Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-4780

meas_extensions_shapeHSM seems to be broken

    Details

      Description

      I have installed the meas_extensions_shapeHSM package together with galsim and tmv (I documented it at : https://github.com/DarkEnergyScienceCollaboration/ReprocessingTaskForce/wiki/Installing-the-LSST-DM-stack-and-the-related-packages#installing-meas_extensions_shapehsm) and tried to run it on CFHT cluster data.

      My config file is the following:

      import lsst.meas.extensions.shapeHSM
      config.measurement.plugins.names |= ["ext_shapeHSM_HsmShapeRegauss", "ext_shapeHSM_HsmMoments",
                                          "ext_shapeHSM_HsmPsfMoments"]
      config.measurement.plugins['ext_shapeHSM_HsmShapeRegauss'].deblendNChild=''
      config.measurement.slots.shape = "ext_shapeHSM_HsmMoments"
      

      When I run measCoaddSources.py, I get the following error :

      Traceback (most recent call last):
        File "/sps/lsst/Library/lsstsw/stack/Linux64/pipe_tasks/2015_10.0-10-g1170fd0/bin/measureCoaddSources.py", line 3, in <module>
          MeasureMergedCoaddSourcesTask.parseAndRun()
        File "/sps/lsst/Library/lsstsw/stack/Linux64/pipe_base/2015_10.0-3-g24e103a/python/lsst/pipe/base/cmdLineTask.py", line 444, in parseAndRun
          resultList = taskRunner.run(parsedCmd)
        File "/sps/lsst/Library/lsstsw/stack/Linux64/pipe_base/2015_10.0-3-g24e103a/python/lsst/pipe/base/cmdLineTask.py", line 192, in run
          if self.precall(parsedCmd):
        File "/sps/lsst/Library/lsstsw/stack/Linux64/pipe_base/2015_10.0-3-g24e103a/python/lsst/pipe/base/cmdLineTask.py", line 279, in precall
          task = self.makeTask(parsedCmd=parsedCmd)
        File "/sps/lsst/Library/lsstsw/stack/Linux64/pipe_base/2015_10.0-3-g24e103a/python/lsst/pipe/base/cmdLineTask.py", line 363, in makeTask
          return self.TaskClass(config=self.config, log=self.log, butler=butler)
        File "/sps/lsst/Library/lsstsw/stack/Linux64/pipe_tasks/2015_10.0-10-g1170fd0/python/lsst/pipe/tasks/multiBand.py", line 530, in __init__
          self.makeSubtask("measurement", schema=self.schema, algMetadata=self.algMetadata)
        File "/sps/lsst/Library/lsstsw/stack/Linux64/pipe_base/2015_10.0-3-g24e103a/python/lsst/pipe/base/task.py", line 255, in makeSubtask
          subtask = configurableField.apply(name=name, parentTask=self, **keyArgs)
        File "/sps/lsst/Library/lsstsw/stack/Linux64/pex_config/2015_10.0-1-gc006da1/python/lsst/pex/config/configurableField.py", line 77, in apply
          return self.target(*args, config=self.value, **kw)
        File "/sps/lsst/dev/lsstprod/clusters/my_packages/meas_base/python/lsst/meas/base/sfm.py", line 247, in __init__
          self.initializePlugins(schema=self.schema)
        File "/sps/lsst/dev/lsstprod/clusters/my_packages/meas_base/python/lsst/meas/base/baseMeasurement.py", line 298, in initializePlugins
          self.plugins[name] = PluginClass(config, name, metadata=self.algMetadata, **kwds)
        File "/sps/lsst/dev/lsstprod/clusters/my_packages/meas_base/python/lsst/meas/base/wrappers.py", line 15, in __init__
          self.cpp = self.factory(config, name, schema, metadata)
        File "/sps/lsst/dev/lsstprod/clusters/my_packages/meas_base/python/lsst/meas/base/wrappers.py", line 223, in factory
          return AlgClass(config.makeControl(), name, schema)
        File "/sps/lsst/dev/lsstprod/clusters/my_packages/meas_extensions_shapeHSM/python/lsst/meas/extensions/shapeHSM/hsmLib.py", line 964, in __init__
          def __init__(self, *args, **kwargs): raise AttributeError("No constructor defined - class is abstract")
      AttributeError: No constructor defined - class is abstract
      

        Attachments

          Issue Links

            Activity

            Hide
            jbosch Jim Bosch added a comment -

            John Swinbank Dominique Boutigny This seems to have gotten lost. Could we perhaps merge the fix that's already in, and open a new issue for the unsolved problem, if that's still worth chasing down?

            Show
            jbosch Jim Bosch added a comment - John Swinbank Dominique Boutigny This seems to have gotten lost. Could we perhaps merge the fix that's already in, and open a new issue for the unsolved problem, if that's still worth chasing down?
            Hide
            swinbank John Swinbank added a comment -

            Not quite lost, in that I'm still aware it needs to be finished, but I agree that it's pointless to block the fix we already have from going in until we track down some likely unrelated issue. I'm happy to merge & open a new ticket if Dominique Boutigny signs off on this.

            Show
            swinbank John Swinbank added a comment - Not quite lost, in that I'm still aware it needs to be finished, but I agree that it's pointless to block the fix we already have from going in until we track down some likely unrelated issue. I'm happy to merge & open a new ticket if Dominique Boutigny signs off on this.
            Hide
            boutigny Dominique Boutigny added a comment -

            I am now back to France and almost ready to start working again. I will have a look to it in the coming days

            Show
            boutigny Dominique Boutigny added a comment - I am now back to France and almost ready to start working again. I will have a look to it in the coming days
            Hide
            boutigny Dominique Boutigny added a comment -

            Sorry for the delay. That's ok to merge this branch and to re-open another ticket for the segfault which is thrown at : https://github.com/lsst/meas_extensions_shapeHSM/blob/master/src/HsmMoments.cc#L115

            Show
            boutigny Dominique Boutigny added a comment - Sorry for the delay. That's ok to merge this branch and to re-open another ticket for the segfault which is thrown at : https://github.com/lsst/meas_extensions_shapeHSM/blob/master/src/HsmMoments.cc#L115
            Hide
            swinbank John Swinbank added a comment -

            Changes as reviewed merged to master. DM-5247 created to catch the segfault.

            Show
            swinbank John Swinbank added a comment - Changes as reviewed merged to master. DM-5247 created to catch the segfault.

              People

              • Assignee:
                swinbank John Swinbank
                Reporter:
                boutigny Dominique Boutigny
                Reviewers:
                Dominique Boutigny
                Watchers:
                Dominique Boutigny, Jim Bosch, John Swinbank
              • Votes:
                0 Vote for this issue
                Watchers:
                3 Start watching this issue

                Dates

                • Created:
                  Updated:
                  Resolved:

                  Summary Panel