Repository navigation
Expose min_pixperbeam in loop_postprocess - #411
Merged
Merged
Conversation
thomaswilliamsastro
force-pushed
the
expose-min-pixperbeam
branch
from
October 14, 2026 09:06
b6fd355 to
9a4e87b
Compare
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #411 +/- ##
======================================
Coverage 7.95% 7.95%
======================================
Files 42 42
Lines 16635 16635
Branches 3983 3983
======================================
Hits 1324 1324
Misses 15288 15288
Partials 23 23
|
Thread a typed min_pixperbeam argument (default 3, unchanged behaviour) from loop_postprocess through recipe_cleanup_one_target into task_compress, replacing the hardcoded value passed to trim_cube for both the casa and spectralcube postprocessing methods. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
thomaswilliamsastro
force-pushed
the
expose-min-pixperbeam
branch
from
October 14, 2026 21:52
9a4e87b to
abcc37a
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The user may want more control over how many pixels per beam there are in the final image. This is probably more important for point-source science, and the fact that the ALMA pipeline aims for 5, while we aim for 3. Visually, there's a significant step up in visual fidelity going from 3 to 5 pixels per beam, but we currently have no control over that in postprocessing.
Summary
Exposes the
min_pixperbeamparameter used when compressing cubes, so it can be set from theloop_postprocesscall:loop_postprocessgainsmin_pixperbeam: int = 3, passed torecipe_cleanup_one_targetin the cleanup step.recipe_cleanup_one_targetandtask_compressgain the same typed argument, replacing the hardcodedmin_pixperbeam=3passed totrim_cubefor both thecasaandspectralcubemethods.Testing
trim_cubemocked,loop_postprocess(do_cleanup=True, min_pixperbeam=5)passes5totrim_cubefor bothcasaandspectralcube, and3when omitted. Onmasterthe same call raisesunexpected keyword argument 'min_pixperbeam'.ruffreports the same 4 pre-existing warnings asmaster.🤖 Generated with Claude Code