


GAUSS(1)                 USER COMMANDS                   GAUSS(1)



NAME
     gauss - gaussian classifier for  pattern  recognition  work-
     bench

SYNOPSIS
     gauss (-l patterns|-t patterns|-T patterns|-r patterns) (-d)
     (-a) (-s) (-f falrate) paramfile

DESCRIPTION
     gauss  provides  a  general  purpose   gaussian   classifier
     designed to operate within the UCL pattern recognition work-
     bench.  The program reads classifier parameters from a  sup-
     plied  parameter  file,  and  stores updated parameters back
     into the file (creating it if  required).   Pattern  vectors
     are  supplied  in two files <pattern>.vec which contains the
     binary floating point data, and <pattern>.lab which contains
     the  labels  and offsets into the .vec file for each pattern
     vector.  The format of the patternfiles follows the UCL pat-
     tern  recognition workbench standard.  The program has three
     modes of operation: learn, test  and  recognise.   For  each
     mode,  the  user must supply the basename of the two pattern
     vector files.

  LEARNING MODE
     In learning mode, the current parameters  in  the  parameter
     file  are  updated  using the patterns in the pattern vector
     files.  If the parameter file does not  exist  then  a  NULL
     parameter set is created.  The updated parameters are stored
     back into the parameter file.

  TEST MODE
     In test mode, the current parameters in the  parameter  file
     are  used to generate outputs for each vector in the pattern
     vector files which is then compared with the supplied label.
     For  switch  '-t'  performance  statistics  are printed, for
     switch '-T' statistics  are  saved  into  the  PRW  database
     /usr/prw/records.

     The recognised class is chosen (in the default case) as  the
     nearest  distribution, and the results are printed as a con-
     fusion matrix.  The '-a' switch may be employed for 2  class
     continuous  problems,  where the value (p1+p2)/p1 is used an
     'activation' of the classifier and thresholded in  an  anal-
     gous  way to the neural-net classifiers.  With the -a option
     the statistics are based on the location of a threshold that
     equalises the miss-rate and the false-alarm rate.  An option
     (-f falrate) allows the setting of a  specific  false  alarm
     rate.  An  option  (-s)  produces  files  for  plotting  the
     Receiver Operating Characteristic.

  RECOGNITION MODE
     In recognition mode, the current parameters in the parameter



PRW                     Last change: UCL                        1






GAUSS(1)                 USER COMMANDS                   GAUSS(1)



     file  are  used  to  generate outputs for each vector in the
     pattern vector files.  The input label  and  the  recognised
     output   are   written   to  the  files  <pattern>.gc.i  and
     <pattern>.gc.r respectively.  With the -a option, the recog-
     nised output is the 'activation' of the classifier.

     Options:

     -I         Identify the program name and version.

     -l patterns
                Select learn mode.  Update parameters by training
                with supplied pattern vector files.

     -t patterns
                Select test mode.  Print  statistics  of  perfor-
                mance  of  classifier  on supplied pattern vector
                files. Equal error based on frame percentages.

     -T patterns
                Select test mode.  Save statistics of performance
                of  classifier  on  supplied pattern vector files
                into PRW record database. Equal  error  based  on
                frame percentages.

     -r patterns
                Select recognise mode.  Output labels  into  sup-
                plied  pattern  vector  files.   -d  Use Diagonal
                covariance matrices only.   This  simplifies  the
                calculations  and  can  lead  to improved perfor-
                mance.  In learn mode, the stored model  is  con-
                verted  to  a  diagonal matrix.  In test mode the
                stored model is unaffected.  -a Use the  'activa-
                tion'  of  the classifier in testing and recogni-
                tion.  Two-class continuous problems only.

     -s         (Test  mode  only)   Produce   statistics   files
                <model>.rcx   (False-alarm   rate   as  threshold
                increases), <model>.rcy (Hit  rate  as  threshold
                increases)  for  plotting  ROC.  Implies 'activa-
                tion' mode.

     -ffalrate  (Test mode only) Produce  performance  statistics
                for  a  given  false-alarm  rate.   Default is to
                choose equal-error point).   Implies  'activation
                mode'.

FILES
     /usr/prw/records - PRW statistics records database

VERSION/AUTHOR
     2.0 - Mark Huckvale



PRW                     Last change: UCL                        2






GAUSS(1)                 USER COMMANDS                   GAUSS(1)



SOURCE
     /users/mark/sfs/man/../prw/gauss.src/gauss.c





















































PRW                     Last change: UCL                        3



