Leopar main page

Leopar's usual options

Leopar is a parser for Natural Language (more info on leopar.loria.fr)

Usage: leopar [OPTION]

If no mode is specified, the main Graphical User Interface is opened. Other modes are:
  -corpus                      	parse each sentence of a corpus and print small result informations
  -param                        print parameters on stdout
  -save                         save the command line parameters in config file
  -version                      print version number
  
Options are:
  -grammar <string>             choose the grammar file used by leopar
  -lexicon <string>             choose the lexicon file used by lexicomp & toktok
  -toktok_system <string>       choose the rules file used by lexicomp & toktok
  -examples <string>            choose the examples file for the GUI (set to "" to disable)
  -filters <string>             set the list of filters to apply (separated by ';')
  -algo <string>                algorithm used for deep parsing
  -dep <string>                 specify the model representation used (value are: 'complete', 'epsilon' and 'tree')

  -timeout <float>              specify the timeout that computation would not exceed
  -config_path <path>           the path where to find config.xml and dot_config.xml (if needed)
  -config <file>                the config file to use
  -out_dir                      the directory where output are store (to use with -corpus option)

Options documentation

the -aux option

With the -aux option, you can load several auxiliary lexicon into leopar. These auxiliaty lexicons are used by Toktok in addition to the main lexicon. Of course, auxiliary lexicons must be defined a the same domain as the main one.

For example, the small auxiliary lexicon tdm80.lex is use in the parsing of Jules Verne's Novel.

leopar -aux tdm80.zen

If several auxiliary lexicons are required, the syntax is:

leopar -aux "aux1.zen aux2.zen aux3.zen"