Changelog
1.3.1 (Unreleased)
Fix occasional all-NaN issue in
single_tile_destripeFix backgrounds being wrongly moved in
lv1_stepMoved
astrometric_alignto a combined step, to avoid list ordering issuesUpdated dependencies, added specific pins, dependabot, and CODEOWNERS
1.3.0 (2025-10-02)
Fixed crash for
astrometric_catalogfor background observationsFixed crash when filtering out guide stars
Fixed issue with background files in
move_raw_obs_stepFixed logging, will now properly output to a log file
Fix crash for background observations in
get_short_band_nameFor background association, strip whitespace and lowercase for more robust matching
Imports update - now supports numpy>2.0!
Keep track of expected stripe direction in
multi_tile_destripe_stepTrim off reference pixels for median filter in
single_tile_destripe_stepKeep track of expected stripe direction in
single_tile_destripe_stepAdd support for NIRISS
Fix crash in the persistence correction in
lv1_stepif some earlier observation is missingFix crash in
multi_tile_destripe_stepfor heavily masked data when doing smoothingFix crash in
astrometric_align_stepwith new gwcs updatesFix crash in
lv3step when degrouping NIRCam modulesAllow for degrouping of NIRCam modules in
get_wcs_adjust_stepandapply_wcs_adjust_stepFix for crash if trying to anchor without first aligning
Fix for crash if trying to PSF match without some combination of anchoring, aligning
Change how the level 3 step handles models for newer JWST versions
1.2.0 (2024-09-03)
Removed raised warning in
multi_tile_destripe_stepwhen input images have a variety of rotationsAdded
combine_nircam_shortoption inlevel_match_step, which will match levels between the four NIRCam short imaging chips before doing matching between mosaic tileslv2_stepwill now propagate through individual exposure offset times from backgrounds, which is necessary for selecting a reference image inlevel_match_stepAdded
fit_typeoption tolevel_match_stepfor every option, which allows for plane fitting in the level matching in a modular waylevel_match_stepfitting method has been changed to resemble the iterative Montage method, which is necessary for plane fittingreproject_imageinutilscan now also reproject error arraysmake_stacked_imageinutilscan now also reproject the error and readnoise mapsExposed
auto_rotateinmake_stacked_imageinutilsAdded prefilter option to
get_wcs_adjust, which uses constrained diffusion to remove large-scale structureAdded
recombine_lyotoption tolevel_match_step, which allows for recombining the lyot coronagraph into the main MIRI science chip before matching between mosaic tilesTidied up
level_match_stepto make more modular and functionalFix bug in
astrometric_align_stepif first pass of alignment succeeds but second failsAdd cross-correlation option to
get_wcs_adjust_step, which usesspacepylotAdd control over reproject functions, can be
reproject_interp(default),reproject_exact, orreproject_adaptive. For MIRI,reproject_exactmay work better. This applies toanchoring_step,astrometric_align_step,level_match_step,multi_tile_destripe_step,psf_matching_step, andrelease_stepFix bug for lv1 hanging on newer Macbooks
Fix bug for parameters with ‘pix’ in getting picked up like numbers of pixels
Updated PHANGS Cy3 config
Include links to Francesco Belfiore’s kernel generation repository in the docs
Updated various package requirements
Updated indexing to align with new
reprojectupdates
1.1.0 (2024-03-04)
If science lv1 data already exists, just copy it over for background data to save processing time
Include BMAJ in
psf_matching_stepAuto rotate the WCS in
level_match_stepFix SIP approximation issues in
apply_wcs_adjust_stepastrometric_catalog_stepcan now use either DAOStarFinder or IRAFStarFinder. Observatory recommendations seemto be IRAFStarFinder in general
Removed factor of 2 for FWHM in lv3 tweakreg-related steps to more closely match observatory recommendations
Add lower version pins to project requirements to avoid incompatibilities
Fixed bug in
astrometric_align_stepwhere file might already be closed before savingUpdated PHANGS Cycle 1 config to v1p1 (last couple of observations, and a few small changes to the config)
Fix bug with
process_bgr_like_scienceinlv3_stepAdded ‘filename’ option for background association
Fix in
parse_fits_to_tablewhen there is not a defined observation labelPoint Zenodo DOI to “concept” DOI, rather than latest
Cleaner labels in
anchoring_stepAdd acknowledgment for PCA (thanks Liz!)
Added CITATION.cff
1.0.4 (2024-01-04)
Fix del in
get_wcs_adjustif we skip alignmentMade plots more consistent and “publication ready” across the board
Fix bug in
anchoring stepRename
anchor_to_external_step.pytoanchor_step.pyfor consistencyFix “too many plots open” warning in
anchoring_stepInclude useful pjpipe version info in files
Save anchoring background in metadata properly in
anchoring_stepAdd Zenodo badge
1.0.2 (2023-12-01)
Fix packaging to minimise file size
Error in
pyproject.toml
1.0.0 (2023-12-01)
Added documentation
Ensure
multi_tile_destripe_stepis properly imported in the pipelineRemove an error in
multi_tile_destripe_stepstripes in stacked image don’t align with any particular axis of the arrayFix crash where quadrants are turned off in median filtering in
single_tile_destripe_stepFix potential memory leaks in
get_wcs_adjust_stepandapply_wcs_adjust_stepDiagnostic plots are compressed by default in
release_step
0.10.1 (2023-11-20)
Can specify CRDS context at the pipeline level
median_filter is now the default in
single_tile_destripe_stepAdded fallback for median_filter in
single_tile_destripe_stepwhen too much data is maskedRemove mask option in median_filter in
single_tile_destripe_step, since it’s always used anywayChanged up how
do_large_scaleworks inmulti_tile_destripe_step, which seems significantly improved and simplifiedAdded fallback for in
multi_tile_destripe_stepwhen too much data is masked in quadrantsChanged how tweakreg grouping is done in
get_wcs_adjust_stepandlv3_stepto account for code changes in the pipelineAdded option to decouple the short NIRCam chips for tweakreg in
lv3_stepAdded option to move the various diagnostic plots in
release_stepremove_bloatis now False by default inrelease step, to maintain datamodels compatibilityremove_bloatalso applies to PSF matched files inrelease_step
0.10.0 (2023-11-14)
Added PSF modelling routines. These are currently very preliminary, but at least exist
Allow for external, absolute catalog in
get_wcs_adjust_stepAdd PSF matching routines (
psf_matching)Add anchoring routines (
anchoring)Include useful outputs from these in the
release_stepIf not grouping dithers in
get_wcs_adjust_step, respect that in how the transforms are written outsingle_tile_destripe_stepcan now run on rate files (pre-flat fielding)Bugfixing in
single_tile_destripe_stepDecoupled horizontal/vertical destriping methods in
single_tile_destripe_step, since the noise properties are distinct in these two axesAdd
smoothoption tosingle_tile_destripe_step, based on Dan Coe’s smooth1overf algorithmAdded control over how values are extended beyond array edge for filtering in
single_tile_destripe_stepLots of bugfixing in
multi_tile_destripe_stepmulti_tile_destripe_stepcan un-flat before correctingLevel between amplifiers in
multi_tile_destripe_stepMake vertical stripe subtraction optional in
multi_tile_destripe_stepAdded median option to
multi_tile_destripe_stepto do a median rather than mean imageAdded iterative option to
multi_tile_destripe_stepthat will keep things going until sigma-based convergenceAdded sigma-clip median option for creating stacked images
do_large_scalenow works completely differently inmulti_tile_destripe_step, instead attempting to clean up the average imageAdded support for different
do_large_scalemethods inmulti_tile_destripe_step, which may work better in certain situationsAdded a median filter
do_large_scalemethod, which may be more robust than the boxcar. THIS IS NOW THE DEFAULTAdded a sigma-clipped
do_large_scalemethod, as should be optimal in observations that aren’t full of emissionAdded a smooth convolution
do_large_scalemethod, based on Dan Coe’s smooth1overf algorithmAdded control over how values are extended beyond array edge for
do_large_scaleAdded option in
lv3_stepto degroup dithers for tweakregAdded option in
lv3_stepto degroup NIRCam modules, since the WCS is currently inconsistent between the twoFixed crash in
lv3_stepif one of the group/degroup parameters is not definedBe smarter about keeping track of exposure numbers in
lv3_stepregress_against_previouswill now search for files in priority order, for fallback between versionsMake sure backgrounds are included in label for
regress_against_previousf-string fixes
0.9.2 (2023-09-18)
Allow multiple options for e.g. proposal ID in
download_stepFix potential error with file validation in
download_stepmove_raw_obs_stepis smarter about missing filterslv1_stepis smarter about grouping dithersFix plotting error if quadrants=False and using median filter in
single_tile_destripe_stepFix potential subarray issues with
lyot_separate_step/lyot_mask_stepdo_large_scaledefaults to False inmulti_tile_destripe_stepMuch improved diagnostic plots in
multi_tile_destripe_stepCatch errors in
level_match_stepwhere all data might be NaNRename
do_vertical_subtractionin config filesCaught some typos from lazy copy/pasting docstrings
Additions and updates for #2130 (Local Group) and #3707 (Cy2 Treasury)
0.9.1 (2023-09-04)
If not supplied, will default to running on all CPUs, not 1
Include option for producing background images
Include Gaia query for astrometric catalogs (
gaia_query_step)Include option to produce mosaics for each individual field (
mosaic_individual_fields_step)Parallelise up the download integrity verification in
download_stepCatch warnings as errors in integrity verification in
download_stepInclude array information when creating asn files to ensure we don’t associate backgrounds incorrectly
Change naming system for lyot separate to ensure compatibility with later steps
get_wcs_stepnow sorts shifts to be more human-readableAdd local background subtraction to
level_match_step, which may help for mosaics without overlapsrelease_stepnow takes the lv3 directory as an argument, rather than parsing any progress dictionariesrelease_stepwill now also move any individual field mosaics
0.9 (2023-07-25)
Modular refactor
Include subtracted backgrounds in release