Cosipy asi pipeline#605
Conversation
In docs/tutorials/apps add an example of an end-to-end pipeline for s point like source using all the apps. Updates the documentation apps.rst accordingly. Updates the unit test.
Documentation build overview
37 files changed ·
|
adjust parameters for example pipeline.
…_asi_pipeline # Conflicts: # cosipy/pipeline/src/fitting.py # cosipy/pipeline/src/preprocessing.py # cosipy/pipeline/task/task.py
Codecov Report❌ Patch coverage is
... and 3 files with indirect coverage changes 🚀 New features to boost your workflow:
|
…comment line in run_pipeline_zero.sh
|
The orientation file you want is now in COSI-SMEX/DC4/Data/Orientation. The name was changed to start with DC4 instead of DC3 but is otherwise the same. COSI-SMEX/DC4/Data/Orientation/DC4_final_530km_3_month_with_slew_1sbins_GalacticEarth_SAA.fits |
…sponse files. Update all the configuration files accordingly
…i_threemlfit crashes. Now If the fit fails, the error message is printed to stdout, an empty hdf is saved and no plot is produced. This serves to avoid pipeline crashes.
|
Thanks @jdbuhler. I think that solves the first of @ldigesu's points. As for the other:
@ldigesu Can you please clarify what you meant? Is it that the result don't make sense but the unit test still pass? If so, maybe it's due to the test files being very poor --i.e. to coarse. It's enough to test the mechanics, so maybe that's fine for now if you understand what's going on under the hood of the test. |
In this draft PR, I add the command-line app cosi-tsdetect to the available pipeline tasks. This automates the TS map fitting and extracts the coordinates at the maximum TS value.
To demonstrate the usage I added in docs/tutorials/apps the file run_pipeline_zero.sh, which used all the available apps for an end-to-end analysis pipeline of a point-like source with known tstart-tstop. This is what is currently being used by Francesco Fenu in the prototype of the comprehensive pipeline in cosiflow. I updated the documentation accordingly apps.rst accordingly with all the explanations. I also updated the unit test for the app. I think this PR is of interest for people working in the pipeline @NicoloParmiggiani @falric05 @ckierans
I still have the following minor problems @israelmcmc :
-for the tutorial pipeline_zero: the orientation file DC3_final_530km_3_month_with_slew_1sbins_GalacticEarth_SAA.fits has either changed name or location in wasabi, and currently the fetching command throws an error
-for the unit test: the apps are not working with the files already available in test_data. However, the test is passed anyway
Once these are solved, this can be converted to a normal PR. I am also open to feedback if things are better arranged in another way.