EranOfek
commented
1 year ago Hi Yarin,
Please send me the PSF matrix in mat or ascii file so I can test it. Note that sending mat files via the WIS network is not possible (send it to my gmail). Thanks, Eran
On Tue, Jul 25, 2023 at 9:10 PM YarinMS @.***> wrote:
In order to use the forced photometry tool, we would like to run the pipeline with an option to save the proc images as well, by executing : [image: image] https://user-images.githubusercontent.com/118757319/255983141-74aebdc0-03b7-4256-ba1f-914c4514b50e.png
It looks like the pipeline fails for all the images, although they look ok (to me). For this report I've used images from : group 1 - First image: /last04w/data2/archive/LAST.01.04.04/new/LAST.01.04.04_20230720.230849.762_clear_264+65_020_001_001_sci_raw_Image_1.fits
up to group 3 - First image: /last04w/data2/archive/LAST.01.04.04/new/LAST.01.04.04_20230720.215515.584_clear_264+65_020_001_001_sci_raw_Image_1.fits
For the time being I'll leave these images in the failed directory, them and only them. In case anyone would like to examine them. A failed image to demonstrate, [image: image] https://user-images.githubusercontent.com/118757319/255985524-f80befb6-fe8e-4f2b-90b8-00d3f23b8906.png
While running the pipeline it fails in processing all the groups. We get the following try error: [ERR] pipeline.DemonLAST try error: Subscripted assignment between dissimilar structures. / funname: AstroPSF.curve_of_growth @ line: 432 [image: image] https://user-images.githubusercontent.com/118757319/255982710-ae18e6b0-8b2d-41bc-a29f-745c9590610d.png
Eran asked to check what is the content of the PSFData in the Coadd AstroImage BEFORE the call to fhwm, to do so, we run the pipeline on debug mode and taking the next break points: [image: image] https://user-images.githubusercontent.com/118757319/255986609-8cb6a757-ed6a-413b-8455-57e34d54828c.png Which uses the function,
pipeline.generic.multiRaw2procCoadd
Next, we are sent to,
pipeline.generic.singleRaw2proc [image: image] https://user-images.githubusercontent.com/118757319/255987252-fe2669bb-e34d-42d0-a56f-b71e68d8d0d9.png Then, we encounter the first call to fwhm : [image: image] https://user-images.githubusercontent.com/118757319/255987561-dd7464ef-b103-479f-a0f2-7224188dad58.png
Assuming the content we are looking for lies in : SI.PPSFData We get a non-empty AstroPSF object : [image: image] https://user-images.githubusercontent.com/118757319/255988108-6dfb0a42-47c3-44f4-b53c-10cbbf72597b.png
Looking further, SI.PSFData.Data or SI.PSFData.DataPSF demands to choose a subframe (out of 24), looking at an arbitrary sub image (15) around the center of the detector :
SI.PSFData.Data == SI.PSFData.DataPSF and it looks like the PSF spreads the light in a gaussian manner, (as expected?) : [image: image] https://user-images.githubusercontent.com/118757319/255989112-e7069f4f-526f-4b2d-9b7e-bbf492987613.png And the lower part : [image: image] https://user-images.githubusercontent.com/118757319/255989237-5e280bd3-4681-4bf9-91e2-cf5f568d23c0.png Making sure it is normalized.
So, assuming I understood what Eran asked, the problem originates deeper in AstroPSF and specifically in imUtil.psf.curve_of_growth
— Reply to this email directly, view it on GitHub https://github.com/EranOfek/AstroPack/issues/199, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJUQ4LR2HCVACDU4N6CPBDXSADYVANCNFSM6AAAAAA2XOEDFY . You are receiving this because you are subscribed to this thread.Message ID: @.***>
YarinMS
In order to use the forced photometry tool, we would like to run the pipeline with an option to save the proc images as well, by executing :
It looks like the pipeline fails for all the images, although they look ok (to me). For this report I've used images from : group 1 - First image:
/last04w/data2/archive/LAST.01.04.04/new/LAST.01.04.04_20230720.230849.762_clear_264+65_020_001_001_sci_raw_Image_1.fitsup to group 3 - First image:/last04w/data2/archive/LAST.01.04.04/new/LAST.01.04.04_20230720.215515.584_clear_264+65_020_001_001_sci_raw_Image_1.fitsFor the time being I'll leave these images in the failed directory, them and only them. In case anyone would like to examine them. A failed image to demonstrate,
While running the pipeline it fails in processing all the groups. We get the following try error:
[ERR] pipeline.DemonLAST try error: Subscripted assignment between dissimilar structures. / funname: AstroPSF.curve_of_growth @ line: 432Eran asked to check what is the content of the PSFData in the Coadd AstroImage BEFORE the call to fhwm, to do so, we run the pipeline on debug mode and taking the next break points:
Which uses the function,
pipeline.generic.multiRaw2procCoaddNext, we are sent to,
pipeline.generic.singleRaw2procThen, we encounter the first call to fwhm :
Assuming the content we are looking for lies in :
SI.PPSFDataWe get a non-empty AstroPSF object :Looking further,
SI.PSFData.DataorSI.PSFData.DataPSFdemands to choose a subframe (out of 24), looking at an arbitrary sub image (15) around the center of the detector :SI.PSFData.Data == SI.PSFData.DataPSFand it looks like the PSF spreads the light in a gaussian manner, (as expected?) :And the lower part :
Making sure it is normalized.
So, assuming I understood what Eran asked, the problem originates deeper in
AstroPSFand specifically inimUtil.psf.curve_of_growth