Coverage for python / lsst / pipe / base / pipelineTask.py: 76%
17 statements
« prev ^ index » next coverage.py v7.13.1, created at 2026-01-18 01:04 -0800
1# This file is part of pipe_base.
2#
3# Developed for the LSST Data Management System.
4# This product includes software developed by the LSST Project
5# (http://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 software is dual licensed under the GNU General Public License and also
10# under a 3-clause BSD license. Recipients may choose which of these licenses
11# to use; please see the files gpl-3.0.txt and/or bsd_license.txt,
12# respectively. If you choose the GPL option then the following text applies
13# (but note that there is still no warranty even if you opt for BSD instead):
14#
15# This program is free software: you can redistribute it and/or modify
16# it under the terms of the GNU General Public License as published by
17# the Free Software Foundation, either version 3 of the License, or
18# (at your option) any later version.
19#
20# This program is distributed in the hope that it will be useful,
21# but WITHOUT ANY WARRANTY; without even the implied warranty of
22# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
23# GNU General Public License for more details.
24#
25# You should have received a copy of the GNU General Public License
26# along with this program. If not, see <http://www.gnu.org/licenses/>.
28"""Define `PipelineTask` class and related methods."""
30from __future__ import annotations
32__all__ = ["PipelineTask"] # Classes in this module
34from collections.abc import Callable
35from typing import TYPE_CHECKING, Any, ClassVar
37from .connections import InputQuantizedConnection, OutputQuantizedConnection
38from .task import Task
40if TYPE_CHECKING:
41 import logging
43 from lsst.utils.logging import LsstLogAdapter
45 from ._quantumContext import QuantumContext
46 from .config import PipelineTaskConfig
47 from .struct import Struct
50class PipelineTask(Task):
51 """Base class for all pipeline tasks.
53 This is an abstract base class for PipelineTasks which represents an
54 algorithm executed by framework(s) on data which comes from data butler,
55 resulting data is also stored in a data butler.
57 PipelineTask inherits from a `~lsst.pipe.base.Task` and uses the same
58 configuration mechanism based on :ref:`lsst.pex.config`. `PipelineTask`
59 classes also have a `PipelineTaskConnections` class associated with their
60 config which defines all of the IO a `PipelineTask` will need to do.
61 PipelineTask sub-class typically implements `run()` method which receives
62 Python-domain data objects and returns `lsst.pipe.base.Struct` object with
63 resulting data. `run()` method is not supposed to perform any I/O, it
64 operates entirely on in-memory objects. `runQuantum()` is the method (can
65 be re-implemented in sub-class) where all necessary I/O is performed, it
66 reads all input data from data butler into memory, calls `run()` method
67 with that data, examines returned `Struct` object and saves some or all of
68 that data back to data butler. `runQuantum()` method receives a
69 `QuantumContext` instance to facilitate I/O, a `InputQuantizedConnection`
70 instance which defines all input `lsst.daf.butler.DatasetRef`, and a
71 `OutputQuantizedConnection` instance which defines all the output
72 `lsst.daf.butler.DatasetRef` for a single invocation of PipelineTask.
74 Subclasses must be constructable with exactly the arguments taken by the
75 PipelineTask base class constructor, but may support other signatures as
76 well.
78 Attributes
79 ----------
80 canMultiprocess : bool, True by default (class attribute)
81 This class attribute is checked by execution framework, sub-classes
82 can set it to ``False`` in case task does not support multiprocessing.
84 Parameters
85 ----------
86 config : `~lsst.pex.config.Config`, optional
87 Configuration for this task (an instance of ``self.ConfigClass``,
88 which is a task-specific subclass of `PipelineTaskConfig`).
89 If not specified then it defaults to ``self.ConfigClass()``.
90 log : `logging.Logger`, optional
91 Logger instance whose name is used as a log name prefix, or ``None``
92 for no prefix.
93 initInputs : `dict`, optional
94 A dictionary of objects needed to construct this PipelineTask, with
95 keys matching the keys of the dictionary returned by
96 `getInitInputDatasetTypes` and values equivalent to what would be
97 obtained by calling `~lsst.daf.butler.Butler.get` with those
98 DatasetTypes and no data IDs. While it is optional for the base class,
99 subclasses are permitted to require this argument.
100 **kwargs : `~typing.Any`
101 Arbitrary parameters, passed to base class constructor.
102 """
104 ConfigClass: ClassVar[type[PipelineTaskConfig]]
105 canMultiprocess: ClassVar[bool] = True
107 def __init__(
108 self,
109 *,
110 config: PipelineTaskConfig | None = None,
111 log: logging.Logger | LsstLogAdapter | None = None,
112 initInputs: dict[str, Any] | None = None,
113 **kwargs: Any,
114 ):
115 super().__init__(config=config, log=log, **kwargs)
117 run: Callable[..., Struct] # The 'run' method for subclasses will have a different signature
119 def run(self, **kwargs: Any) -> Struct: # type: ignore
120 """Run task algorithm on in-memory data.
122 This method should be implemented in a subclass. This method will
123 receive keyword-only arguments whose names will be the same as names of
124 connection fields describing input dataset types. Argument values will
125 be data objects retrieved from data butler. If a dataset type is
126 configured with ``multiple`` field set to ``True`` then the argument
127 value will be a list of objects, otherwise it will be a single object.
129 If the task needs to know its input or output DataIds then it also has
130 to override the `runQuantum` method.
132 This method should return a `Struct` whose attributes share the same
133 name as the connection fields describing output dataset types.
135 Parameters
136 ----------
137 **kwargs : `~typing.Any`
138 Arbitrary parameters accepted by subclasses.
140 Returns
141 -------
142 struct : `Struct`
143 Struct with attribute names corresponding to output connection
144 fields.
146 Examples
147 --------
148 Typical implementation of this method may look like:
150 .. code-block:: python
152 def run(self, *, input, calib):
153 # "input", "calib", and "output" are the names of the
154 # connection fields.
156 # Assuming that input/calib datasets are `scalar` they are
157 # simple objects, do something with inputs and calibs, produce
158 # output image.
159 image = self.makeImage(input, calib)
161 # If output dataset is `scalar` then return object, not list
162 return Struct(output=image)
163 """
164 raise NotImplementedError("run() is not implemented")
166 def runQuantum(
167 self,
168 butlerQC: QuantumContext,
169 inputRefs: InputQuantizedConnection,
170 outputRefs: OutputQuantizedConnection,
171 ) -> None:
172 """Do butler IO and transform to provide in memory
173 objects for tasks `~Task.run` method.
175 Parameters
176 ----------
177 butlerQC : `QuantumContext`
178 A butler which is specialized to operate in the context of a
179 `lsst.daf.butler.Quantum`.
180 inputRefs : `InputQuantizedConnection`
181 Datastructure whose attribute names are the names that identify
182 connections defined in corresponding `PipelineTaskConnections`
183 class. The values of these attributes are the
184 `lsst.daf.butler.DatasetRef` objects associated with the defined
185 input/prerequisite connections.
186 outputRefs : `OutputQuantizedConnection`
187 Datastructure whose attribute names are the names that identify
188 connections defined in corresponding `PipelineTaskConnections`
189 class. The values of these attributes are the
190 `lsst.daf.butler.DatasetRef` objects associated with the defined
191 output connections.
192 """
193 inputs = butlerQC.get(inputRefs)
194 outputs = self.run(**inputs)
195 butlerQC.put(outputs, outputRefs)