.TH fst-train 1 "October 2005" "" "fst-train"
.SH NAME
fst-train \- learning transducer weights
.SH SYNOPSIS
.B fst-train [ options ]
.I file
[
.I input-file
]
.SH OPTIONS
.TP
.B \-t file
use multiple transducers in the same way as 
.B fst-infl2.
.TP
.B \-b
This option is used for supervised training with disambiguated data.
.TP
.B \-d
Disambiguate the analyses symbolically as described in the man pages
of 
.B fst-infl2.
.TP
.B \-q
quiet mode
.SH DESCRIPTION
.I fst-train
is used to learn statistical weights for the transducers transitions
based on training data. Training is either unsupervised (default) or
supervised (option -b).
.br
In supervised mode, the input contains fully disambiguated data with
the surface and the analysis form. The format restrictions are
identical to those applying for lexicon entries, i.e. all operators
other than the colon operator (:) are interpreted literally.
.br
In unsupervised mode, the input data consists of surface strings. The
format is identical to the input format of 
.I fst-infl
and
.I fst-infl2.
.br
The transducer weights are stored in files whose names are obtained by
appending 
.I .prob
to the names of the transducer files.
.SH BUGS
No bugs are known so far.
.SH "SEE ALSO"
fst-infl2, fst-compiler
.SH AUTHOR
Helmut Schmid,
Institute for Computational Linguistics,
University of Stuttgart,
Email: schmid@ims.uni-stuttgart.de,
This software is available under the GNU Public License.