>> %-- Make a scatter plot of the two masses for each event candidate
>> plot(ins.mass1,ins.mass2,'.')
>> %-- Display a histogram of snr only for events with both masses > 6
>> hist(ins.snr(find(ins.mass1>6 & ins.mass2>6)))
The LIGOtools dataflow package includes a Matlab function (implemented as a MEX-file) called "readMeta" which reads a LIGO_LW file containing a table of database entries and places the contents of each table column into a Matlab array. Assuming that you set up your environment in the usual way (by invoking eval `<dir>/use_ligotools` in your .cshrc or .login file) your MATLABPATH environment variable will include $LIGOTOOLS/matlab, so the readMeta function will be available for use without any extra effort on your part. The syntax is rather simple, as you can see by typing "help readMeta" within Matlab:
readMeta --- Read LIGO metadata table into a MATLAB structure
By Peter Shawhan (shawhan_p@ligo.caltech.edu), April 2001
Revised by Peter Shawhan, Dec 2002, to automatically unpack spectra from BLOBs
Revised by Peter Shawhan, Feb 2003, to be able to read a specific table
Uses "metaio" parsing library by Philip Charlton
This returns a MATLAB structure with fields which correspond to
columns in the table. Each numeric column is converted to a vector
of double-precision real values, while each string or binary column
is converted to a cell array containing individual char arrays.
Usage: a = readMeta( file [,table] [,row [,col]] )
file is a "LIGO lightweight" XML file containing a Table object.
table (optional) specifies the name of the table to be read; this allows a
specific table to be read from a file containing multiple tables.
The table name is case-insensitive.
row is a row number (counting from 1), or a vector of row numbers
(in ascending order), e.g. [2:5,10] gives rows 2 through 5 and
row 10. If omitted, or equal to 0, then all rows are included.
col is a string containing one or more column names, separated by
commas and/or spaces. If omitted, all columns are included.
Spectrum BLOBs (from the summ_spectrum table) are automatically unpacked into
numeric arrays, assuming the mimetype is one of the standard LIGO ones.
Note: to access the contents of a Matlab cell array, use curly braces around
the index. For example, "a.spectrum{1}" gives the first spectrum array.
If readMeta(file) is called without assigning the output to a variable,
then it simply prints the column names and types and the number of rows.
As noted in the documentation above, there is an optional argument
which allows you to read a specific table out of a file containing
multiple tables. (This feature was added in version 4.10 of the
dataflow package, as was the code which unpacks BLOBs into
numeric arrays if they have a known mimetype.)
getMeta -d lho -q "select search,end_time,end_time_ns,mass1,mass2,amplitude,snr,chisq,event_id from sngl_inspiral where end_time > `tconvert jan 3` order by end_time,end_time_ns fetch first 1000 rows only" -o
< M A T L A B >
Copyright 1984-2000 The MathWorks, Inc.
Version 6.0.0.88 Release 12
Sep 21 2000
To get started, select "MATLAB Help" from the Help menu.
>> %-- First, just scan the file to see what it contains
>> readMeta('inspiral_events.xml')
search LSTRING
end_time INT_4S
end_time_ns INT_4S
mass1 REAL_4
mass2 REAL_4
amplitude REAL_4
snr REAL_4
chisq REAL_4
event_id ILWD:CHAR_U
Read 1000 rows
>> %-- Now read in the contents of the file
>> ins = readMeta('inspiral_events.xml')
Read 1000 rows
ins =
search: {1000x1 cell}
end_time: [1000x1 double]
end_time_ns: [1000x1 double]
mass1: [1000x1 double]
mass2: [1000x1 double]
amplitude: [1000x1 double]
snr: [1000x1 double]
chisq: [1000x1 double]
event_id: {1000x1 cell}
>> %-- Display a histogram of the signal-to-noise ratio
>> hist(ins.snr)
>> %-- Make a scatter plot of the two masses for each event candidate
>> plot(ins.mass1,ins.mass2,'.')
>> %-- Display a histogram of snr only for events with both masses > 6
>> hist(ins.snr(find(ins.mass1>6 & ins.mass2>6)))