id
int64 0
5.38k
| issuekey
stringlengths 4
16
| created
stringlengths 19
19
| title
stringlengths 5
252
| description
stringlengths 1
1.39M
| storypoint
float64 0
100
|
|---|---|---|---|---|---|
1,999 |
DM-9766
|
03/10/2017 16:24:59
|
FITS "can opener"
|
It would be useful to expose in Firefly applications the ability to read an arbitrary FITS file, browse its headers and any extensions, and be able to visualize whatever subset of data types Firefly can natively handle (e.g., image, basic table). Mutatis mutandi for other file types we support. This would be a useful debugging tool, but it would also be useful in the JupyterLab environment where plugins can be associated with specific "file" types in the file browser component. Then, if users have FITS files in their workspace storage, they can open them in the JupyterLab environment with simple UI actions. It may also be of interest in other Firefly application contexts, e.g., in IRSA, to offer this capability, but of course there are security issues associated with providing a general "upload a FITS file and we'll tell you all about it" capability.
| 0 |
2,000 |
DM-9769
|
03/10/2017 17:37:25
|
Provide SLAC information on image cutout, reference IRSA existing practice
|
SLAC has delivered imageServ APIs to retrieve the image cutouts but they have questions about how to provide the cutouts for coadds (per Kenny Lo). They would like our team to provide how-to. Thanks to Tatiana who points out that IRSA has been using the cutouts algorithm in http://irsatest.ipac.caltech.edu//ibe/cutouts.html. Basically users provide the center coordinates and the size of the cutout. "The size parameter consists of one or two (comma separated) values followed by an optional units specification. Units can be pixels (px, pix, pixels) or angular (arcsec, arcmin, deg, rad); the default is degrees. The first size value (x) is taken to be the full-width of the desired cutout along the first image axis (NAXIS1), and the second (y) is taken to be the full-height along the second axis (NAXIS2). If only one size value is specified, it is used as both the full-width and full-height. Negative sizes are illegal." Should we suggest SLAC team to consider to adapt the same algorithm for LSST image cutouts, both single epoch and coadd? Need David C. and Gregory to consider this and approve it. Please see the full document in the link below.
| 3 |
2,001 |
DM-9775
|
03/13/2017 06:26:49
|
Create NMF deblender presentation
|
Create a presentation for the Princeton Monday Meeting to outline NMF, proximal operators, and the new deblender.
| 3 |
2,002 |
DM-9784
|
03/13/2017 11:42:05
|
Implement test for overlapping galaxies
|
Once DM-9644 is completed we need to create a test that calculates the amount of overlap between galaxies, basically (S{~}i~ S{~}j~)^2^ / ((∑S{~}i~^2^) (∑S{~}j~^2^)) for both the measured intensities and simulated intensities.
| 5 |
2,003 |
DM-9789
|
03/13/2017 16:37:14
|
pipelines.lsst.io v13 needs some fixes
|
After perusing the latest version of pipelines.lsst.io, I made the following comments on the merge commit: * {{metrics/v13_0.rst}} line 9: Should we explain that we are not trying to compare to the "glide path" in LDM-240 anymore and that the targets are hence goals for much later rather than thresholds we are supposed to meet with this release? * {{releases/note_source/v13_0.rst}} line 659: There's no release note that says that {{daf_butlerUtils}} was moved to {{obs_base}}? And I think the title of this would be a bit better if it had "definitions" appended. * {{releases/note_source/v13_0.rst}} line 786: If there were no documentation improvements, this section should be omitted. But I think there were some, at least to pipelines.lsst.io itself...
| 1 |
2,004 |
DM-9790
|
03/13/2017 17:07:43
|
Items missing from the pybind11 wrapping guide
|
I've been using the [pybind11 wrapping guide|https://developer.lsst.io/coding/python_wrappers_for_cpp_with_pybind11.html] to wrap jointcal, and it's generally been quite helpful, but there were a few questions it didn't answer for me. * what to change/add in {{python/SConscript}} and {{__init__.py}}. * how many of the wrapped modules to include in {{__init__}}: should they all be included as a general rule, or just those that one definitely wants to be accessible from python? * How to choose between {{clsBlah}} and {{cls}} when naming the wrapped code (it's discussed in the style guide, but it would be useful to say something about it in the wrapping guide, with a link to the style guide sections). * how to import new pybind11 modules in python (no more blahLib). It might be useful to also have a list of "how to remove all your SWIG stuff", for those dealing with packages that were previously SWIGed.
| 1 |
2,005 |
DM-9792
|
03/13/2017 18:11:10
|
Mapping::getInverse not exception-safe
|
The current implementation of {{Mapping::getInverse}} allocates a new {{AstMapping}}, but does not free it if AST's internal error flag triggers an exception via {{assertOK}}. -In addition, {{Object::fromAstObject}} does not free {{rawObjCopy}} if there is a dynamic casting error.- The leaks in these methods should be fixed using either {{assertOK}} or a {{catch}} or {{finally}} block. Unfortunately, as I understand it astshim's object and type juggling would interfere with using a {{unique_ptr}}. In addition, both methods should document the exceptions they throw and their (intended) exception safety.
| 2 |
2,006 |
DM-9794
|
03/13/2017 18:20:36
|
Pass both LSST_LIBRARY_PATH and DYLD_LIBRARY_PATH in scons on Mac OS
|
To allow both Python code and shell scripts to inherit the necessary env settings, I'd like {{sconsUtils.utils.libraryLoaderEnvironment}} to pass both {{LSST_LIBRARY_PATH}} and {{DYLD_LIBRARY_PATH}} to things it calls on Mac OS. Currently {{libraryLoaderEnvironment}} is used within {{sconsUtils}} only by {{sconsUtils.tests.Control.run}} to ensure that, on Mac OS, that {{DYLD_LIBRARY_PATH}} is passed with either {{DYLD_LIBRARY_PATH}} or {{LSST_LIBRARY_PATH}}, whichever is defined first in that search order. But a shell script will be called out of {{/bin}} or an equivalently system-protected directory and needs {{LSST_LIBRARY_PATH}} passed (which will not be stripped) so that that shell script can then reset {{DYLD_LIBRARY_PATH}} (which will be stripped by SIP). The particularly motivating usage is to enable Python test code to call a shell script as part of DM-9740.
| 2 |
2,007 |
DM-9795
|
03/13/2017 20:02:27
|
CModel priors are weighted incorrectly relative to likelihood
|
When per-pixel variances are turned off, CModel likelihoods are computed without using the variance at all. This would not matter in a pure likelihood fit, but it means the prior and likelihood are not given the appropriate relative weights - and the relative weighting is not even consistent; it depends on the noise level of the image. Since the typical effect of this is to make the prior much more informative, there is some danger that fixing this bug will cause other problems due to poorly-constrained fits. To avoid this, I'll add a configuration option to tune the relative weighting of the prior via a constant (which we could set to the typical variance level of the images to get behavior like what we have now without the inconsistency).
| 2 |
2,008 |
DM-9797
|
03/14/2017 01:59:19
|
Investigate HSC-Y background configs
|
Default background configs fail to remove a rotating ellipse-shaped ring of scattered light in the y-band HSC data: !https://paper.dropbox.com/ep/redirect/image?url=https%3A%2F%2Fd2mxuefqeaa7sj.cloudfront.net%2Fs_CE3712A3B91055C667598BDAAFD85C792A8BC38908F5D92E32C19E3223FF4694_1489481658883_Screen%2BShot%2B2017-03-14%2Bat%2B5.23.22%2BPM.png&hmac=8V%2F3Mr6bDlLZbvSguW8pJDZRR9CzFz1%2Fr85NKoWsCjc%3D&width=600! This ticket will capture work to identify appropriate background configs for removing.
| 2 |
2,009 |
DM-9798
|
03/14/2017 08:09:17
|
Remove superfluous print message
|
During the pybind11 port someone (meaning me, probably) left a print statement that print "STRAY FLUX". This should be removed.
| 1 |
2,010 |
DM-9802
|
03/14/2017 10:30:16
|
Update base docker images for alert stream
|
The base kafka image in the docker-compose file for the current alert_stream repo is deprecated already. :( Need to swap this image out with the current stable version of kafka, go through new release docs to update deprecated environment variables, swap out the zookeeper image, and upgrade the python base image of the main Dockerfile. We want to do this before testing because the currently used version of kafka and also the python base image lack features/performance we will want later, so we should test the versions we want to use.
| 2 |
2,011 |
DM-9808
|
03/14/2017 15:59:44
|
astshim does not build on Ubuntu
|
The stack cannot be built on Ubuntu because the build of {{astshim}} complains about a missing symbol: bq. lsstsw2/stack/Linux64/starlink_ast/lsst-dev-g4bd2b55bd7/lib/libast_pass2.so: undefined reference to 'astPutErr_'
| 1 |
2,012 |
DM-9810
|
03/14/2017 17:11:05
|
Make PSFEx oversampling configurable
|
Add a configuration option that lets PSFEx decide when to oversample PSF images.
| 1 |
2,013 |
DM-9811
|
03/14/2017 20:06:23
|
Add 1.3 arcsec target seeing for convolved flux measurement
|
Masayuki Tanaka asks that we also include a 1.3 arcsec target seeing in the {{ConvolvedFluxPlugin}}. This target seeing is larger than any expected seeing in the coadd, and so we'll always get a useful result.
| 1 |
2,014 |
DM-9812
|
03/14/2017 22:41:03
|
Clean up outputs from CharacterizeImageTask and CalibrateTask
|
We're writing the {{icExp}} files in {{CharacterizeImageTask}}, which we don't use, but greatly increases our disk usage. We want the denormalized match catalogs from {{CalibrateTask}} and from multiband processing.
| 2 |
2,015 |
DM-9816
|
03/15/2017 10:54:19
|
Evaluate Plotly.js as a replacement to Highcharts
|
Plotly is an open source plotting package that is more oriented toward science than Highcharts. Study the following: * Can we control tooltips? * What control do we have during selection? * In general, can monitor or have hooks into their events? * How hard to port histograms, scatter and density? * How hard to create a heat map to replace density? * Do they support optimized updates and not complete redraws? Overall we want a sense if we should go to the next step of replacing Highcharts. We need to understand how much efforts would be involved.
| 8 |
2,016 |
DM-9817
|
03/15/2017 11:30:58
|
Doxygen tries to parse pybind11 wrappers
|
Doxygen currently reads {{.cc}} files in the {{python/}} directory, leading to silly documentation like https://lsst-web.ncsa.illinois.edu/doxygen/x_masterDoxyDoc/namespacelsst_1_1afw_1_1camera_geom.html#a2a868ed61ae08543a5cc793c35f5674e. I propose that any C++ files in {{python/}} be excluded using an {{EXCLUDE}} or {{EXCLUDE_PATTERNS}} configuration. I'm not sure whether this only needs to be changed in {{base}} or in all packages.
| 2 |
2,017 |
DM-9818
|
03/15/2017 11:41:17
|
Design dochub-prototype (www.lsst.io builder)
|
Write up a lightweight specification+design for the {{dochub-prototype}} project (the build infrastructure for www.lsst.io) to help [~athornton] make the initial implementation. {{dochub-prototype}} is itself a prototype for what will become DocHub (https://sqr-013.lsst.io).
| 1 |
2,018 |
DM-9824
|
03/15/2017 14:04:01
|
Fail to upload additional catalog after initial upload
|
Select Catalogs -> Load Catalog Upload the sample catalog file below. Click Search. Everything works as intended. Now, try to upload another catalog. Clicking Search does nothing. No error indications. You'll find uncaught errors in the console. sample catalog file: {noformat} | ra | dec | | double | double | 66.76957 26.10453 194.25975 22.03131 {noformat}
| 0 |
2,019 |
DM-9826
|
03/15/2017 17:07:31
|
Add hasTranForward and hasTranInverse to Transform
|
Like the {{Mappings}} they adapt, a {{Transform}} may or may not have a forward transformation, and may or may not have an inverse transformation. Test methods should be added to {{Transform}} to let clients defend against calling a missing transformation.
| 2 |
2,020 |
DM-9828
|
03/16/2017 00:25:39
|
Enable rectangular binSizes in SubtractBackgroundTask
|
While validating on HSC data, we encountered a use case for rectangular bin sizes in {{SubtractBackgroundTask}}. We propose two new config fields: * binSizeX * binSizeY By default they will be None and populated with the value in binSize. This change will be 100% backwards compatible. Everyone's config files will produce the same results.
| 2 |
2,021 |
DM-9846
|
03/16/2017 13:41:50
|
Improve handling of error messages
|
astshim presently prints AST error messages to stderr. Improve this so that the messages and up as the text of exceptions, instead. This work requires the new AST capability of registering an error handler which has just been released.
| 1 |
2,022 |
DM-9848
|
03/16/2017 14:20:30
|
obs_subaru test failures possibly related to daf_persistence
|
Both [~wmwood-vasey] and myself have seen {{obs_subaru}} test failures today. In my case, the test failure was non-reproducible. I am concerned that this might be a concurrent butler access issue. https://ci.lsst.codes/job/release/job/run-rebuild/472/consoleFull#console-section-13 {code:java} [2017-03-16T18:08:21.499615Z] Traceback (most recent call last): [2017-03-16T18:08:21.499681Z] Traceback (most recent call last): [2017-03-16T18:08:21.499703Z] File "bin.src/genDefectFits.py", line 86, in <module> [2017-03-16T18:08:21.499727Z] File "bin.src/genDefectFits.py", line 86, in <module> [2017-03-16T18:08:21.499748Z] genDefectFits(args.cameraName, args.defectsFile, args.targetDir) [2017-03-16T18:08:21.499771Z] genDefectFits(args.cameraName, args.defectsFile, args.targetDir) [2017-03-16T18:08:21.499789Z] File "bin.src/genDefectFits.py", line 20, in genDefectFits [2017-03-16T18:08:21.499808Z] File "bin.src/genDefectFits.py", line 20, in genDefectFits [2017-03-16T18:08:21.499825Z] mapper = mapperMap[cameraName.lower()](root=".") [2017-03-16T18:08:21.499844Z] mapper = mapperMap[cameraName.lower()](root=".") [2017-03-16T18:08:21.499875Z] File "/home/lsstsw/jenkins/release/lsstsw/build/obs_subaru/python/lsst/obs/hsc/hscMapper.py", line 31, in __init__ [2017-03-16T18:08:21.499912Z] File "/home/lsstsw/jenkins/release/lsstsw/build/obs_subaru/python/lsst/obs/hsc/hscMapper.py", line 31, in __init__ [2017-03-16T18:08:21.499936Z] super(HscMapper, self).__init__(policy, policyFile.getRepositoryPath(), **kwargs) [2017-03-16T18:08:21.499974Z] File "/home/lsstsw/jenkins/release/lsstsw/stack/Linux64/obs_base/13.0-5-g36cc3d4+1/python/lsst/obs/base/cameraMapper.py", line 223, in __init__ [2017-03-16T18:08:21.500009Z] super(HscMapper, self).__init__(policy, policyFile.getRepositoryPath(), **kwargs) [2017-03-16T18:08:21.500050Z] File "/home/lsstsw/jenkins/release/lsstsw/stack/Linux64/obs_base/13.0-5-g36cc3d4+1/python/lsst/obs/base/cameraMapper.py", line 223, in __init__ [2017-03-16T18:08:21.500392Z] calibStorage = dafPersist.Storage.makeFromURI(uri=calibRoot) [2017-03-16T18:08:21.500448Z] File "/home/lsstsw/jenkins/release/lsstsw/stack/Linux64/daf_persistence/13.0-4-g886c4ba/python/lsst/daf/persistence/storage/storageContinued.py", line 144, in makeFromURI [2017-03-16T18:08:21.500470Z] calibStorage = dafPersist.Storage.makeFromURI(uri=calibRoot) [2017-03-16T18:08:21.500516Z] File "/home/lsstsw/jenkins/release/lsstsw/stack/Linux64/daf_persistence/13.0-4-g886c4ba/python/lsst/daf/persistence/storage/storageContinued.py", line 144, in makeFromURI [2017-03-16T18:08:21.500969Z] CameraMapper INFO: Unable to locate registry registry in root: registry.sqlite3 [2017-03-16T18:08:21.501428Z] CameraMapper INFO: Unable to locate registry registry in current dir: ./registry.sqlite3 [2017-03-16T18:08:21.501771Z] CameraMapper INFO: Loading Posix registry from . [2017-03-16T18:08:21.502249Z] CameraMapper INFO: Unable to locate calibRegistry registry in root: calibRegistry.sqlite3 [2017-03-16T18:08:21.502676Z] CameraMapper INFO: Unable to locate calibRegistry registry in current dir: ./calibRegistry.sqlite3 [2017-03-16T18:08:21.503010Z] CameraMapper INFO: Loading Posix registry from ./CALIB [2017-03-16T18:08:21.506297Z] CameraMapper INFO: Unable to locate registry registry in root: registry.sqlite3 [2017-03-16T18:08:21.507001Z] CameraMapper INFO: Unable to locate registry registry in current dir: ./registry.sqlite3 [2017-03-16T18:08:21.507448Z] CameraMapper INFO: Loading Posix registry from . [2017-03-16T18:08:21.507975Z] CameraMapper INFO: Unable to locate calibRegistry registry in root: calibRegistry.sqlite3 [2017-03-16T18:08:21.508473Z] CameraMapper INFO: Unable to locate calibRegistry registry in current dir: ./calibRegistry.sqlite3 [2017-03-16T18:08:21.508875Z] CameraMapper INFO: Loading Posix registry from ./CALIB [2017-03-16T18:08:21.513985Z] ret = theClass(uri=uri) [2017-03-16T18:08:21.514008Z] ret = theClass(uri=uri) [2017-03-16T18:08:21.514060Z] File "/home/lsstsw/jenkins/release/lsstsw/stack/Linux64/daf_persistence/13.0-4-g886c4ba/python/lsst/daf/persistence/posixStorage.py", line 55, in __init__ [2017-03-16T18:08:21.514114Z] File "/home/lsstsw/jenkins/release/lsstsw/stack/Linux64/daf_persistence/13.0-4-g886c4ba/python/lsst/daf/persistence/posixStorage.py", line 55, in __init__ [2017-03-16T18:08:21.520395Z] os.makedirs(self.root) [2017-03-16T18:08:21.520437Z] File "/home/lsstsw/jenkins/release/lsstsw/miniconda/lib/python2.7/os.py", line 157, in makedirs [2017-03-16T18:08:21.520468Z] os.makedirs(self.root) [2017-03-16T18:08:21.520512Z] File "/home/lsstsw/jenkins/release/lsstsw/miniconda/lib/python2.7/os.py", line 157, in makedirs [2017-03-16T18:08:21.536157Z] mkdir(name, mode) [2017-03-16T18:08:21.536186Z] mkdir(name, mode) [2017-03-16T18:08:21.536237Z] OSError: [Errno 17] File exists: './CALIB' [2017-03-16T18:08:21.536264Z] OSError: [Errno 17] File exists: './CALIB' [2017-03-16T18:08:21.619152Z] scons: *** [hsc/defects/2014-06-01/defects.dat-fits] Error 1 [2017-03-16T18:08:21.623222Z] scons: *** [hsc/defects/2013-01-31/defects.dat-fits] Error 1 {code}
| 1 |
2,023 |
DM-9851
|
03/16/2017 15:59:46
|
Improve DCR code use of the butler
|
The current DCR template generation code requires the user to supply the name of the telescope, and has telescope-specific code that interfaces with the butler. This code should be re-written to be less fragile and to remove the lines that depend on the telescope.
| 2 |
2,024 |
DM-9857
|
03/17/2017 08:53:35
|
Research Support of Evaluation of Plotly as a replacement to Highchartts
|
Plotly is an open source plotting package that is more oriented toward science than Highcharts. Provide a second set of eyes in evaluation of Plotly.js and support DM-9816. Study some of the following: Can we control tooltips? What control do we have during selection? In general, can monitor or have hooks into their events? How hard to port histograms, scatter and density? Do they support optimized updates and not complete redraws? How would a Plotly react component work? Overall we want a sense if we should go to the next step of replacing Highcharts. We need to understand how much efforts would be involved.
| 8 |
2,025 |
DM-9862
|
03/17/2017 13:18:15
|
Update meas_mosaic's wcs/fcr output files to reflect LSST coordinate system
|
When importing {{meas_mosaic}}, the coordinate system for writing out the wcs/fcr files was not adapted to that expected from LSST (which always associates the detector origin with the electronics, whereas HSC's is such that a given detector's origin, pixel (0, 0) is associated with its LLC w.r.t. to the focal plane), but rather the tasks in {{meas_mosaic}}'s *updateExposure.py* were adapted to account for the rotated CCDs. It was assumed that this was the only place those corrections were every used. This turns out not to be the case since the wcs used to create the coadd gets attached to it in *coaddInputs*. If {{meas_mosaic}} was run and *doApplyUberCal=True* (which are both the case for our HSC data processing), the wcs’s that are getting attached to the coaddInputs are not in the coordinate system appropriate for LSST for any nQuarter != 0 ccds. Since this is already causing an issue in {{pipe_tasks}}'s *propogateVisitFlags.py*, see issue highlighted in DM-9383, (and could very well be causing issues elsewhere as yet undiscovered), we have decided these outputs need to be written out in the coordinate system expected by LSST.
| 3 |
2,026 |
DM-9863
|
03/17/2017 13:34:45
|
Replace use of makeVisitInfo(... with VisitInfo(...)
|
{{lsst.afw.image.makeVisitInfo(...)}} is superseded by {{lsst.afw.image.VisitInfo(...)}}. Remove existing usage of {{makeVisitInfo}}. {{makeVisitInfo}} is used in one or two places in most obs_ packages and a few other packages as well. Note that this can be done on a package by package basis as time permits. Once all usage is gone we can get rid of {{makeVisitInfo}}.
| 0.5 |
2,027 |
DM-9866
|
03/17/2017 15:49:42
|
Make change to remove flagDecorator (RFC-302)
|
This is a follow up to the FlagHandler modification DM-9249. Posted this last deletion to flagDecorator.py in case anyone unknown was using it.
| 1 |
2,028 |
DM-9871
|
03/20/2017 07:43:03
|
Move wcs and fcr datasets out of {pointing} directory in obs_subaru
|
We should change the templates for {{fcr}} and {{wcs}} to the following: {code} jointcal-results/%(tract)d/wcs-%(visit)07d-%(ccd)03d.fits jointcal-results/%(tract)d/fcr-%(visit)07d-%(ccd)03d.fits {code} This will avoid problems in copying/updating directories at NAOJ during incremental releases.
| 0.5 |
2,029 |
DM-9873
|
03/20/2017 09:16:42
|
PropertySet does not support values of None
|
In DM-5466, we needed to pass the results from {{ParseTask.getInfo()}} to the butler as a dataId. This is normally valid, since both are dictionaries, and even though {{getInfo()}} often contains extraneous entries that aren't relevant, the butler will usually ignore them. However, when parsing calibration files this dictionary contains some values that are set to {{None}}, since they will be filled in later. These extraneous keys are then placed in {{ButlerLocation.additionalData}} ([butlerLocation.py:221|https://github.com/lsst/daf_persistence/blob/master/python/lsst/daf/persistence/butlerLocation.py#L221]), which throws an exception as it is a PropertySet and does not support python {{None}} as a value. DM-5466 has a work-around that strips these None values from the dictionary, but this is inelegant. The main driver for excluding None from PropertySet seems to compatibility with FITS headers. This seems like an unwarranted mixing of data model and persistence formats. Unless there is some advantage to not being able to store None in our dictionary-like objects, it seems preferable to shift the burden of accommodating FITS's peculiarities onto the persistence layer rather than PropertySet.
| 3 |
2,030 |
DM-9882
|
03/20/2017 17:17:59
|
Add integrate interface to BoundedField/ChebyshevBoundedField
|
In order to implement the "mean zero point" functionality of PhotoCalib, I need to compute the integral over the bounding box. {{BoundedField}} does not include facilities for integration/differentiation. Fortunately, my primary usage is {{ChebyshevBoundedField}}, which has a relatively straight-forward recurrence relation for integration. I suggest an interface like the following on {{BoundedField}}, with a virtual "not implemented" default: {code} // Compute the integral of this function over its bounding-box. double integrate() const; {code}
| 8 |
2,031 |
DM-9885
|
03/21/2017 03:39:40
|
Rename deepCoadd_srcMatch as deepCoadd_measMatch
|
Implement RFC-306.
| 1 |
2,032 |
DM-9893
|
03/21/2017 11:40:32
|
Write SQuaRE User Stories for SuperTask
|
Write a set of user stories and consequent SuperTask design implications that address SQuaRE needs.
| 1 |
2,033 |
DM-9894
|
03/21/2017 12:09:39
|
SQuaRE SuperTask Collaboration Week of 2017-03-20
|
Bucket ticket to cover non-specific SuperTask collaboration activities (see linked meeting notes).
| 1 |
2,034 |
DM-9895
|
03/21/2017 12:17:01
|
FrameSet frames not preserved by Transform(frameSet) constructor
|
When an {{lsst.afw.geom.Transform}} is constructed from an {{astshim.FrameSet}} the internal frames in the frame set are lost. I strongly suspect what is happening is that the mapping constructor of {{Transform}} is being called, instead of the {{FrameSet}} constructor. Reversing the order in the pybind11 wrapper file should fix the problem. A unit test is required.
| 2 |
2,035 |
DM-9897
|
03/21/2017 13:39:39
|
conda channel errors causing lsstsw/bin/deploy to fail
|
We see errors installing both osx and linux packages like the following: {code:java} CondaRuntimeError: Runtime error: RuntimeError: Runtime error: Could not open u'/Users/square/jenkins/workspace/release/tarball/DarwinX86/miniconda2/4.2.12.lsst1/pkgs/mkl-11.3.3-0.tar.bz2.part' for writing (HTTPSConnectionPool(host='repo.continuum.io', port=443): Read timed out.). {code} I believe the best resolution is to mirror the public conda channels internally. Another, less desirable, option would be build retrying into the conda command.
| 3 |
2,036 |
DM-9900
|
03/21/2017 16:08:00
|
Recreate DCR simulation images for testing
|
The existing simulation data used for testing the DCR template building code is missing some important metadata. Those simulations should be brought up to date, and re-created with all the required metadata.
| 2 |
2,037 |
DM-9916
|
03/22/2017 15:06:03
|
Re-write internal DCR template wcs-matching
|
In the DCR template generation code, when exposures are warped that operation is currently performed in-place, over-writing the array values and wcs info. This could potentially lead to bugs down the road, so either the implementation should be made more robust or checks should be put in place to verify that it is safe.
| 3 |
2,038 |
DM-9917
|
03/22/2017 18:49:40
|
Add a callback to cameraGeom.showCamera
|
It is sometimes very convenient to modify the data that {{cameraGeom.utils.showCamera}} displays; the classic case if trimming, bias subtracting, or gain-correcting raw data. Please port old RHL HSC code that registers a callback to do the work (much better than adding more and more special-case code to {{showCamera}})
| 1 |
2,039 |
DM-9920
|
03/23/2017 11:20:59
|
The catalog search in IRSAViewer does not work for multi-object search
|
While working on LSST multi-object search, I found that the multi-object search in IRSA viewer does not work. After spending hours to look into the reason, I found that the uploaded table in CatalogSelectViewPanel.jsx is not proper defined. {code} if (spatial === SpatialMethod.get('Multi-Object').value) { var filename = get(catPart, 'fileUpload'); {code} The correct way should be: {code} if (spatial === SpatialMethod.get('Multi-Object').value) { var filename = get(spacPart, 'fileUpload'); {code}
| 2 |
2,040 |
DM-9921
|
03/23/2017 11:24:41
|
Uploader for dochub-prototype HTML to LSST the Docs
|
Publish dochub-prototype HTML to {{www.lsst.io}} via LSST the Docs.
| 2 |
2,041 |
DM-9925
|
03/23/2017 17:49:26
|
PolyTran should not provide an iterative inverse by default
|
{{PolyTran}}'s constructor that takes only forward coefficients provides an iterative inverse by default (as does its other constructor if the forward coefficient array is empty). This seems like a bad idea for several reasons: - Most polynomials do not have unique inverses. That said, the iterative inverse will do sensible things in that situation (nan if no value exists, else a valid value if multiple choices exist). - The iterative inverse is much less efficient to compute than the fit inverse provided by `PolyMap.polyTran`. I propose to disable the iterative inverse by default, using the philosophy of "when in doubt, refuse to guess".
| 1 |
2,042 |
DM-9931
|
03/24/2017 10:27:04
|
Add dependency of "refcat" on "preSfm" in ci_hsc
|
{{ci_hsc}} fails in my lsst-dev/lsstvc/slurm env (using 1 node) although it passes on Jenkins. The problem arose when {{preSfm}} ran before {{refcat}}. I think this is because {{preSfm}}, effectively {{processCcd.py DATA --rerun ci_hsc --doraise}}, needs the {{ref_cats}} link already in the repo. Running the command in a repo without the {{ref_cats}} gives this error: {code:java} root INFO: Running: /home/hchiang2/lsstsw/stack/Linux64/pipe_tasks/13.0-8-g148fcaf+1/bin/processCcd.py /home/hchiang2/ciHscTest/asMaster/ci_hsc/DATA --rerun test2 --doraise Traceback (most recent call last): File "/home/hchiang2/lsstsw/stack/Linux64/pipe_tasks/13.0-8-g148fcaf+1/bin/processCcd.py", line 25, in <module> ProcessCcdTask.parseAndRun() File "/home/hchiang2/lsstsw/stack/Linux64/pipe_base/13.0-1-g92030ba+10/python/lsst/pipe/base/cmdLineTask.py", line 482, in parseAndRun resultList = taskRunner.run(parsedCmd) File "/home/hchiang2/lsstsw/stack/Linux64/pipe_base/13.0-1-g92030ba+10/python/lsst/pipe/base/cmdLineTask.py", line 202, in run if self.precall(parsedCmd): File "/home/hchiang2/lsstsw/stack/Linux64/pipe_base/13.0-1-g92030ba+10/python/lsst/pipe/base/cmdLineTask.py", line 300, in precall task = self.makeTask(parsedCmd=parsedCmd) File "/home/hchiang2/lsstsw/stack/Linux64/pipe_base/13.0-1-g92030ba+10/python/lsst/pipe/base/cmdLineTask.py", line 391, in makeTask return self.TaskClass(config=self.config, log=self.log, butler=butler) File "/home/hchiang2/lsstsw/stack/Linux64/pipe_tasks/13.0-8-g148fcaf+1/python/lsst/pipe/tasks/processCcd.py", line 160, in __init__ astromRefObjLoader=astromRefObjLoader, photoRefObjLoader=photoRefObjLoader) File "/home/hchiang2/lsstsw/stack/Linux64/pipe_base/13.0-1-g92030ba+10/python/lsst/pipe/base/task.py", line 237, in makeSubtask subtask = taskField.apply(name=name, parentTask=self, **keyArgs) File "/home/hchiang2/lsstsw/stack/Linux64/pex_config/13.0-1-g41367f3+1/python/lsst/pex/config/configurableField.py", line 83, in apply return self.target(*args, config=self.value, **kw) File "/home/hchiang2/lsstsw/stack/Linux64/pipe_tasks/13.0-8-g148fcaf+1/python/lsst/pipe/tasks/calibrate.py", line 322, in __init__ self.makeSubtask('astromRefObjLoader', butler=butler) File "/home/hchiang2/lsstsw/stack/Linux64/pipe_base/13.0-1-g92030ba+10/python/lsst/pipe/base/task.py", line 237, in makeSubtask subtask = taskField.apply(name=name, parentTask=self, **keyArgs) File "/home/hchiang2/lsstsw/stack/Linux64/pex_config/13.0-1-g41367f3+1/python/lsst/pex/config/configurableField.py", line 83, in apply return self.target(*args, config=self.value, **kw) File "/home/hchiang2/lsstsw/stack/Linux64/meas_algorithms/13.0-6-g6f8a914+2/python/lsst/meas/algorithms/loadIndexedReferenceObjects.py", line 50, in __init__ dataset_config = butler.get("ref_cat_config", name=self.config.ref_dataset_name, immediate=True) File "/home/hchiang2/lsstsw/stack/Linux64/daf_persistence/13.0-5-g87674b4/python/lsst/daf/persistence/butler.py", line 945, in get raise NoResults("No locations for get:", datasetType, dataId) lsst.daf.persistence.butlerExceptions.NoResults: No locations for get: datasetType:ref_cat_config dataId:DataId(initialdata={'name': 'ps1_pv3_3pi_20170110'}, tag=set([])) {code}
| 1 |
2,043 |
DM-9937
|
03/24/2017 12:37:15
|
Add noexcept specifiers to applicable methods in afw
|
C++11 methods that are declared {{noexcept}} enable compilers to streamline code that calls them, in some cases (e.g., STL code) unlocking more efficient algorithms that would be unsafe with throwing methods. These changes are most useful for low-level types, whose methods are also the most likely to be provably non-throwing. This ticket shall add the {{noexcept}} specifier to any methods that are guaranteed not to throw and are unlikely to be modified to throw exceptions in the future (e.g., trivial getters will usually qualify, but methods for which some inputs are invalid will not even if the existing code performs no input validation). Particular attention shall be paid to constructors, assignment operators, and {{std::swap}} implementations, as these are the methods where {{noexcept}} provides the biggest gains. The classes covered by this ticket are (based on assumed simplicity, and therefore subject to change): * {{afw::cameraGeom::CameraPoint}} * {{afw::cameraGeom::CameraSys}} * {{afw::cameraGeom::CameraSysPrefix}} * {{afw::cameraGeom::Orientation}} * {{afw::coord::Coord}} and its subclasses * {{afw::coord::Observatory}} * {{afw::coord::Weather}} * {{afw::detection::Threshold}} * {{afw::geom::Angle}} * {{afw::geom::AngleUnit}} * {{afw::geom::Box}} * {{afw::geom::CoordinateBase}} and its subclasses * {{afw::geom::CoordinateExpr}} * {{afw::geom::Span}} * {{afw::geom::SpherePoint}} * {{afw::geom::polygon::Polygon}} * {{afw::image::Calib}} * {{afw::image::Color}} * {{afw::image::DefectBase}} * {{afw::image::Filter}} * {{afw::image::FilterProperty}} * {{afw::math::FitResults}} * {{afw::math::Function}} and its subclasses * {{afw::math::MaskedVector}} * {{afw::math::Statistics}} * {{afw::table::BaseRecord}} and its subclasses * {{afw::table::ConstFunctorKey}} * {{afw::table::FieldBase}} and its subclasses * {{afw::table::InputFunctorKey}} and its subclasses * {{afw::table::KeyBase}} and its subclasses * {{afw::table::Match}} * {{afw::table::OutputFunctorKey}} and its subclasses * {{afw::table::ReferenceFunctorKey}} * {{afw::table::SchemaItem}} * {{afw::table::io::Persistable}} * {{afw::table::io::PersistableFacade}}
| 3 |
2,044 |
DM-9940
|
03/24/2017 14:11:06
|
Simplify package dependencies of template generation
|
The DCR matched template code has dependencies on sims packages that might be replaced with functionality from core software. Once DM-9615 is completed, some of these (such as the bandpass) may be obtained from the relevant obs package.
| 3 |
2,045 |
DM-9942
|
03/24/2017 15:09:43
|
Plotly Change: Create Plotly loading infrastructure and React wrapper
|
Setup plotly environment: * Set up how we are going the load plotly using a deferred load * Create a plotly react wrapper * set up Firefly environment to contain a property to switch between plotting packages
| 3 |
2,046 |
DM-9946
|
03/24/2017 16:45:49
|
Remove debugging example refcount.cc
|
The file {{examples/refcount.cc}} was intended for debugging, not distribution. Remove it.
| 0 |
2,047 |
DM-9948
|
03/24/2017 16:52:50
|
LSST catalog search processor
|
make sure the current catalog search processor handles WISE catalog data search properly
| 8 |
2,048 |
DM-9949
|
03/24/2017 16:56:00
|
Update the PDAC sample queries and test cases page
|
Update the PDAC sample queries and test cases page [https://confluence.lsstcorp.org/display/DM/PDAC+sample+queries+and+test+cases] to include WISE data
| 2 |
2,049 |
DM-9954
|
03/27/2017 09:36:39
|
Optimize PSF convolution to use sparse matrices
|
Currently the PSF convolution operator is a collection of 2D sparse arrays, which are different for each band, that converted are into full arrays in order to use [~pmelchior]'s proximal operator algorithm. Since the current simulated data is PSF matched, the processing time can be significantly reduced (by several orders of magnitude) by using the PSF operator as a sparse matrix in each band. This fix is temporary so that we can profile and test PSF convolution and other features of the new deblender but it is expected that this section of the code will be re-written in C++ in the final deblender before it is merged into the stack in a future sprint.
| 2 |
2,050 |
DM-9955
|
03/27/2017 09:53:50
|
Investigate building stack on CentOS7 with devtoolset-6
|
RFC-303 approved the principle that we can mandate a CentOS7 baseline operating system that requires a devtoolset to build the stack. This ticket is to demonstrate that a stack can build and run using devtoolset-6 on CentOS7 without running in to problems. A successful outcome should result in an RFC requesting a new compiler baseline.
| 0.5 |
2,051 |
DM-9958
|
03/27/2017 10:43:26
|
SQuaRE SuperTask Collaboration Week of 2017-03-13
|
Bucket ticket to cover non-specific SuperTask collaboration activities (see linked meeting notes).
| 1 |
2,052 |
DM-9959
|
03/27/2017 11:04:30
|
SQuaRE SuperTask Collaboration Week of 2017-03-27
|
Bucket ticket to cover non-specific SuperTask collaboration activities.
| 1 |
2,053 |
DM-9961
|
03/27/2017 11:25:16
|
Add treecorr to validate_drp and lsst_ci dependencies
|
Create a package for {{treecorr}} to allow for calculation of correlation functions in a fast, robust, and tested code. This will not add dependencies to {{lsst_apps}} or {{lsst_distrib}}. Implementation is likely 1. Create wrapper install packages for {{treecorr}}. 2. Figure out how to avoid adding the FITS reader dependency in {{treecorr}}. 3. Add treecorr dependence to {{validate_drp}} table file. TreeCorr is a "[c]ode for efficiently computing 2-point and 3-point correlation functions" and is being proposed for use in {{validate_drp}} to calculate the correlation function of the PSF residual ellipticity (DM-8951). The GitHub repo is here: https://github.com/rmjarvis/TreeCorr TreeCorr depends on the following Python packages. numpy future fitsio pandas pyyaml cffi
| 3 |
2,054 |
DM-9962
|
03/27/2017 11:28:59
|
Write proposal for storing coadd HeavyFootprints
|
Deblending results in the form of HeavyFootprints should be available to users, which means they must be stored. This proposal should include an estimate of how much storage this will require for a given area of sky, which can probably be derived from the average HeavyFootprint size in HSC data. This will be a function of depth, so it will not be the same at the beginning of the survey as it will be at the end (but we should have HSC data that spans those depths). This will be a conservative estimate - it is possible that deblender outputs may take a form that allows them to be highly compressed and quickly reconstituted given coadd data. We will not consider the possibility that it will be necessary to store per-epoch HeavyFootprints for ForcedSources.
| 2 |
2,055 |
DM-9964
|
03/27/2017 12:14:16
|
Change LSSTCatalogSearch 's behavior to the same as IRSA's catalog search
|
When the search result in IRSA catalog search is empty, the IRSA catalog search shows the table with empty row data. The LSSTCatalogSearch returns the message with "no data found". I think that the behaviors in two catalog searches should be consistent. Since the meta server is not available, the data definitions are obtained from the search result. When there no data is available, there is no data type definition either. The current implementation threw an exception when the empty table is returned from the PDA search. But for multi-object search, the exception will stop further searches for the other object. For now, it is modified as below: # Let getDataFromUR return null if the result is empty # When the PDA search returns empty, the exception in LSSTCatalogSearch is deferred at the loadFile # In LSSTMutliObjectSearch, when one of the callable returns empty result, it will be skipped. Thus, all targets in the input file will be searched. NOTE: When the meta server is working, the above implementation will be changed accordingly. So is the meta definition in LSSTQuery.
| 1 |
2,056 |
DM-9965
|
03/27/2017 12:39:33
|
Fix bug in monotonicity operator
|
There appears to be a bug in the monotonicity operator that causes some edge pixels to wrap monotonicity around the edge to the next (or previous) line. The cause of this needs to be found and fixed.
| 1 |
2,057 |
DM-9966
|
03/27/2017 12:49:55
|
Recommend that TODO comments link to Jira
|
Following discussion on RFC-307, this ticket adds the following rule to the Python style guide: h3. To-do comments SHOULD include a Jira issue key If the commented code is a workaround for a known issue, this rule makes it easier to find and remove the workaround once the issue has been resolved. If the commented code itself is the problem, this rule ensures the issue will be reported on Jira, making it more likely to be fixed in a timely manner. {code} # TODO: DM-12345 is triggered by this line {code} {code} # TODO: workaround for DM-6789 {code} ---- and the following rule to the C++ style guide: h4. 6-26. To-do comments SHOULD include a Jira issue key If the commented code is a workaround for a known issue, this rule makes it easier to find and remove the workaround once the issue has been resolved. If the commented code itself is the problem, this rule ensures the issue will be reported on Jira, making it more likely to be fixed in a timely manner. {code} // TODO: DM-12345 is triggered by this line {code} {code} // TODO: workaround for DM-6789 {code}
| 1 |
2,058 |
DM-9967
|
03/27/2017 13:16:05
|
ctrl_pool should not accept a default for --time on real batch systems
|
ctrl_pool's {{--time}} is easily confused with the inherited {{-t}} from CmdLineTask, which makes it easy to use the wrong one and get your batch jobs killed too early. We should always be forced to provide a time estimate when submitting to a real batch system.
| 1 |
2,059 |
DM-9968
|
03/27/2017 14:25:25
|
Add eupspkg build for igprof
|
The profiler we recommend using in the developer guide isn't provided as part of the stack, and isn't entirely trivial to build on most operating systems (though it's not hard). This ticket will provide eupspkg builds for igprof and its dependencies. Hopefully that will make it easy to also make it eups-distrib-installable in the near future, though it shouldn't ever be a dependency of any stack packages (or metapackages).
| 1 |
2,060 |
DM-9970
|
03/27/2017 14:56:10
|
Fix Composite Policy Docs
|
in this block: {quote} 2.16.9 Composite Policy To indicate that a dataset should be serialized/deserialized from components, the policy’s dataset definition has a keyword composite. The structure is: {code} <dataset type name>: { composite: { <component name>: { datasetType: <dataset type> setter: <method name of setter> getter: <method name of getter> assembler: <importable function to do custom deserialization> disassembler: <importable function to do custom serialization> subset: bool inputOnly: bool } ... } {code} {quote} assembler & disassembler are in the wrong spot, the need to be below ‘composite’ not ‘componentName’
| 1 |
2,061 |
DM-9975
|
03/27/2017 17:46:20
|
Port LDM-151 to new latex style
|
LDM-151 is currently using a default Latex style. It needs to be ported to use the standard look and feel {{lsstdoc.cls}}.
| 0.5 |
2,062 |
DM-9989
|
03/28/2017 18:46:17
|
Incorporate ZOGY and A&L decorrelation option into imageDifference.py
|
Incorporate ZOGY and A&L decorrelation option into imageDifference.py. This is specifically regarding the spatially-varying variants of each, which rely upon the ImageMapReduce tasks. This epic includes fixing various bugs discovered after incorporation into imageDifference.py and running them on real images.
| 40 |
2,063 |
DM-9991
|
03/29/2017 06:50:26
|
Provide functional obs_ctio0m9
|
Building on the work done by [~aguyonnet] in DM-9363, provide an obs_ package that can be used with data from the 0.9m at CTIO.
| 2 |
2,064 |
DM-9992
|
03/29/2017 06:54:29
|
Provide CTIO 0.9m data in NCSA
|
Please transfer appropriate CTIO 0.9m datasets to NCSA in order to facilitate the construction and test of obs_ctio0m9.
| 1 |
2,065 |
DM-9993
|
03/29/2017 06:59:58
|
Prepare presentation on crowded field processing
|
Audience is the PST-SC Chairs telecon scheduled for April 18th. Should be able to mostly reuse slides from last P&CW, with some updates from LDM-151.
| 2 |
2,066 |
DM-9999
|
03/29/2017 09:48:45
|
Review literature on CModel
|
Where "literature" likely means the relevant section of the HSC pipeline paper, provided by [~jbosch].
| 2 |
2,067 |
DM-10008
|
03/29/2017 15:04:38
|
MapBox.maxOutCoord not set to nout if specified as 0 during construction
|
MapBox claims that maxOutCoord will be set to nout if specified as 0 in the constructor, but that does not happen. Fix it and add a unit test. Also consider making some or all fields const. (Making the computed vectors const is almost certainly more work than it is worth).
| 0.5 |
2,068 |
DM-10009
|
03/29/2017 16:18:25
|
Ensure masks are valid from ImageMapReduceTask
|
Ensure that the reducer subtask correctly sets the mask to invalid for pixels where the reduced exposure is infinite or NaN.
| 2 |
2,069 |
DM-10023
|
03/30/2017 13:45:07
|
Build prototype alert filtering system
|
This is to do the research and coding necessary to produce a prototype alert filtering mechanism (simple broker) for the LSST.
| 40 |
2,070 |
DM-10035
|
03/30/2017 16:52:33
|
Rotate class needs some minor refactoring
|
While working on DM-8854, the unit test for Rotate class, I noticed that it needs some minor refactoring. After investigating how this class is used, I found that the class just needs some document. I added the algorithm and brief description what it is doing. This ticket can be closed.
| 1 |
2,071 |
DM-10039
|
03/31/2017 08:19:37
|
Incorrect docs for CatalogCalculationConfig
|
meas_base's [{{CatalogCalculationConfig}}|https://github.com/lsst/meas_base/blob/0b6c81b6fe40f3d0e39237a9a83394b219cce41a/python/lsst/meas/base/catalogCalculation.py#L104] claims that it is: {quote} Default CatalogCalculationConfig. Currently this is an empty list, meaning that there are no default plugins run. The config object for each plugin must use this variable to specify the names of all plugins to be run {quote} This is wrong in a number of ways — it's not a list, it's not empty, and it doesn't make sense for each plugin to have to specify the names of all other plugins in use. Please fix it.
| 1 |
2,072 |
DM-10042
|
03/31/2017 09:00:06
|
Update mariadb statistics on 35TB dataset
|
Join query has a very strange query execution plan on 35TB dataset. Statistics have to be updated in order to fix that.
| 5 |
2,073 |
DM-10046
|
03/31/2017 11:10:48
|
Remove deprecated upload URL from firefly_client
|
In spring 2017, a change was made to the URL for uploading data to a Firefly server. The Python API handles this change in DM-9843 by checking the old and the new URL. After some months, the old URL can be considered deprecated and can be removed from the upload functions in {{firefly_client}}.
| 2 |
2,074 |
DM-10047
|
03/31/2017 11:30:01
|
Add LSST refs to LDM-151
|
LDM-151 doesn't cite LSST documents so they don't appear in references. They need to be added.
| 0.5 |
2,075 |
DM-10051
|
03/31/2017 12:37:51
|
Add core documents to shared bibliography
|
The lsstdoc.cls file defines standard macros for referring to core documents but not all of these are found in the shared bibliography in lsst-texmf. Add these missing documents. Also redefine the {{\ds}} macro to use a citation rather than a direct docushare link.
| 0.5 |
2,076 |
DM-10066
|
03/31/2017 16:15:26
|
Provide utility function for wrapping operator<<
|
We frequently want to implement a class's {{\_\_str\_\_}} or {{\_\_repr\_\_}} methods in terms of its {{operator<<}}. The current approach has led to a fair amount of code duplication throughout the stack. Once a function template that takes a printable object and returns a string is available, explicit wrappers for {{operator<<}} should be rewritten to use it.
| 3 |
2,077 |
DM-10069
|
03/31/2017 18:18:31
|
Remove boost_thread Qserv dependency
|
The sole remaining boost_thread dependency in Qserv seems to be some thread-local storage used by the mysql qserv module. Since the baseline minimal compiler is gcc 4.8, we can use C++11 thread_local instead.
| 0.5 |
2,078 |
DM-10089
|
04/03/2017 16:36:38
|
Simplify validate_base Blob into a concrete collection type
|
validate_base's BlobBase type is currently intended to be an abstract base class that apps create subclasses for. The original intention was that blob data would be computed within the Blob subclass's code. With the new validate_base we're going towards a composition model with container types. This ticket is to convert BlobBase into a Blob class, and permit applications to store data into the Blob instance through standard mapping idioms.
| 1 |
2,079 |
DM-10091
|
04/03/2017 20:31:02
|
Fix problems left over from DM-9952
|
DM-9952 is merged to master, but I typoed the Jenkins run {{DM-9952}} not {{tickets/DM-9952}} and didn't notice failing tests from afwdata; afwdata is not present on tiger in Princeton where I did my local testing. This ticket is to capture the errors turned up when I *did* run the tests.
| 1 |
2,080 |
DM-10093
|
04/04/2017 09:54:04
|
Revert disabling of meas_modelfit dependency in lsst_apps
|
In DM-8467, the meas_modelfit dependency of lsst_apps was temporarily removed: {code} commit 041ae975d9c60a41b24488ec0ba1768431330fd8 Author: Pim Schellart <[email protected]> Date: Tue Feb 14 15:51:46 2017 -0500 Disable meas_modelfit until pybind11 wrapped diff --git a/ups/lsst_apps.table b/ups/lsst_apps.table index dce8fd4..b239e74 100644 --- a/ups/lsst_apps.table +++ b/ups/lsst_apps.table @@ -1,5 +1,6 @@ setupRequired(meas_deblender) -setupRequired(meas_modelfit) +# temporarily disable meas_modelfit until pybind11 wrapped +#setupRequired(meas_modelfit) setupRequired(pipe_tasks) setupRequired(obs_lsstSim) setupRequired(obs_sdss) {code} meas_modelfit is now pybind wrapped (DM-8465), so this should be reverted.
| 1 |
2,081 |
DM-10094
|
04/04/2017 10:19:04
|
Copy CTIO 0.9m data to /datasets/ctio0m9/raw
|
See RFC-313 and DM-9992
| 1 |
2,082 |
DM-10096
|
04/04/2017 11:48:10
|
Add unit test asserts for SpherePoint, SpherePointList and PointList
|
Add unit test asserts to lsst.afw.geom.utils: - assertSpherePointsAlmosttEqual - assertSpherePointListsAlmostEqual - assertPointListsAlmostEqual and unit tests for them. This work is needed (or at least very useful) for the new WCS classes.
| 2 |
2,083 |
DM-10101
|
04/04/2017 19:04:56
|
bad exception handling in afw for python3
|
Looks like a missed 2to3 compatibility check: {code} *** error building product afw. *** exit code = 2 *** log is in /Users/josh/src/lsstsw3/build/afw/_build.log *** last few lines: ::::: [2017-04-05T00:57:00.101207Z] import lsst.afw.cameraGeom.utils as cameraGeomUtils ::::: [2017-04-05T00:57:00.101210Z] File "/Users/josh/src/lsstsw3/build/afw/python/lsst/afw/cameraGeom/utils.py", line 397 ::::: [2017-04-05T00:57:00.101214Z] except Exception, e: {code}
| 1 |
2,084 |
DM-10102
|
04/05/2017 07:05:15
|
Reset lsst-dev01:/ssd shared stack
|
Implement RFC-317 by wiping out the shared stack in {{lsst-dev01:/ssd/lsstsw/stack}} (or, perhaps, moving it aside) and creating a new stack in its place. Take precautions to make sure that the toolchain used to build the new stack is always consistent. Consider bumping to newer versions of EUPS, Anaconda etc than come by default in the {{shared-stack.py}} script. Also install AstroPy through {{conda}}.
| 1 |
2,085 |
DM-10103
|
04/05/2017 07:08:20
|
Investigate segfaults in lsst-dev01:/ssd/lsstsw/stack shared stack
|
The {{w_2017_14}} weekly fails to build as part of the shared stack in {{lsst-dev01:/ssd/lsstsw/stack}}. This may be related to segfaults previously reported by [~tmorton].
| 1 |
2,086 |
DM-10105
|
04/05/2017 07:19:31
|
Inconsistency in meas/forced wcs leads to CModel failure
|
From Sogo Mineo: {code} During processing a dataset with multiBandDriver.py in hscPipe 5.0-beta1, I ran into an error: Traceback (most recent call last): File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/ctrl_pool/13.0-1-g0a48345/python/lsst/ctrl/pool/pool.py", line 113, in wrapper return func(*args, **kwargs) File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/ctrl_pool/13.0-1-g0a48345/python/lsst/ctrl/pool/pool.py", line 1067, in run while not menu[command](): File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/ctrl_pool/13.0-1-g0a48345/python/lsst/ctrl/pool/pool.py", line 237, in wrapper return func(*args, **kwargs) File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/ctrl_pool/13.0-1-g0a48345/python/lsst/ctrl/pool/pool.py", line 1085, in reduce result = self._processQueue(context, func, [(index, data)], *args, **kwargs)[0] File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/ctrl_pool/13.0-1-g0a48345/python/lsst/ctrl/pool/pool.py", line 544, in _processQueue return self._reduceQueue(context, None, func, queue, *args, **kwargs) File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/ctrl_pool/13.0-1-g0a48345/python/lsst/ctrl/pool/pool.py", line 570, in _reduceQueue resultList = [func(self._getCache(context, i), data, *args, **kwargs) for i, data in queue] File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/pipe_drivers/13.0-1-g07ace30+4/python/lsst/pipe/drivers/multiBandDriver.py", line 388, in runForcedPhot self.forcedPhotCoadd.run(dataRef) File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/meas_base/13.0-3-gc489882+1/python/lsst/meas/base/forcedPhotImage.py", line 147, in run self.measurement.run(measCat, exposure, refCat, refWcs, exposureId=self.getExposureId(dataRef)) File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/meas_base/13.0-3-gc489882+1/python/lsst/meas/base/forcedMeasurement.py", line 330, in run beginOrder=beginOrder, endOrder=endOrder) File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/meas_base/13.0-3-gc489882+1/python/lsst/meas/base/baseMeasurement.py", line 268, in callMeasure self.doMeasurement(plugin, measRecord, *args, **kwds) File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/meas_base/13.0-3-gc489882+1/python/lsst/meas/base/baseMeasurement.py", line 287, in doMeasurement plugin.measure(measRecord, *args, **kwds) File "/data/mineo/opt/hscpipe/5/stack_5.0-beta1/Linux64/meas_modelfit/13.0-3-g177c5a2+1/python/lsst/meas/modelfit/cmodel.py", line 98, in measure "CModel forced measurement currently requires the measurement image to have the same" FatalAlgorithmError: CModel forced measurement currently requires the measurement image to have the same Wcs as the reference catalog (this is a temporary limitation). application called MPI_Abort(MPI_COMM_WORLD, 1) - process 10 The exception was thrown from meas_modelfit/13.0-3-g177c5a2+1/python/lsst/meas/modelfit/cmodel.py: def measure(self, measRecord, exposure, refRecord, refWcs): if refWcs != exposure.getWcs(): ### Condition (A) ### raise lsst.meas.base.FatalAlgorithmError( "CModel forced measurement currently requires the measurement image to have the same" " Wcs as the reference catalog (this is a temporary limitation)." ) `refWcs` and `exposure` in Condition (A) was from meas_base/13.0-3-gc489882+1/python/lsst/meas/base/forcedPhotImage.py:139 refWcs = self.references.getWcs(dataRef) exposure = self.getExposure(dataRef) `self.references.getWcs` was probably defined in meas_base/13.0-3-gc489882+1/python/lsst/meas/base/references.py:202: def getWcs(self, dataRef): """Return the WCS for reference sources. The given dataRef must include the tract in its dataId. """ skyMap = dataRef.get(self.config.coaddName + "Coadd_skyMap", immediate=True) return skyMap[dataRef.dataId["tract"]].getWcs() Therefore, in Condition (A), one of the compared WCS was from skyMap, and the other was from a FITS file. I suspected that WCS had been changed when saved in the FITS file, due to limited number of digits written in FITS headers. To prove it, I wrote the following script. It actually proved that there are some tracts where Condition (A) is violated. ====================== import lsst.afw.image as afwImage import cPickle as pickle def main(): skyMap = pickle.load(open("skyMap.pickle", "rb")) for tract, tractInfo in enumerate(skyMap): testWcs(tract, tractInfo.getWcs()) def testWcs(tract, wcs): exp = afwImage.ExposureF(0, 0) exp.setWcs(wcs) exp.writeFits("/dev/shm/foo.fits") exp = afwImage.ExposureF("/dev/shm/foo.fits") # This message must always be "{tract}, True" print tract, exp.getWcs() == wcs main() {code}
| 1 |
2,087 |
DM-10111
|
04/05/2017 11:19:11
|
Add overscan exclusion regions as an ISR config parameter
|
As per the discussion in RFC-315, two config parameters need to be added to overscanCorrection() to allow excluding the first and last n rows. Both parameters should default to 0, so that the default configuration reproduces current functionality.
| 1 |
2,088 |
DM-10112
|
04/05/2017 11:37:15
|
Visualization algorithm research (f17)
|
This epic captures the algorithm related research for visualization.
| 20 |
2,089 |
DM-10118
|
04/05/2017 18:56:57
|
SQuaRE SuperTask Collaboration Week of 2017-04-03
|
Bucket ticket to cover non-specific SuperTask collaboration activities.
| 1 |
2,090 |
DM-10132
|
04/06/2017 16:31:20
|
Make DCR template unit test data robust
|
The current unit tests for the DCR template generation prototype uses `numpy.save()` and `numpy.load()` to persist and depersist test data. This format is fragile and has already been broken once with the conversion from swig to pybind11, so a more robust test data format should be used.
| 2 |
2,091 |
DM-10138
|
04/07/2017 09:54:17
|
Compare the convergence of ADMM, SDMM, and GLM algorithms
|
The convergence of our new algorithms, SDMM and GLMM are unknown as of now, so we should study the convergence rates of both algorithms (and compare them to the ordinary ADMM) to understand convergence in all three algorithms.
| 3 |
2,092 |
DM-10142
|
04/07/2017 11:17:23
|
Change HDU used in obs_ctio0m9 from 1 to 0
|
obs_ctio0m9 looks in the wrong HDU for data. Config parameter just needs to be changed from 1 to 0.
| 1 |
2,093 |
DM-10144
|
04/07/2017 12:54:41
|
Document lsst-texmf
|
People are starting to use lsst-texmf for document writing but there is some confusion. This ticket is to write a document explaining how to use the lsst latex classes. The consensus is that this document should match the LSST "software guide" standard and be written using sphinx in a subdirectory of lsst-texmf (rather than writing it in Latex). The documentation will be published to https://lsst-texmf.lsst.io and will include links to the rendered example PDFs. This ticket covers the writing of the guide. Publishing will be handled separately.
| 2 |
2,094 |
DM-10146
|
04/07/2017 16:09:07
|
Fix minor doc typos
|
Object.h and WcsMap.h both refer to {{astshim::}} when they should refer to {{ast::}}. I'll fix that for now, but I am becoming increasingly aware that {{astshim::}} would be less confusing. Still...changing that would be another (simple) ticket.
| 0 |
2,095 |
DM-10149
|
04/07/2017 17:54:10
|
The range of the phase value and period value for periodogram in time series are not correct
|
The range of phase value should be [0, 2) plotly based phase folding chart doesn't react properly while an invalid period value is selected from the periodogram table. The phase folded chart should show empty content when an invalid period value is selected.
| 2 |
2,096 |
DM-10151
|
04/07/2017 20:43:15
|
Investigate a better symmetry operator
|
The current symmetry operator work by comparing each pixel to its symmetric partner and projecting the difference to zero using a proximal operator, forcing the solution to be symmetric. This works relatively well, but not as well as the current deblender in certain cases. For example, in a blend with only two sources, with one source much brighter than the other, the faint source using the NMF deblender attempts to steal a small amount of flux from its brighter neighbor. We had hoped that the near zero flux opposite the bright source would be enough to limit this effect (when combined with symmetry) but the algorithm breaks when the fainter object wants to steal flux near the same level as the noise. Sparsity is not much help, as l0 sparsity limits the size of brighter objects more than the smaller ones. This blending issue is accounted for in the current deblender (adapted from SDSS) because it doesn't just force a symmetric solution, it forces the symmetric solution to use the minimum of each pixel and its symmetric partner. This is a much stronger constraint (and more useful) constraint. It would be useful to update the NMF deblender to have the same minimum pixel value constraint. One way to do this is to use a proximal operator that projects each template onto a space that uses the minimum of each symmetric pair. This will no longer be a linear operation, as choosing the minimum pixel value is non-linear, so it is unclear how this will affect convergence. If we can make this work, combining the new symmetry operator with monotonicity might eliminate the need to use a sparsity constraint, which is advantageous as [~pmelchior] and I are realizing that sparsity is a wilder beast than it first appeared to be, and taming it is not a simple task. This is because the sparsity of an image is dependent on the total flux in the image, and the brightness of each source, so it might be that using sparsity would require each source in a blend to have a different sparsity requirement dependent on it's brightness, which could be a difficult scheme to implement. Taking care of this directly using a better symmetry constraint is preferable.
| 3 |
2,097 |
DM-10152
|
04/08/2017 12:16:19
|
Fix bug in Box Python stringification
|
The implementation of {{Box2I.\_\_repr\_\_}} in the pybind11 wrappers incorrectly labels it as a {{Box2D}} which is very confusing.
| 0.5 |
2,098 |
DM-10161
|
04/10/2017 11:06:35
|
Remove maxtasksperchild=1 during pool initialization in meas_mosaic
|
Testing for DM-10043 has revealed that having {{maxtasksperchild=1}} in the pool initialization for {{mosaic.py}} (the pool being used to read input catalogs in parallel) causes a bug where {{mosaic.py}} hangs indefinitely. Curiously, this happens only in with the LSST stack (pybind11 version), not with the HSC stack. It has been demonstrated that removing this constraint removes this hanging behavior.
| 1 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.