mayzup: A parser for grammars developed with MAYZ

Japanese version

This is a parser for specialized a grammar and a probabilistic model that are developed using the MAYZ toolkit.

mayzup [options] [-a arguments]
Arguments following "-a" are passed to LiLFeS programs as command-line arguments
Options
-L directoryDirectory of LiLFeS modules
-W numberLimit number of words
-E numberLimit number of edges
-l moduleLoad LiLFeS program
-e commandExecute LiLFeS command
-iInteractive mode (show LiLFeS prompt)
-nNon-interactive mode (default)
-iterWith iterative beam thresholding (default)
-fomWith disambiguation
-nofomWithout disambiguation

This tool loads LiLFeS modules specified with "-l" and executes a parser. If a lilfes command is specified, it is executed. If interactive mode is specified, it prints a command prompt of LiLFeS after executing the lilfes command.

This tool supports only loading of modules and up interfaces. It does nothing about loading of a lexicon and initialization of FOM models. For these, utilize the interfaces of "up" such as "parser_init".

In addition to the function of "up", "mayzup" provides the following predicates. The following predicates assist the usage of probabilistic models in parsing.

init_amis_model(+$ModelName, +$ModelFile)
$ModelNamename of a probabilistic model
$ModelFileparameter file
Makes a new amis model, loads paramters from $ModelFile, and initialize feature masks.
delete_amis_model(+$ModelName)
$ModelNamename of a probabilistic model
Delete a model created by "init_amis_model/2"
amis_event_weight(+$ModelName, +$Category, +$Event, -$FOM)
$ModelNamename of a probabilistic model
$Categoryname of a category
$Eventevent (a list of strings)
$FOMFOM of the event (log probability)
Computes the FOM of the event $Event by summing up weights of features that are obtained by applying features masks to the event.
amis_log_probability(+$ModelName, +$Category, +$EventList, -$FOMList)
$ModelNamename of a probabilistic model
$Categoryname of a category
$EventListlist of events (a list of lists of strings)
$FOMListlist of FOM (log probability) of the events
Computes FOM of each events by "amis_event_weight/4", and returns a list of FOMs by normalizing them.

The other functions are the same as "up". For details, see the manual of UP.


MAYZ Toolkit Manual MAYZ Home Page Tsujii Laboratory
MIYAO Yusuke (yusuke@is.s.u-tokyo.ac.jp)