search

ACES-CMZ / reduction_ACES

Reduction scripts and tools for ACES
https://worldwidetelescope.org/webclient/?wtml=https://data.rc.ufl.edu/pub/adamginsburg/ACES/mosaics/mosaics.wtml
15 stars 12 forks source link

Execution Block ID uid://A001/X15a0/Xd0 Sgr_A_st_i_03_TM1 #138

Open keflavich opened 2 years ago

keflavich commented 2 years ago

Sgr_A_st_i_03_TM1 uid://A001/X15a0/Xd0

Product Links:

Reprocessed Product Links:

xinglunju commented 2 years ago

We will use this one for offline training at SHAO, together with the 7m SB toward the same region (#37)

xinglunju commented 2 years ago
keflavich commented 2 years ago

@xinglunju Usually spw33 doesn't image at all b/c of size mitigation - stopping because of divergence is more than we usually get! It's actually strange that all of the windows got binned instead of just ditching one.

xinglunju commented 2 years ago

@keflavich I just submitted a change to the override_tclean_commands.json file, but I still have a few questions:

keflavich commented 2 years ago

How do we deal with the CLEAN diverging problem? Should I set 'cycleniter' to 100?

Probably better to set cyclefactor=2 or 3.

Do we need to change the imagename?

Not in override_tclean_commands, no, but if you run your own clean, you should add reclean to the name following the pattern set by other recleans - instructions for that are somewhere but I don't have them immediately to hand

How do we know if the re-processing has started or finished on hipegator?

You just have to check for the files. I run the imaging semi-regularly, but it takes days to complete an image run.

xinglunju commented 1 year ago

The spw33 cube in globus looks great. The other spws are still 2x binned, and are probably still undergoing re-clean.

xinglunju commented 1 year ago

Update on Mar 1 2023: except for SPW33 (which looks good after re-imaging), the other cubes in Globus are not updated.

keflavich commented 1 year ago

Moved all except spw33 to pre_mar1_2023/

xinglunju commented 1 year ago

Re-imaged SPW25/27/33 cubes look good. SPW29/31/35 cubes still missing on Globus. If the pipeline is not running, I will run it on the local machine.

keflavich commented 1 year ago

Reimaging is working on hipergator now, but I have a lot of bugs to work out still.

There's some cool stuff in HC3N in spw35: image

keflavich commented 1 year ago

spw29 is reimaging. The second subcube is at:

2023-06-18 20:38:51     INFO    task_tclean::SynthesisImager::executeMajorCycle         ----------------------------------------------------------- Run Major Cycle 37 -------------------------------------

😱

keflavich commented 1 year ago

spw 29 is done.

spw 29, 31 have stippling, which is a form of failed convergence. These are two adjacent channels.

image image
xinglunju commented 10 months ago

As noted by Adam, the SPW29 and 31 cubes have stippling issue. Why does this issue only happen in the two narrow SPWs? Because these SPWs are easier to diverge?

SPW35 cube looks good. More interesting HC3N features in addition to that arc noted by Adam, a collimated jet or filament? Screenshot_2023-10-17_10-33-47

xinglunju commented 9 months ago

Confetti in SPW29 seems to be fixed after increasing cyclefactor from 2.5 to 3.5. Now running re-imaging of SPW31. Red: spectrum from the new cube. Blue: spectrum from the old cube.

image
xinglunju commented 9 months ago

Confetti in SPW31 seems to be fixed after increasing cyclefactor from 2.5 to 3.5. But brightnesses clearly change. I will discuss this on WP1 meetings and then decide whether to consider this SB as done. Red: spectrum from the new cube. Blue: spectrum from the old cube. image

xinglunju commented 9 months ago

Ash's suggestion: double check the model/mask to make sure it has actually been cleaned

xinglunju commented 9 months ago

Confirmed that models and masks exist in the recleaned data. I will mark this one as 'done' for now, and will upload the new cubes to Globus, what do you think? @keflavich @ashleythomasbarnes

ashleythomasbarnes commented 9 months ago

Could you share a screenshot of the mask, model, and image as one of the confetti channels?

xinglunju commented 9 months ago

Top left: recleaned image. Top right: old pipeline image Bottom left: model. Bottom right: mask (yellow = 1)

Spectra: red -- reclean image; blue -- old pipeline image

image

nbudaiev commented 6 months ago

QA - Line contamination in continuum images from high/low frequencies

Looks okay. Maybe slight contamination in the spw27 selection.

uid___A001_X15a0_Xca s36_0 Sgr_A_star_sci spw33_35 cont I iter1 image tt0-uid___A001_X15a0_Xca s36_0 Sgr_A_star_sci oldhigh_spw33_35 cont I iter1 image tt0-uid___A001_X15a0_Xca s36_0 Sgr_A_star_sci sp-2024-02-28-13-20-31 uid___A001_X15a0_Xd0 s38_0 Sgr_A_star_sci spw35 cube I iter1 image pbcor statcont contsub_diagnostic_spectra