Skip to content

Latest commit

 

History

History
84 lines (69 loc) · 4.41 KB

CHANGELOG.rst

File metadata and controls

84 lines (69 loc) · 4.41 KB
Raw

Changes from v2.4 to v2.5

This release adds support for Pyton 3.11 and 3.12. We currently support 3.7 through 3.12.

API Changes

  • Deprecated the use of filter W149 in roman module. New name is W146. (#1017)
  • Deprecated automatic allocation of PhotonArray arrays via "get" access rather than "set" access for the arrays that are not initially allocated. E.g. writing dxdz = photon_array.dxdz will emit a warning (and eventually this will be an error) if the angle arrays have not been either set or explicitly allocated. One should be sure to either set them (e.g. using photon_array.dxdz = [...]) or explicitly allocate them (using photon_array.allocateAngles()). (#1191)
  • Changed the .SED attribute name of ChromaticObject to lowercase .sed. (#1245)

Config Updates

  • Fixed a bug in Scattered type, where it didn't respect world_pos being specified in the stamp field, as it should. (#1190)
  • Added a new initial_image input type that lets you read in an existing image file and draw onto it. (#1237)
  • Added skip_failures option in stamp fields. (#1238)
  • Let input items depend on other input items, even if they appear later in the input field. (#1239)
  • Allow profiling output to reach the logger when running with -v0. (#1245)
  • Added Eval type for GSObjects. (#1250)

New Features

  • Updated Roman telescope data to Phase C (aka Cycle 9) specifications (#1017)
  • Added ShapeData.applyWCS method to convert HSM shapes to sky coordinates. Also added the option use_sky_coords=True to hsm.FindAdaptiveMom to apply this automatically. (#1219)
  • Added DoubleZernike class and related functionality. (#1221)
  • Added some utility functions that we have found useful in our test suite, and which other people might want to use to the installed galsim.utilities. (#1240)
  • Added utilities.merge_sorted which merges two or more sorted numpy arrays faster than the available numpy options. (#1243)
  • Added EmissionLine class to represent emission line SEDs. (#1247, #1249)
  • Updated data in roman module to Phase C (Cycle 9) information. (#1017, #1251)

Performance Improvements

  • Added support for GPU offloading. So far this has only been applied to the Silicon sensor calculations of the brighter-fatter effect. (#1212, #1217, #1218, #1222, #1224, #1230)
  • Drawing chromatic objects with photon shooting automatically adds a WavelengthSampler photon_op. It used to do this regardless of if one was already in a photon_ops list, which is inefficient. Now it only adds it if there is not already one given by the user. (#1229, #1236)
  • Work around an OMP bug that disables multiprocessing on some systems when omp_get_max_threads is called. (#1241)
  • Sped up the combine_wave_list function, using the new merge_sorted function. (#1243)
  • No longer keep a separate wave_list array in ChromaticObject. These are always equal to the wave_list in the sed attribute, so there is no need to duplicate the work of computing the wave_list. (#1245)
  • Delayed the calculation of the sed attributes of ChromaticObject until they are actually needed. Since sometimes they are not needed, this is a performance improvement in those cases. (#1245)
  • Reduce long-term memory usage of Silicon class after drawing onto a very large stamp and then moving on to smaller stamps. (#1246)

Bug Fixes

  • Fixed a bug that could lead to overflow in extremely large images. (#1017)
  • Fixed a slight error in the Moffat maxk calculation. (#1208, #1210)
  • Fixed a bug that prevented Eval types from generating lists in config files in some contexts. (#1220, #1223)
  • Fixed the absorption depth calculation in the Silicon class to allow wavelengths that are outside the given range of the absorption lookup table. It now just uses the limiting values, rather than raising an exception. (#1227)
  • Changed the SED class to correctly broadcast over waves when the SED is constant. (#1228, #1235)
  • Fixed some errors when drawing ChromaticTransformation objects with photon shooting. (#1229)
  • Fixed the flux drawn by ChromaticConvolution with photon shooting when poisson_flux=True. (#1229)
  • Fixed a slight inaccuracy in the FFT phase shifts for single-precision images. (#1231, #1234)
  • Fixed a bug that prevented a convolution of two PhaseScreenPSF objects from being drawn with photon shooting. (#1242)