xmatcmp(1) xmatcmp(1)
NAME
xmatcmp - graphical tool (for X using Tcl/Tk) for comparing and
visualizing matrices
SYNOPSIS
xmatcmp [-sparse] [-tol ] [operator]
[|]
DESCRIPTION
Visualizes the contents of a matrix, or performs a comparison between
two matrices (or between a matrix and a constant) and displays the
output in graphical format. The input files, file1 and file2 , contain
snapshot summaries of matrix contents in the format described in mcf(5).
The last argument to the command may be either a filename or a constant.
It is assumed to be a file, but if no such file exists, it will be
considered a constant. The constant should be specified as either an
integer or a floating point number (see options for examples).
An optional operator determines how the comparison is performed. Six
predefined operators may be specified using C-like or Fortran-like
notation. The following list provides all the valid operators:
EQ "=" GT ">" GE ">=" LT "<" LE "<="
If the argument is not a predefined operator, it is assumed to be a
user-defined comparison function. See mcf(5) for description of
comparison functions. If an operator is not provided, a default of
equality is assumed.
An optional tolerance determines if an element in the result matrix is
a mismatch or not. Only elements whose value exceeds the tolerance
value are counted as mismatches. If no tolerance value is specified,
a default of 0.0 is assumed.
The main display visualizes the result matrix in the form of a
rectangular grid where each square represents a matrix element.Squares
are colored according to their value (i.e., the magnitude of the
difference calculated by comparing the corresponding elements of the
two inputs). Positioning the cursor over a square, reveals the indices
and value of the element. Button controls allow the user to select a
different predefined operator, apply a user-defined function to compare
the matrices, or change the tolerance value; the display is updated to
show the new results. A color key shows the mapping of colors to
magnitude ranges, and also summarizes the results, in terms of the
number of mismatches and the magnitudes of the average and worst-case
mismatches.
QUICK REFERENCE GUIDE
To see the indices/value of an element, position the cursor over the
corresponding square in the main display.
To change the comparison operator, select the appropriate button from
the button selection at the bottom of the main window.
To change the tolerance value, type the new value in the entry field
at the bottom of the main window and hit the Return key.
To hide/redisplay the Color Key, choose the appropriate setting on
the Options menu.
To change the size ranges for mismatches, choose Change Color Mapping
from the Options menu.
To change a color, click on that colorUs square in the Change Color
Mapping window.
To save the matrix being displayed, select Save Results File from the
File menu.
To display this Quick Reference guide, select Quick Reference from the
Help menu.
To exit the tool, select Quit from the File menu.
OPTIONS
-sparse indicates that the matrix snapshots are in
sparse format (see mcf(5) for file formats).
Dense format is assumed by default.
-tol apply tolerance instead of the default of 0.0.
The tolerance is specified as either an integer
or a floating point number. Examples of valid
values for tolerance are: 8.197620, -4,
0.495, -13.045, etc.
ENVIRONMENT
DISPLAY specifies the default host and display number
(for X Window System calls).Tcl and Tk are required to run the tool.
SEE ALSO
X(1), Tcl, Tk, matcmp(1)
COPYRIGHT
The design and implementation of the Matrix Comparison Facility
(MCF) are available for royalty-free adoption and use for
non-commercial purposes, by any public or private organization.
Copyright is retained by Oregon State University.
Redistribution of any part of MCF or any derivative works must
include this notice.
MCF was developed by Bhanu Balasubramaniam, Department of Computer
Science, Oregon State University, 1995.
Back to MCF home page