search

linan7788626 / Arc_Finding_CNN

0 stars 1 forks source link

Modification #14

Closed WentaoLuo closed 8 years ago

WentaoLuo commented 8 years ago

figure_1 figure_2

I think the lens_galaxy.py contains a bug that make the pixel value very large. Now I compared to Rachel's SHERA and get reasonable pixel values for SDSS.

WentaoLuo commented 8 years ago

figure_2 figure_3

The top figure is the converted ACS image to SDSS image without Gaussian noise. The colorbar denotes the pixel value.

Guassian noise has been added to the bottom one.

The last thing to check is the unit of the magnitude from SDSS website and make sure the magnitude is in either AB or Pogson. It seems that the schema crashed since yesterday.

WentaoLuo commented 8 years ago

figure_4

Sorry, wrong figure.

WentaoLuo commented 8 years ago

I did not add softbias=1000.0 and skybackground~1100.0

WentaoLuo commented 8 years ago

I will push these modified codes after I confirm the units.

Good luck guys!

Wentao

linan7788626 commented 8 years ago

Hi Wentao,

It looks reasonable now. We can talk later today.

Thanks, Nan

—— Nan Li Postdoctoral Fellow University of Chicago & Argonne National Lab linan7788626.github.io

On Feb 9, 2016, at 10:48 AM, Wentao Luo notifications@github.com wrote:

I will push these modified codes after I confirm the units.

Good luck guys!

Wentao

— Reply to this email directly or view it on GitHub https://github.com/linan7788626/Arc_Finding_CNN/issues/14#issuecomment-181952141.

WentaoLuo commented 8 years ago

Sure. Let's discuss it later.

linan7788626 commented 8 years ago

I will close issue. And we will focus on producing DES images and LSST images in the next step.