These draft notes cover the major updates made by Data Release Production to the LSST stack since release 12 (Winter/Extra 2016). Please add material here which should ultimately be incorporated into the notes accompanying the Fall 2016 release.
The KronFluxAlgorithm measures the flux in an elliptical aperture as defined by Kron (1980). The plugin is not enabled by default, but may be switched on by requesting ext_photometryKron_KronFlux in measurement configuration.
A package version tracking and checking capability has been included in the stack and integrated with the ``CmdLineTask`` system. This keeps track of what versions of packages are being used in order to prevent the user from unwittingly processing data with multiple, subtly different, versions of the code. This provides a provisional implementation of a provenance tracking
system: it is essential for short-term use, but will be rendered obsolete before we move to operations. A more detailed description of this system is available on the LSST Community Forum.
Erin Sheldon's ngmix has been integrated with the LSST codebase: it can be installed as package ngmix, and meas_extensions_ngmix package provides an interface between it and the LSST measurement framework. At time of writing, these packages are purely optional, and have not been included in any of the standard Science Pipelines top-level packages (lsst_apps, lsst_distrib). Currently, we provide an ngmix-based plugins for approximating the PSF and for fitting sources using a mixture of Gaussians in both single frame and forced measurement.
Metadata describing the operations of algorithms within the multiband coadd processing framework is now correctly persisted along with their outputs.
The meas_mosaic package for performing simultaneous astrometric calibration across multiple overlapping exposures has been ported from Hyper Suprime-Cam. While this is an essential capability for LSST, we do not expect meas_mosaic itself to survive to operations: it is slated for replacement by Jointcal. However, meas_mosaic provides an essential point of comparison for Jointcal, and is a key capability until Jointcal is fully available.
Experiments with Hyper Suprime-Cam have established that oversampling does not work well in PSFEx. It has therefore been disabled in the default configuration used for LSST.
This filter is now defined in the obs_subaru camera package.
The algorithm used to select the reference band in multi-band coadd processing has been updated to avoid bands with certain user-specified flags set. By default, these flags include the source having an interpolated centre, or errors detected with PSF, Kron or CModel fluxes.
CatalogCalculation
The "afterburner" system the stack in the previous (v12) release; it provides a capability to run algorithms akin to our regular measurements, but which only take inputs from catalogs (i.e. not pixel data). To avoid ambiguity, this system has been renamed CatalogCalculation
.
Resolved a bug whereby comparison of (intrinsically unordered) configuration dictionaries incorrectly compared their ordering rather than just their contents. This lead to unexpected failure of re-runs due to incorrectly identified “mismatched configurations”.
Units for the time field changed from seconds to second to comply with the Astropy convention.
Our SourceDetectionTask
provides the optional capability of fitting and temporarily removing a local background before detecting sources. This capability was accidentally disabled in the X/W16 release (v12); it has now been restored. It is disabled by default.
A memory handling error could cause our convolution code to generate incorrect results on some platforms (problems observed on OS X when the code is built without optimization; no errors seen on Linux, or when building with -O2
or -O3
on OS X). This appears to originate in the Boost Generic Image Library, an external dependency of the LSST stack. We have been unable to isolate the problem in Boost, but have worked around it for our particular use case.
On certain error conditions, GalSim, a third party package used by the LSST stack, generates an assertion failure which is hard for our code to handle appropriately. LSST's version of GalSim has been patched to raise a more appropriate exception instead. This change has been submitted for inclusion in GalSim itself.
Resolved an issue whereby the star selection could change depending on the number of CPU cores in use.
Resolved an issue whereby the SingleFrameDriverTask
failed to use more than one CPU core regardless of configuration.
A typo which rendered CoaddSrcTransformTask inoperable was resolved. In the process, the tests for this task were significantly upgraded to avoid a repeat of this, or other, problems.
AR_NAME
headers when persisting dataWhen persisting objects to FITS, we were incorrectly writing duplicate AR_NAME
headers. This has been resolved.
We now correctly write TUNIT
headers describing the units used when storing angles, and are more careful about checking units when reading them.
A bug has been discovered in the afw table library which could lead to infinite recursion. While the ultimate cause of the bug is still under investigation, a temporary workaround has been put in place to mitigate its impact.
Continuing work started in the previous release, we have audited the stack for use of the Boost libraries. Where possible, Boost has been replaced with C++11 standard library equivalents.
Previously, non-fatal NumPy warnings of the form RuntimeWarning: invalid value encountered in less could appear on the console, but not be caught and logged anywhere. It was therefore easy to miss problems arising when running the algorithm. These warnings are now ingested into the LSST logging system, and are hence recorded properly.
A set of unit tests have been written which properly exercise the colorterm framework which was adapted from Hyper Suprime-Cam in the S15 cycle.
The FlagHandler system for efficiently and safely setting flags corresponding to measurement algorithm failures on source records is now accessible from Python. Previously, it was only available for use in C++ code.
The assertImagesEqual
and assertMaskedImagesEqual
methods have been added to added to LSST's test cases. These provide explicit mechanisms for checking for exact equality of image data.
Added a test to the ci_hsc package to monitor the fraction of sources used to model the PSF in when processing a CCD image and ensure that the vast majority of them are ultimately classified as stars. This provides an automated check that we are not inadvertantly estimating the PSF based on extended objects.
The CatalogCalculation facility was added to the stack in the previous (v12) release (where it was referred to as "afterburners"); it provides a capability to run algorithms akin to our regular measurements, but which only take inputs from catalogs (i.e. not pixel data). This has now been used to clean up the logic used to apply aperture corrections within our regular data processing flow. This old method for handling aperture corrections was a persistent source of bugs and confusion: these should be avoided by the new approach.
Following an RFC, sconsUtils was updated to enable optimization by default.
Added extra logic to ensure we target a version of the OS X SDK which is compatible both with the version of Mac OS installed and the Python interpreter in use.
Butler dataset definitions which are common to multiple cameras have been consolidated in a central location in daf_butlerUtils rather than being repeated in each camera package.
The sconsUtils build system now reads SConscript
files, which describe the build, in a user configurable order. This makes it possible to specify that some components of a package must be built before others.