Coverage for python/lsst/ap/association/association.py : 26%

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 ap_association.
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/>.
22"""A simple implementation of source association task for ap_verify.
23"""
25__all__ = ["AssociationConfig", "AssociationTask"]
27import numpy as np
28import pandas as pd
29from scipy.spatial import cKDTree
31import lsst.geom as geom
32import lsst.pex.config as pexConfig
33import lsst.pipe.base as pipeBase
35# Enforce an error for unsafe column/array value setting in pandas.
36pd.options.mode.chained_assignment = 'raise'
39class AssociationConfig(pexConfig.Config):
40 """Config class for AssociationTask.
41 """
42 maxDistArcSeconds = pexConfig.Field(
43 dtype=float,
44 doc='Maximum distance in arcseconds to test for a DIASource to be a '
45 'match to a DIAObject.',
46 default=1.0,
47 )
50class AssociationTask(pipeBase.Task):
51 """Associate DIAOSources into existing DIAObjects.
53 This task performs the association of detected DIASources in a visit
54 with the previous DIAObjects detected over time. It also creates new
55 DIAObjects out of DIASources that cannot be associated with previously
56 detected DIAObjects.
57 """
59 ConfigClass = AssociationConfig
60 _DefaultName = "association"
62 @pipeBase.timeMethod
63 def run(self,
64 diaSources,
65 diaObjects):
66 """Associate the new DiaSources with existing DiaObjects.
68 Parameters
69 ----------
70 diaSources : `pandas.DataFrame`
71 New DIASources to be associated with existing DIAObjects.
72 diaObjects : `pandas.DataFrame`
73 Existing diaObjects from the Apdb.
75 Returns
76 -------
77 result : `lsst.pipe.base.Struct`
78 Results struct with components.
80 - ``"matchedDiaSources"`` : DiaSources that were matched. Matched
81 Sources have their diaObjectId updated and set to the id of the
82 diaObject they were matched to. (`pandas.DataFrame`)
83 - ``"unAssocDiaSources"`` : DiaSources that were not matched.
84 Unassociated sources have their diaObject set to 0 as they
85 were not associated with any existing DiaObjects.
86 (`pandas.DataFrame`)
87 - ``"nUpdatedDiaObjects"`` : Number of DiaObjects that were
88 matched to new DiaSources. (`int`)
89 - ``"nUnassociatedDiaObjects"`` : Number of DiaObjects that were
90 not matched a new DiaSource. (`int`)
91 """
92 diaSources = self.check_dia_source_radec(diaSources)
93 if len(diaObjects) == 0:
94 return pipeBase.Struct(
95 matchedDiaSources=pd.DataFrame(columns=diaSources.columns),
96 unAssocDiaSources=diaSources,
97 nUpdatedDiaObjects=0,
98 nUnassociatedDiaObjects=0)
100 matchResult = self.associate_sources(diaObjects, diaSources)
102 mask = matchResult.diaSources["diaObjectId"] != 0
104 return pipeBase.Struct(
105 matchedDiaSources=matchResult.diaSources[mask].reset_index(drop=True),
106 unAssocDiaSources=matchResult.diaSources[~mask].reset_index(drop=True),
107 nUpdatedDiaObjects=matchResult.nUpdatedDiaObjects,
108 nUnassociatedDiaObjects=matchResult.nUnassociatedDiaObjects)
110 def check_dia_source_radec(self, dia_sources):
111 """Check that all DiaSources have non-NaN values for RA/DEC.
113 If one or more DiaSources are found to have NaN values, throw a
114 warning to the log with the ids of the offending sources. Drop them
115 from the table.
117 Parameters
118 ----------
119 dia_sources : `pandas.DataFrame`
120 Input DiaSources to check for NaN values.
122 Returns
123 -------
124 trimmed_sources : `pandas.DataFrame`
125 DataFrame of DiaSources trimmed of all entries with NaN values for
126 RA/DEC.
127 """
128 nan_mask = (dia_sources.loc[:, "ra"].isnull()
129 | dia_sources.loc[:, "decl"].isnull())
130 if np.any(nan_mask):
131 nan_idxs = np.argwhere(nan_mask.to_numpy()).flatten()
132 for nan_idx in nan_idxs:
133 self.log.warning(
134 "DiaSource %i has NaN value for RA/DEC, "
135 "dropping from association." %
136 dia_sources.loc[nan_idx, "diaSourceId"])
137 dia_sources = dia_sources[~nan_mask]
138 return dia_sources
140 @pipeBase.timeMethod
141 def associate_sources(self, dia_objects, dia_sources):
142 """Associate the input DIASources with the catalog of DIAObjects.
144 DiaObject DataFrame must be indexed on ``diaObjectId``.
146 Parameters
147 ----------
148 dia_objects : `pandas.DataFrame`
149 Catalog of DIAObjects to attempt to associate the input
150 DIASources into.
151 dia_sources : `pandas.DataFrame`
152 DIASources to associate into the DIAObjectCollection.
154 Returns
155 -------
156 result : `lsst.pipe.base.Struct`
157 Results struct with components.
159 - ``"diaSources"`` : Full set of diaSources both matched and not.
160 (`pandas.DataFrame`)
161 - ``"nUpdatedDiaObjects"`` : Number of DiaObjects that were
162 associated. (`int`)
163 - ``"nUnassociatedDiaObjects"`` : Number of DiaObjects that were
164 not matched a new DiaSource. (`int`)
165 """
166 scores = self.score(
167 dia_objects, dia_sources,
168 self.config.maxDistArcSeconds * geom.arcseconds)
169 match_result = self.match(dia_objects, dia_sources, scores)
171 return match_result
173 @pipeBase.timeMethod
174 def score(self, dia_objects, dia_sources, max_dist):
175 """Compute a quality score for each dia_source/dia_object pair
176 between this catalog of DIAObjects and the input DIASource catalog.
178 ``max_dist`` sets maximum separation in arcseconds to consider a
179 dia_source a possible match to a dia_object. If the pair is
180 beyond this distance no score is computed.
182 Parameters
183 ----------
184 dia_objects : `pandas.DataFrame`
185 A contiguous catalog of DIAObjects to score against dia_sources.
186 dia_sources : `pandas.DataFrame`
187 A contiguous catalog of dia_sources to "score" based on distance
188 and (in the future) other metrics.
189 max_dist : `lsst.geom.Angle`
190 Maximum allowed distance to compute a score for a given DIAObject
191 DIASource pair.
193 Returns
194 -------
195 result : `lsst.pipe.base.Struct`
196 Results struct with components:
198 - ``"scores"``: array of floats of match quality updated DIAObjects
199 (array-like of `float`).
200 - ``"obj_idxs"``: indexes of the matched DIAObjects in the catalog.
201 (array-like of `int`)
202 - ``"obj_ids"``: array of floats of match quality updated DIAObjects
203 (array-like of `int`).
205 Default values for these arrays are
206 INF, -1, and -1 respectively for unassociated sources.
207 """
208 scores = np.full(len(dia_sources), np.inf, dtype=np.float64)
209 obj_idxs = np.full(len(dia_sources), -1, dtype=np.int64)
210 obj_ids = np.full(len(dia_sources), 0, dtype=np.int64)
212 if len(dia_objects) == 0:
213 return pipeBase.Struct(
214 scores=scores,
215 obj_idxs=obj_idxs,
216 obj_ids=obj_ids)
218 spatial_tree = self._make_spatial_tree(dia_objects)
220 max_dist_rad = max_dist.asRadians()
222 vectors = self._radec_to_xyz(dia_sources)
224 scores, obj_idxs = spatial_tree.query(
225 vectors,
226 distance_upper_bound=max_dist_rad)
227 matched_src_idxs = np.argwhere(np.isfinite(scores))
228 obj_ids[matched_src_idxs] = dia_objects.index.to_numpy()[
229 obj_idxs[matched_src_idxs]]
231 return pipeBase.Struct(
232 scores=scores,
233 obj_idxs=obj_idxs,
234 obj_ids=obj_ids)
236 def _make_spatial_tree(self, dia_objects):
237 """Create a searchable kd-tree the input dia_object positions.
239 Parameters
240 ----------
241 dia_objects : `pandas.DataFrame`
242 A catalog of DIAObjects to create the tree from.
244 Returns
245 -------
246 kd_tree : `scipy.spatical.cKDTree`
247 Searchable kd-tree created from the positions of the DIAObjects.
248 """
249 vectors = self._radec_to_xyz(dia_objects)
250 return cKDTree(vectors)
252 def _radec_to_xyz(self, catalog):
253 """Convert input ra/dec coordinates to spherical unit-vectors.
255 Parameters
256 ----------
257 catalog : `pandas.DataFrame`
258 Catalog to produce spherical unit-vector from.
260 Returns
261 -------
262 vectors : `numpy.ndarray`, (N, 3)
263 Output unit-vectors
264 """
265 ras = np.radians(catalog["ra"])
266 decs = np.radians(catalog["decl"])
267 vectors = np.empty((len(ras), 3))
269 sin_dec = np.sin(np.pi / 2 - decs)
270 vectors[:, 0] = sin_dec * np.cos(ras)
271 vectors[:, 1] = sin_dec * np.sin(ras)
272 vectors[:, 2] = np.cos(np.pi / 2 - decs)
274 return vectors
276 @pipeBase.timeMethod
277 def match(self, dia_objects, dia_sources, score_struct):
278 """Match DIAsources to DiaObjects given a score.
280 Parameters
281 ----------
282 dia_objects : `pandas.DataFrame`
283 A SourceCatalog of DIAObjects to associate to DIASources.
284 dia_sources : `pandas.DataFrame`
285 A contiguous catalog of dia_sources for which the set of scores
286 has been computed on with DIAObjectCollection.score.
287 score_struct : `lsst.pipe.base.Struct`
288 Results struct with components:
290 - ``"scores"``: array of floats of match quality
291 updated DIAObjects (array-like of `float`).
292 - ``"obj_ids"``: array of floats of match quality
293 updated DIAObjects (array-like of `int`).
294 - ``"obj_idxs"``: indexes of the matched DIAObjects in the catalog.
295 (array-like of `int`)
297 Default values for these arrays are
298 INF, -1 and -1 respectively for unassociated sources.
300 Returns
301 -------
302 result : `lsst.pipe.base.Struct`
303 Results struct with components.
305 - ``"diaSources"`` : Full set of diaSources both matched and not.
306 (`pandas.DataFrame`)
307 - ``"nUpdatedDiaObjects"`` : Number of DiaObjects that were
308 associated. (`int`)
309 - ``"nUnassociatedDiaObjects"`` : Number of DiaObjects that were
310 not matched a new DiaSource. (`int`)
311 """
312 n_previous_dia_objects = len(dia_objects)
313 used_dia_object = np.zeros(n_previous_dia_objects, dtype=bool)
314 used_dia_source = np.zeros(len(dia_sources), dtype=bool)
315 associated_dia_object_ids = np.zeros(len(dia_sources),
316 dtype=np.uint64)
317 n_updated_dia_objects = 0
319 # We sort from best match to worst to effectively perform a
320 # "handshake" match where both the DIASources and DIAObjects agree
321 # their the best match. By sorting this way, scores with NaN (those
322 # sources that have no match and will create new DIAObjects) will be
323 # placed at the end of the array.
324 score_args = score_struct.scores.argsort(axis=None)
325 for score_idx in score_args:
326 if not np.isfinite(score_struct.scores[score_idx]):
327 # Thanks to the sorting the rest of the sources will be
328 # NaN for their score. We therefore exit the loop to append
329 # sources to a existing DIAObject, leaving these for
330 # the loop creating new objects.
331 break
332 dia_obj_idx = score_struct.obj_idxs[score_idx]
333 if used_dia_object[dia_obj_idx]:
334 continue
335 used_dia_object[dia_obj_idx] = True
336 used_dia_source[score_idx] = True
337 obj_id = score_struct.obj_ids[score_idx]
338 associated_dia_object_ids[score_idx] = obj_id
339 dia_sources.loc[score_idx, "diaObjectId"] = obj_id
340 n_updated_dia_objects += 1
342 return pipeBase.Struct(
343 diaSources=dia_sources,
344 nUpdatedDiaObjects=n_updated_dia_objects,
345 nUnassociatedDiaObjects=(n_previous_dia_objects
346 - n_updated_dia_objects))