First, try running TARGET on the example in the examples
directory.
(1) setup target
(2) cd to a directory where you want to work
(3) Execute the command:
unix> target
You will get a "ta>" prompt
(4) Now you are in a tcl shell. The example input files for target
are in the $TARGET_DIR/examples directory. The output will
go in your working directory. Here's how you generate the
selected objects:
ta> targetSelectFromChunk \
>> [envscan \$TARGET_DIR]/examples/tsChunk-25-647905.par \
>> -quasars [envscan \$TARGET_DIR]/examples/tsQuasarsParams.par \
>> -objDir [envscan \$TARGET_DIR]/examples
This should have generated the file tsSelectedObjects-000581-1-0-0004.par in
your current directory. This is an ASCII file that you should be able to
look at. (NOTE: The current examples do not run on sdssdp2 - OSF1 - due to
NaN problems. The simulations should work on OSF1, however.)
(5) We can now generate the figure of merit by:
ta> taQsoSimTargetEvalFromChunk \
>> [envscan \$TARGET_DIR]/examples/tsChunk-25-647905.par \
>> tsSelectedObjects-000581-1-0-0004.par [envscan \$TARGET_DIR]/examples
Since the keys in the examples in the current cut product are zero,
the figure of merit is unable to determine the completeness and efficiency
by redshift and type - and can only tell you how many objects you
selected out of the total number of objects.
(6) You can also convert one one Xiaohui's simulations to the input
format for the target selection pipeline and run the above. Suppose you
have a file called NGP9d.ugriz-1010 that is such a simulation in your
current working directory. Note that the converstion routine has only
been tested on chunks containing one field, but it was designed to work
on multiple-field chunks. A sample session follows:
ta> taQsoSimConvert NGP9d.ugriz-1010
Creating chunk 1
Creating segment 1
Creating fieldinfo for field 0
Appending 13402 objects in field 0 to file tsObj-000001-3-0-0000.fit
Writing segment 1 to file tsFieldInfo-000001-3-0-0000.fit
Writing chunk 1 to file tsChunk-21-000182.par
ta> targetSelectFromChunk tsChunk-21-000182.par -quasars tsQuasarsParams.par
field 0 has 334 objects (out of 13402) with Target flag(s) set
0
ta> taQsoSimTargetEvalFromChunk tsChunk-21-000182.par \
>> tsSelectedObjects-000001-3-0-0000.par .
Targets selected (by type): 1 13/991
2 35/9801
3 5/2324
4 0/0
5 0/0
6 0/0
7 0/0
8 58/58
9 14/16
10 209/212
-----------------
334/13402
QSO completeness (z): -10 - 1 85/85
1 - 1.5 48/48
1.5 - 2 32/32
2 - 2.5 24/24
2.5 - 3 15/16
3 - 3.5 5/6
3.5 - 4 0/1
4 - 6 0/0
6 - 10 0/0
-----------------
209/212
QSO completeness (m): -10 - 15 0/0
15 - 16 1/1
16 - 17 1/1
17 - 18 20/20
18 - 19 91/92
19 - 30 96/98
-----------------
209/212
ta>
The conversion of the simulated data into a FITS file is done within the RADIO product. The matching of the radio sources with optical objects is done within TARGET.
Assume that the original quasar simulation is: /data/sim/sim.dat
The actual setting of flags is done within TARGET. The FIRST FITS catalog can be moved to the directory that has the simulated chunk file. Assume that this is in /data/sim/tsChunk-sim.par and that there is a corresponding field info file in the same directory. The tunable parameters used for the known object matching are given in a parameter file, with the format given by $TARGET_DIR/examples/koMatch.par
For additional options and details about sim2First and first2Fits go to the RADIO Home Page.
There are also additional options for koMatch as well as details about the currently implemented matching algorithms.