- ←Prev
- PICARD — a PIpeline for Combining and Analyzing Reduced Data
- Next→
- TOC ↑
Description:
Apply PCA processing to raw SCUBA-2 data to determine dominant modes.
The input data should contain a fast-ramp flatfield (taken prior to the target data).
Notes:
-
Input files must contain raw SCUBA-2 data.
-
Input data should be from a single subarray only, and for a single observation. However, no checks are
made that this is actually the case.
-
If results are to be calculated by the recipe, then all the input data are used. This could lead to long run
times with a large number of files as the data are pre-processed with SMURF sc2clean.
Available
Parameters :
The following parameters can be set via the -recpars option:
LOGFILE
Flag to
denote whether to write results to a log file at the end of processing. Default is 1 (write a
log file).
PCA_COMP
PCA components to analyze and display. Default is 0 to 5. The
components may be specified either as a comma-separated list (e.g. 0,1,2,3 etc - they need not be
contiguous or in order), or as a Perl array slice (e.g. 0..3). The number of components must be no
more than 8. If more than 8 are given, only the first 8 are used.
PCA_KEEPFILES
Flag to
indicate which files should be kept on disk. Default is 1 which keeps the PCA amplitude,
component and power spectrum files on disk. A value of 0 deletes all files, while a value of
PCA_REUSE
Flag to indicate that existing data should
be used if present. Default is 1 (reuse).
Display
The results for each chosen PCA component are
displayed in up to two KAPVIEW windows. The left-hand column displays the amplitude scaled between
/-2
sigma, the next column displays the component as a function of time and the third column shows the
power spectrum of each component. Each KAPVIEW window can display results for up to 4 PCA
components.
Copyright © 2014 University of British Columbia and the Science & Technology Facilities Council
-
←Prev
- PICARD — a PIpeline for Combining and Analyzing Reduced Data
- Next→
- TOC ↑