re-make jpegs

[davidwhogg]

Can you re-make the jpegs now down to h50i, but for 5 color quantiles, 3 SB quantiles.

When / if it fails, send me the logs and I will get it unblocked.

Since you saved the jpegs you have already made, this should only have to make the new jpegs.

Ask here if you have questions about what I mean.

Some time sensitivity to this!

[ekta1224]

our current quantiles are already cut at 5x3 and they are for h50i > 20.

we were breaking on quantile 13 with the following error:

% MWRFITS: Appending FITS extension to file NGC_4565_SA_J123620.62+255911.5-i.fits % Array dimensions must be greater than 0. % Execution halted at: HOGG_OPEN_OVERLAY 20 /global/data/products/Linux64/idlutils/v5_5_8/pro/rgbcolor/hogg_open_overlay.pro % SMOSAIC_MAKE_JPG_OVERLAY 49 /global/data/products/Linux64/photoop/v1_11_1/pro/image/smosaic_make_jpg_overl ay.pro % SMOSAIC_MAKE_JPG 192 /global/data/products/Linux64/photoop/v1_11_1/pro/image/smosaic_make_jpg.pro % SLOAN_ATLAS 112 /mount/hydra1/scr/dwh3/ep109/all_images/sloan_atlas.pro % $MAIN$ % Program caused arithmetic error: Floating divide by 0 % Program caused arithmetic error: Floating underflow % Program caused arithmetic error: Floating overflow % Program caused arithmetic error: Floating illegal operand

here are the quantiles: http://broiler.astrometry.net/~ep1091/quantiles/quantile00/index..html

[davidwhogg]

While I figure that crap out, are you saying we have made all jpegs down to h50i > 15.? If so, aweseome! If not, can you run them?

[ekta1224]

hmm no, i guess that was cut at 20, so i will redo and send a new error if it's different.

On Wed, Feb 12, 2014 at 12:12 PM, David W. Hogg notifications@github.comwrote:

While I figure that crap out, are you saying we have made all jpegs down to h50i > 15.? If so, aweseome! If not, can you run them?

Reply to this email directly or view it on GitHubhttps://github.com/davidwhogg/SloanAtlas/issues/55#issuecomment-34891392 .

Ekta Patel

[davidwhogg]

Please do

[davidwhogg]

is this code running now in the background? Let me know if you need help. I am EAGER

[ekta1224]

i will set it up right now.

[ekta1224]

running.

[davidwhogg]

awesome. Stay on target. Do you run the different quantiles as separate processes?

Also, where is the catalog file that you are working from?

Also, can you go to where the jpegs are being created and run

 \rm -vf NGC_4565_*

[ekta1224]

no they run one at a time through a for loop by quantile number on bootes.

the catalog file i am using is the one i created for you the other day (sdss_atlas_plates_quants.fits) in /home/ep1091/penguin/tractor, cut at h50i > 15 in 5x3 quantiles (copied over to bootes).

removed the files.

[ekta1224]

if you want to start looking at these, 00-07 are done and the next is running. http://broiler.astrometry.net/~ep1091/quantiles/quantile07/index..html

[davidwhogg]

Okay cool. Let me know if it fails anywhere.

[davidwhogg]

Are the images being put into the directory /global/data/scr/dwh3/ep109/all_images/ ?

If not, where are they going?

I am finding no new files. Or many missing files. Are we putting some radius cut in that we don't mean to be?

Is the JPEG-making code you are running fully checked in and pushed?

[davidwhogg]

To be clear, in that file I see NGC 337 and it has QUANTILE=0. and I don't see it anywhere on bootes. Any ideas?

[davidwhogg]

[dwh2@bootes ep109]$ python Enthought Python Distribution -- www.enthought.com Version: 7.0-1 (64-bit)

Python 2.7.1 |EPD 7.0-1 (64-bit)| (r271:86832, Nov 29 2010, 13:51:37) [GCC 4.1.2 20080704 (Red Hat 4.1.2-44)] on linux2 Type "help", "copyright", "credits" or "license" for more information.

import pyfits hdulist = pyfits.open("sdss_atlas_plates_quants.fits") table = hdulist[1].data import numpy as np qq = np.where(table.NAME == "NGC 337") print qq (array([388]),) print table.QUANTILE[qq] [ 0.] print table.NAME[qq] ['NGC 337']

[dwh2@bootes ep109]$ pwd /global/data/scr/dwh3/ep109

[davidwhogg]

Okay figured something out: Because you (sensibly!) made quantile_0.log files, I can see that the system couldn't find the data for NGC 337. For now I am going to open a new issue and work around.

This issue can be closed when the re-build of jpegs is complete.

[ekta1224]

that's very odd to me. looking at how many we should have in the first quantile, i get 65, but the webpage seems to only have 62.

[davidwhogg]

How are we going on the quantiles; still building?

[ekta1224]

yep. running 11 and 12 now. should be finished within the next few hours.

[davidwhogg]

Okay cool; stay on target!

[ekta1224]

13 and 14 running now and we are done. i won't be at my computer, but if you check the time stamps for the quantiles, you'll be able to tell if they ran without error in about 1-2 hours.

http://broiler.astrometry.net/~ep1091/quantiles

[ekta1224]

all done! didn't seem to hit an error this time around...

flip through them here: http://broiler.astrometry.net/~ep1091/quantiles

[davidwhogg]

This is GREAT. I owe you big time. Closing