Skip to content

Latest commit

 

History

History
74 lines (50 loc) · 2.51 KB

CHANGELOG.md

File metadata and controls

74 lines (50 loc) · 2.51 KB
Raw

Change Log

Probably irrelevant now

Running v0.5 on K2 C1

Use the pixdecor.sh function to run the complete photometric pipeline and transit search.

pixdecor.sh -c C1 -r C1_02-03 -s 201367065 -t ${K2PHOTFILES}/pixeltrans_C1_ch04.h5

flags

  • -d run in debug mode (shorter versions of all the code)
  • -c which K2 campaign are we working with?
  • -s what is the starname?
  • -t with pixel-transformation file do we use?

v0.3 -> v0.4

Changes to current pipeline

  • New method of solving for orientation of spacecraft based on many stars. Implemented in the channel_centroids.py and centroid.py modules

  • 2D Gaussian process-based detrending with iterative identification of outliers provided a major advance in noise reduction. This is implemented in pixel_decorrelation4.py.

v0.2 -> v0.3

Changes to current pipeline:

  • flatfield.py solves for flatfield
  • pixel_decorrelation2.py decorrelates against 2D position and time

New code (still experimental):

  • diffimage.py (code to compute difference images)
  • pointing.py (code to incorporate Tom B. pointing info)

v0.1 -> v0.2

  • Location of the star is acquired by using the WCS headers to achieve an initial guess. Then we generate a synthetic image and register that image to the reference image to determine the correction to the WCS solution
  • Diagnostic plots show the positions of the stars according to the epic catalogs
  • dx,dy between frames is computed using subpixel registration

Old Examples

Examples from previous runs of the code. Calls may be deprecated. Checkout the older versions to recover.

Running v0.3 on K2 C0

python flatfield.py --tmin 1940 --tmax 1972 <path to pixel .fits file> -d <output directory> <sqlite3 database file>
python pixel_decorrelation2.py --h5file <path to flatfield .ff.h5 file> <output directory>

Running v0.2 on K2 C0

. /global/homes/p/petigura/k2_setup.sh
python pixel_decorrelation.py -f <path to pixel .fits file> --wcs=1 -r 4 --minrad=2 --maxrad=8 --verbose=1 --gausscen=0 --plotmode=gs --tmin=1940 --output=pobj,fits --xymeth=xcorr2D -s <output directory>

Notes:

  • --xymeth=xcorr2D keyword forces use of 2D image registration
  • --output=pobj,fits dumps both fits and pickle versions of the photometry obhect
  • --wcs=1 Determines aperture center by using coordinates (after refinement)

Running v0.1 on K2 C0

python pixel_decorrelation.py -f <path to pixel .fits file> --wcs=1 -r 4 --minrad=2 --maxrad=8 --verbose=1 --gausscen=0 --plotmode=gs --tmin=1940 --output=2