Regarding the variable_PSF/ground/constant branch #17
Comments
|
Hi Jun - You're right that something seems odd here. I have a few thoughts: (2) If you've already unzipped and untarred it, then another option would be for you to e-mail me a single randomly-chosen catalog or image file and I will compare with what I have here to see if they are the same. (However, note that I am going away after 7pm Eastern time on Sunday, and will be completely online for 2 days after that.) (3) Usually a score of 0 means something is actually wrong with the format of the file being submitted. Did you run your submission through the submission checking script at I hope that helps!
|
|
Actually, Jun, given that I'm going away in a few hours, I thought it best to send you some things you can compare with. I've gone into my unzipped version of the variable_psf/ground/constant dataset, and computed md5sums for the image and catalog files in subfield 137 (which I chose semi-randomly). You can use these to check against your files: b9bea8261f31944c6767948ad42cfbdb galaxy_catalog-137.fits If anything is unclear, perhaps someone else can help you while I'm offline. ( @msimet , if you want to know where my unzipped copy of this branch lives, it's on coma, /physics/rmandelb/great3-v11/public/variable_psf/ground/constant - so if any other questions about versions come up, could you please help Jun out? I'm not sure if Barney has an unzipped version of this branch somewhere.) |
|
Noted, thanks--I'm happy to look into things if necessary. |
|
Hi Rachel, Thank you so much for the information!! I will try to compare myself using your data. Best, ----- 原始邮件 ----- Actually, Jun, given that I'm going away in a few hours, I thought it best to send you some things you can compare with. I've gone into my unzipped version of the variable_psf/ground/constant dataset, and computed md5sums for the image and catalog files in subfield 137 (which I chose semi-randomly). You can use these to check against your files: b9bea8261f31944c6767948ad42cfbdb galaxy_catalog-137.fits If anything is unclear, perhaps someone else can help you while I'm offline. ( @msimet , if you want to know where my unzipped copy of this branch lives, it's on coma, /physics/rmandelb/great3-v11/public/variable_psf/ground/constant - so if any other questions about versions come up, could you please help Jun out? I'm not sure if Barney has an unzipped version of this branch somewhere.) Reply to this email directly or view it on GitHub: |
|
Hi Rachel, My files did pass the md5sum check. I will try to figure out the problem. Thanks again for your time! Best, ----- 原始邮件 ----- Actually, Jun, given that I'm going away in a few hours, I thought it best to send you some things you can compare with. I've gone into my unzipped version of the variable_psf/ground/constant dataset, and computed md5sums for the image and catalog files in subfield 137 (which I chose semi-randomly). You can use these to check against your files: b9bea8261f31944c6767948ad42cfbdb galaxy_catalog-137.fits If anything is unclear, perhaps someone else can help you while I'm offline. ( @msimet , if you want to know where my unzipped copy of this branch lives, it's on coma, /physics/rmandelb/great3-v11/public/variable_psf/ground/constant - so if any other questions about versions come up, could you please help Jun out? I'm not sure if Barney has an unzipped version of this branch somewhere.) Reply to this email directly or view it on GitHub: |
|
Hmmm, interesting, but I'm sorry there's no quick fix. Please let me know if you have any other questions that could help answer this problem... |
|
Hi Jun, I just wanted to flag up #18 , which is possibly relevant, and offer my sincere apologies if this problem in the submission checker script contributed to your not being able to successfully submit. I hope that you will nonetheless consider submitting to the post-challenge leaderboards when these go up (soon)! |
|
Hi Barnaby, Thanks a lot! My bug is a little subtle though. It will be very helpful if from somewhere, I can download both the galaxy images and their corresponding PSF images (at the galaxy positions) for the branch of "variable-PSF_ground_constant". It is because I am still not sure if I am using the right data, and if I ran the "psf_models.py" code correctly for reconstructing the PSF field. Would it be a lot trouble for you to generate the PSF images at the positions of the galaxies, and upload them so that I can download? If so, I can quickly re-run my code (the part without PSF reconstruction), and submit my results. Thank you very much for your time! Best, ----- 原始邮件 ----- Hi Jun, I just wanted to flag up #18 , which is possibly relevant, and offer my sincere apologies if this problem in the submission checker script contributed to your not being able to successfully submit. I hope that you will nonetheless consider submitting to the post-challenge leaderboards when these go up (soon)! Reply to this email directly or view it on GitHub: |
|
Hi Jun, OK, well I'm glad that it seems the However, in the near future we will be releasing the GREAT3 simulation code as well as the truth data, so you should be able to edit that and generate the images that you want on your own computing network, without needing to download the images. We'll keep everyone updated about this once it's ready. |
|
Hi Barnaby, Sure. Thanks. I understand that you must be very busy right now. I will return to my issue in the near future. Best, ----- 原始邮件 ----- Hi Jun, OK, well I'm glad that it seems the However, in the near future we will be releasing the GREAT3 simulation code as well as the truth data, so you should be able to edit that and generate the images that you want on your own computing network, without needing to download the images. We'll keep everyone updated about this once it's ready. Reply to this email directly or view it on GitHub: |
Hi Guys,
I've worked on the variable_PSF/ground/constant branch for quite a while. The interesting thing is that I got roughly the same shear values for different sub-fields in the same field (there are 20 of them in each field). I tried two different ways of reconstructing the PSF field myself, and I also tried to use the PSF reconstructed by the code provided by GREAT3. All of these methods yielded very similar answers consistently, but scored "0". Now I suspect that I downloaded the wrong set of data. Have people ever had this situation? Had the data in the "variable_PSF/ground/constant" branch been updated before? Thanks in advance for your time!!
Best,
Jun
The text was updated successfully, but these errors were encountered: