plssvm-scale.1.in

.\" Manpage for plssvm-scale.
.TH PLSSVM-SCALE 1 "13 December 2022" "2.0.0" "plssvm-predict Manual"

.SH NAME
plssvm-scale - LS-SVM with multiple (GPU-)backends

.SH SYNOPSIS
svm-scale [OPTION...] input_file [scaled_file]

.SH DESCRIPTION
plssvm-scale is a utility to scale the data points in a data set to a specific range.

.SH OPTIONS

.TP
.B -l, --lower arg
the lowest (minimal) value allowed in each dimension of the data set (default: -1)

.TP
.B -u, --upper arg
the highest (maximal) value allowed in each dimension of the data set (default: 1)

.TP
.B -f, --format arg
the file format to output the scaled data to: libsvm|arff (default: libsvm)

.TP
.B -s, --save_filename arg
the file name to which the scaling factors should be saved

.TP
.B -r, --restore_filename arg
the file name from which previous scaling factors should be loaded

@PLSSVM_PERFORMANCE_TRACKER_MANPAGE_ENTRY@

.TP
.B --use_string_as_labels arg
must be specified if the labels should be interpreted as strings instead of integers

.TP
.B --use_float_as_real_type arg
must be specified if float should be used instead of double as floating point type

.TP
.B --verbosity arg
choose the level of verbosity: full|timing|libsvm|quiet (default: full)

.TP
.B -q, --quiet
quiet mode (no outputs regardless the provided verbosity level!)

.TP
.B -h, --help
print this helper message

.TP
.B -v, --version
print version information

.TP
.B --input input_file
the file containing the data points that should be scaled

.TP
.B --scaled scaled_file
the file containing the scaled data points

.SH EXIT STATUS
EXIT_SUCCESS
.RS
if OK,
.RE
EXIT_FAILURE
.RS
if any error occurred (e.g., the input file couldn't be opened).

.SH SEE ALSO
plssvm-train(1)
plssvm-predict(1)

.SH AUTHOR
Alexander Van Craen <Alexander.Van-Craen@ipvs.uni-stuttgart.de>
.br
Marcel Breyer <Marcel.Breyer@ipvs.uni-stuttgart.de>