505 if visitCatDataRef is not None:
506 visitCatDataRef.put(visitCat)
508 # Note that the reference ccd is first in the list (if available).
510 # The first dataRef in the group will be the reference ccd (if available)
511 dataRef = groupedDataRefs[visit][0]
513 exp = dataRef.get(datasetType='calexp_sub', bbox=bbox,
514 flags=afwTable.SOURCE_IO_NO_FOOTPRINTS)
516 visitInfo = exp.getInfo().getVisitInfo()
517 f = exp.getFilter()
518 psf = exp.getPsf()
520 rec = visitCat[i]
521 rec['visit'] = visit
522 rec['filtername'] = f.getName()
523 # TODO DM-26991: when gen2 is removed, gen3 workflow will make it
524 # much easier to get the wcs's necessary to recompute the pointing
525 # ra/dec at the center of the camera.
526 radec = visitInfo.getBoresightRaDec()
527 rec['telra'] = radec.getRa().asDegrees()
528 rec['teldec'] = radec.getDec().asDegrees()
529 rec['telha'] = visitInfo.getBoresightHourAngle().asDegrees()
530 rec['telrot'] = visitInfo.getBoresightRotAngle().asDegrees()
531 rec['mjd'] = visitInfo.getDate().get(system=DateTime.MJD)
532 rec['exptime'] = visitInfo.getExposureTime()
533 # convert from Pa to millibar
534 # Note that I don't know if this unit will need to be per-camera config
535 rec['pmb'] = visitInfo.getWeather().getAirPressure() / 100
536 # Flag to signify if this is a "deep" field. Not currently used
537 rec['deepFlag'] = 0
538 # Relative flat scaling (1.0 means no relative scaling)
539 rec['scaling'][:] = 1.0
540 # Median delta aperture, to be measured from stars
541 rec['deltaAper'] = 0.0
543 rec['psfSigma'] = psf.computeShape().getDeterminantRadius()
545 if dataRef.datasetExists(datasetType='calexpBackground'):
546 # Get background for reference CCD
547 # This approximation is good enough for now
548 bgStats = (bg[0].getStatsImage().getImage().array
549 for bg in dataRef.get(datasetType='calexpBackground'))
550 rec['skyBackground'] = sum(np.median(bg[np.isfinite(bg)]) for bg in bgStats)
551 else:
552 self.log.warn('Sky background not found for visit %d / ccd %d' %
553 (visit, dataRef.dataId[self.config.ccdDataRefName]))
554 rec['skyBackground'] = -1.0
556 rec['used'] = 1
558 def _makeSourceMapper(self, sourceSchema):
559 """
560 Make a schema mapper for fgcm sources
562 Parameters
563 ----------
564 sourceSchema: `afwTable.Schema`
565 Default source schema from the butler
567 Returns
568 -------
569 sourceMapper: `afwTable.schemaMapper`
570 Mapper to the FGCM source schema
571 """
573 # create a mapper to the preferred output
574 sourceMapper = afwTable.SchemaMapper(sourceSchema)
576 # map to ra/dec
577 sourceMapper.addMapping(sourceSchema['coord_ra'].asKey(), 'ra')
578 sourceMapper.addMapping(sourceSchema['coord_dec'].asKey(), 'dec')
579 sourceMapper.addMapping(sourceSchema['slot_Centroid_x'].asKey(), 'x')
580 sourceMapper.addMapping(sourceSchema['slot_Centroid_y'].asKey(), 'y')
581 # Add the mapping if the field exists in the input catalog.
582 # If the field does not exist, simply add it (set to False).
583 # This field is not required for calibration, but is useful
584 # to collate if available.
585 try:
586 sourceMapper.addMapping(sourceSchema[self.config.psfCandidateName].asKey(),
587 'psf_candidate')
588 except LookupError:
589 sourceMapper.editOutputSchema().addField(
590 "psf_candidate", type='Flag',
591 doc=("Flag set if the source was a candidate for PSF determination, "
592 "as determined by the star selector."))
594 # and add the fields we want
595 sourceMapper.editOutputSchema().addField(
596 "visit", type=np.int32, doc="Visit number")
597 sourceMapper.editOutputSchema().addField(
598 "ccd", type=np.int32, doc="CCD number")
599 sourceMapper.editOutputSchema().addField(
600 "instMag", type=np.float32, doc="Instrumental magnitude")
601 sourceMapper.editOutputSchema().addField(
602 "instMagErr", type=np.float32, doc="Instrumental magnitude error")
603 sourceMapper.editOutputSchema().addField(
604 "jacobian", type=np.float32, doc="Relative pixel scale from wcs jacobian")
605 sourceMapper.editOutputSchema().addField(
606 "deltaMagBkg", type=np.float32, doc="Change in magnitude due to local background offset")
608 return sourceMapper
610 def fgcmMatchStars(self, butler, visitCat, obsCat):
611 """
612 Use FGCM code to match observations into unique stars.
614 Parameters
615 ----------
616 butler: `lsst.daf.persistence.Butler`
617 visitCat: `afw.table.BaseCatalog`
618 Catalog with visit data for fgcm
619 obsCat: `afw.table.BaseCatalog`
620 Full catalog of star observations for fgcm
622 Returns
623 -------
624 fgcmStarIdCat: `afw.table.BaseCatalog`
625 Catalog of unique star identifiers and index keys
626 fgcmStarIndicesCat: `afwTable.BaseCatalog`
627 Catalog of unique star indices
628 fgcmRefCat: `afw.table.BaseCatalog`
629 Catalog of matched reference stars.
630 Will be None if `config.doReferenceMatches` is False.
631 """
633 if self.config.doReferenceMatches:
634 # Make a subtask for reference loading
635 self.makeSubtask("fgcmLoadReferenceCatalog", butler=butler)
637 # get filter names into a numpy array...
638 # This is the type that is expected by the fgcm code
639 visitFilterNames = np.zeros(len(visitCat), dtype='a10')
640 for i in range(len(visitCat)):
641 visitFilterNames[i] = visitCat[i]['filtername']
643 # match to put filterNames with observations
644 visitIndex = np.searchsorted(visitCat['visit'],
645 obsCat['visit'])
647 obsFilterNames = visitFilterNames[visitIndex]
649 if self.config.doReferenceMatches:
650 # Get the reference filter names, using the LUT
651 lutCat = butler.get('fgcmLookUpTable')
653 stdFilterDict = {filterName: stdFilter for (filterName, stdFilter) in
654 zip(lutCat[0]['filterNames'].split(','),
655 lutCat[0]['stdFilterNames'].split(','))}
656 stdLambdaDict = {stdFilter: stdLambda for (stdFilter, stdLambda) in
657 zip(lutCat[0]['stdFilterNames'].split(','),
658 lutCat[0]['lambdaStdFilter'])}
660 del lutCat
662 referenceFilterNames = self._getReferenceFilterNames(visitCat,
663 stdFilterDict,
664 stdLambdaDict)
665 self.log.info("Using the following reference filters: %s" %
666 (', '.join(referenceFilterNames)))
668 else:
669 # This should be an empty list
670 referenceFilterNames = []
672 # make the fgcm starConfig dict
674 starConfig = {'logger': self.log,
675 'filterToBand': self.config.filterMap,
676 'requiredBands': self.config.requiredBands,
677 'minPerBand': self.config.minPerBand,
678 'matchRadius': self.config.matchRadius,
679 'isolationRadius': self.config.isolationRadius,
680 'matchNSide': self.config.matchNside,
681 'coarseNSide': self.config.coarseNside,
682 'densNSide': self.config.densityCutNside,
683 'densMaxPerPixel': self.config.densityCutMaxPerPixel,
684 'primaryBands': self.config.primaryBands,
685 'referenceFilterNames': referenceFilterNames}
687 # initialize the FgcmMakeStars object
688 fgcmMakeStars = fgcm.FgcmMakeStars(starConfig)
690 # make the primary stars
691 # note that the ra/dec native Angle format is radians
692 # We determine the conversion from the native units (typically
693 # radians) to degrees for the first observation. This allows us
694 # to treate ra/dec as numpy arrays rather than Angles, which would
695 # be approximately 600x slower.
696 conv = obsCat[0]['ra'].asDegrees() / float(obsCat[0]['ra'])
697 fgcmMakeStars.makePrimaryStars(obsCat['ra'] * conv,
698 obsCat['dec'] * conv,
699 filterNameArray=obsFilterNames,
700 bandSelected=False)
702 # and match all the stars
703 fgcmMakeStars.makeMatchedStars(obsCat['ra'] * conv,
704 obsCat['dec'] * conv,
705 obsFilterNames)
707 if self.config.doReferenceMatches:
708 fgcmMakeStars.makeReferenceMatches(self.fgcmLoadReferenceCatalog)
710 # now persist
712 objSchema = self._makeFgcmObjSchema()
714 # make catalog and records
715 fgcmStarIdCat = afwTable.BaseCatalog(objSchema)
716 fgcmStarIdCat.reserve(fgcmMakeStars.objIndexCat.size)
717 for i in range(fgcmMakeStars.objIndexCat.size):
718 fgcmStarIdCat.addNew()
720 # fill the catalog
721 fgcmStarIdCat['fgcm_id'][:] = fgcmMakeStars.objIndexCat['fgcm_id']
722 fgcmStarIdCat['ra'][:] = fgcmMakeStars.objIndexCat['ra']
723 fgcmStarIdCat['dec'][:] = fgcmMakeStars.objIndexCat['dec']
724 fgcmStarIdCat['obsArrIndex'][:] = fgcmMakeStars.objIndexCat['obsarrindex']
725 fgcmStarIdCat['nObs'][:] = fgcmMakeStars.objIndexCat['nobs']
727 obsSchema = self._makeFgcmObsSchema()
729 fgcmStarIndicesCat = afwTable.BaseCatalog(obsSchema)
730 fgcmStarIndicesCat.reserve(fgcmMakeStars.obsIndexCat.size)
731 for i in range(fgcmMakeStars.obsIndexCat.size):
732 fgcmStarIndicesCat.addNew()
734 fgcmStarIndicesCat['obsIndex'][:] = fgcmMakeStars.obsIndexCat['obsindex']
736 if self.config.doReferenceMatches:
737 refSchema = self._makeFgcmRefSchema(len(referenceFilterNames))
739 fgcmRefCat = afwTable.BaseCatalog(refSchema)
740 fgcmRefCat.reserve(fgcmMakeStars.referenceCat.size)
742 for i in range(fgcmMakeStars.referenceCat.size):
743 fgcmRefCat.addNew()
745 fgcmRefCat['fgcm_id'][:] = fgcmMakeStars.referenceCat['fgcm_id']
746 fgcmRefCat['refMag'][:, :] = fgcmMakeStars.referenceCat['refMag']
747 fgcmRefCat['refMagErr'][:, :] = fgcmMakeStars.referenceCat['refMagErr']
749 md = PropertyList()
750 md.set("REFSTARS_FORMAT_VERSION", REFSTARS_FORMAT_VERSION)
751 md.set("FILTERNAMES", referenceFilterNames)
752 fgcmRefCat.setMetadata(md)
754 else:
755 fgcmRefCat = None
757 return fgcmStarIdCat, fgcmStarIndicesCat, fgcmRefCat
759 def _makeFgcmVisitSchema(self, nCcd):
760 """
761 Make a schema for an fgcmVisitCatalog
763 Parameters
764 ----------
765 nCcd: `int`
766 Number of CCDs in the camera
768 Returns
769 -------
770 schema: `afwTable.Schema`
771 """
773 schema = afwTable.Schema()
774 schema.addField('visit', type=np.int32, doc="Visit number")
775 # Note that the FGCM code currently handles filternames up to 2 characters long
776 schema.addField('filtername', type=str, size=10, doc="Filter name")
777 schema.addField('telra', type=np.float64, doc="Pointing RA (deg)")
778 schema.addField('teldec', type=np.float64, doc="Pointing Dec (deg)")
779 schema.addField('telha', type=np.float64, doc="Pointing Hour Angle (deg)")
780 schema.addField('telrot', type=np.float64, doc="Camera rotation (deg)")
781 schema.addField('mjd', type=np.float64, doc="MJD of visit")
782 schema.addField('exptime', type=np.float32, doc="Exposure time")
783 schema.addField('pmb', type=np.float32, doc="Pressure (millibar)")
784 schema.addField('psfSigma', type=np.float32, doc="PSF sigma (reference CCD)")
785 schema.addField('deltaAper', type=np.float32, doc="Delta-aperture")
786 schema.addField('skyBackground', type=np.float32, doc="Sky background (ADU) (reference CCD)")
787 # the following field is not used yet
788 schema.addField('deepFlag', type=np.int32, doc="Deep observation")
789 schema.addField('scaling', type='ArrayD', doc="Scaling applied due to flat adjustment",
790 size=nCcd)
791 schema.addField('used', type=np.int32, doc="This visit has been ingested.")
792 schema.addField('sources_read', type='Flag', doc="This visit had sources read.")
794 return schema
796 def _makeFgcmObjSchema(self):
797 """
798 Make a schema for the objIndexCat from fgcmMakeStars
800 Returns
801 -------
802 schema: `afwTable.Schema`
803 """
805 objSchema = afwTable.Schema()
806 objSchema.addField('fgcm_id', type=np.int32, doc='FGCM Unique ID')
807 # Will investigate making these angles...
808 objSchema.addField('ra', type=np.float64, doc='Mean object RA (deg)')
809 objSchema.addField('dec', type=np.float64, doc='Mean object Dec (deg)')
810 objSchema.addField('obsArrIndex', type=np.int32,
811 doc='Index in obsIndexTable for first observation')
812 objSchema.addField('nObs', type=np.int32, doc='Total number of observations')
814 return objSchema
816 def _makeFgcmObsSchema(self):
817 """
818 Make a schema for the obsIndexCat from fgcmMakeStars
820 Returns
821 -------
822 schema: `afwTable.Schema`
823 """
825 obsSchema = afwTable.Schema()
826 obsSchema.addField('obsIndex', type=np.int32, doc='Index in observation table')
828 return obsSchema
830 def _makeFgcmRefSchema(self, nReferenceBands):
831 """
832 Make a schema for the referenceCat from fgcmMakeStars
834 Parameters
835 ----------
836 nReferenceBands: `int`
837 Number of reference bands
839 Returns
840 -------
841 schema: `afwTable.Schema`
842 """
844 refSchema = afwTable.Schema()
845 refSchema.addField('fgcm_id', type=np.int32, doc='FGCM Unique ID')
846 refSchema.addField('refMag', type='ArrayF', doc='Reference magnitude array (AB)',
847 size=nReferenceBands)
848 refSchema.addField('refMagErr', type='ArrayF', doc='Reference magnitude error array',
849 size=nReferenceBands)
851 return refSchema
853 def _getReferenceFilterNames(self, visitCat, stdFilterDict, stdLambdaDict):
854 """
855 Get the reference filter names, in wavelength order, from the visitCat and
856 information from the look-up-table.
858 Parameters
859 ----------
860 visitCat: `afw.table.BaseCatalog`
861 Catalog with visit data for FGCM
862 stdFilterDict: `dict`
863 Mapping of filterName to stdFilterName from LUT
864 stdLambdaDict: `dict`
865 Mapping of stdFilterName to stdLambda from LUT
867 Returns
868 -------
869 referenceFilterNames: `list`
870 Wavelength-ordered list of reference filter names
871 """
873 # Find the unique list of filter names in visitCat
874 filterNames = np.unique(visitCat.asAstropy()['filtername'])
876 # Find the unique list of "standard" filters
877 stdFilterNames = {stdFilterDict[filterName] for filterName in filterNames}
879 # And sort these by wavelength
880 referenceFilterNames = sorted(stdFilterNames, key=stdLambdaDict.get)
882 return referenceFilterNames