gfsmtrain - EXPERIMENTAL: count successful paths for training string pairs in a transducer
gfsmtrain [OPTIONS] PAIR_FILE(s)...
Arguments:
PAIR_FILE(s)... Input string-pair file(s)
Options
-h --help Print help and exit.
-V --version Print version and exit.
-iLABELS --ilabels=LABELS Specify input (lower) labels file.
-oLABELS --olabels=LABELS Specify output (upper) labels file.
-lLABELS --labels=LABELS Set -i and -o labels simultaneously.
-a --att-mode Parse string(s) in AT&T-compatible mode.
-q --quiet Suppress warnings about undefined symbols.
-u --utf8 Assume UTF-8 encoded alphabet and input.
-B --best Only consider cost-minimal path(s) for each training pair.
-O --ordered Count permutations in arc-order as multiple paths.
-P --distribute-by-path Distribute pair-mass over multiple paths.
-A --distribute-by-arc Distribute path-mass over arcs.
-fFSTFILE --fst=FSTFILE Transducer to apply (required).
-zLEVEL --compress=LEVEL Specify compression level of output file.
-FFILE --output=FILE Specifiy output file (default=stdout).
EXPERIMENTAL: count successful paths for training string pairs in a transducer
PAIR_FILE(s)...
Input string-pair file(s)
One pair per line, TAB-separated.
--help
, -h
Print help and exit.
Default: '0'
--version
, -V
Print version and exit.
Default: '0'
--ilabels=LABELS
, -iLABELS
Specify input (lower) labels file.
Default: 'NULL'
--olabels=LABELS
, -oLABELS
Specify output (upper) labels file.
Default: 'NULL'
--labels=LABELS
, -lLABELS
Set -i and -o labels simultaneously.
Default: 'NULL'
--att-mode
, -a
Parse string(s) in AT&T-compatible mode.
Default: '0'
--quiet
, -q
Suppress warnings about undefined symbols.
Default: '0'
--utf8
, -u
Assume UTF-8 encoded alphabet and input.
Default: '0'
--best
, -B
Only consider cost-minimal path(s) for each training pair.
Default: '0'
If specified and true, only minimal-cost path(s) will be considered for each training pair, otherwise all successful paths will be considered.
--ordered
, -O
Count permutations in arc-order as multiple paths.
Default: '0'
If unspecified or false, only unique successful paths modulo arc-ordering will be considered; e.g. (q --[<epsilon>:a]--> q --[a:<epsilon>]--> q) and (q --[a:<epsilon>]--> q --[<epsilon>:a]--> q) are duplicates in this sense, since they differ only in the ordering of the arcs.
--distribute-by-path
, -P
Distribute pair-mass over multiple paths.
Default: '0'
If true, a total count-mass of 1 will be added for each (input,output) pair, and distributed uniformly among any successful paths for that pair. Otherwise, each successful path for a given pair will receive a count-mass of 1 (one).
--distribute-by-arc
, -A
Distribute path-mass over arcs.
Default: '0'
If true, the total count-mass added to each successful path will be distributed uniformly over all its arcs and its final weight. Otherwise, each arc in the path will receive the full count-mass alotted to that path.
--fst=FSTFILE
, -fFSTFILE
Transducer to apply (required).
Default: 'NULL'
--compress=LEVEL
, -zLEVEL
Specify compression level of output file.
Default: '-1'
Specify zlib compression level of output file. -1 (default) indicates the default compression level, 0 (zero) indicates no zlib compression at all, and 9 indicates the best possible compression.
--output=FILE
, -FFILE
Specifiy output file (default=stdout).
Default: '-'
Documentation file auto-generated by optgen.perl version 0.07 using Getopt::Gen version 0.14. Translation was initiated as:
optgen.perl -l --no-handle-rcfile --nocfile --nohfile --notimestamp -F gfsmtrain gfsmtrain.gog
No negative-cost epsilon cycles are allowed in the transducer.
Perl by Larry Wall.
Getopt::Gen by Bryan Jurish.
Bryan Jurish <moocow.bovine@gmail.com>