edit

tools/tokenize.lua

tokenize.lua options:

  • -h [<boolean>] (default: false)
    This help.
  • -md [<boolean>] (default: false)
    Dump help in Markdown format.
  • -config <string> (default: '')
    Load options from this file.
  • -save_config <string> (default: '')
    Save options to this file.

Tokenizer options

  • -mode <string> (accepted: conservative, aggressive; default: conservative)
    Define how aggressive should the tokenization be. aggressive only keeps sequences of letters/numbers, conservative allows a mix of alphanumeric as in: "2,000", "E65", "soft-landing", etc.
  • -joiner_annotate [<boolean>] (default: false)
    Include joiner annotation using -joiner character.
  • -joiner <string> (default: )
    Character used to annotate joiners.
  • -joiner_new [<boolean>] (default: false)
    In -joiner_annotate mode, -joiner is an independent token.
  • -case_feature [<boolean>] (default: false)
    Generate case feature.
  • -segment_case [<boolean>] (default: false)
    Segment case feature, splits AbC to Ab C to be able to restore case
  • -segment_alphabet <table> (accepted: Tagalog, Hanunoo, Limbu, Yi, Hebrew, Latin, Devanagari, Thaana, Lao, Sinhala, Georgian, Kannada, Cherokee, Kanbun, Buhid, Malayalam, Han, Thai, Katakana, Telugu, Greek, Myanmar, Armenian, Hangul, Cyrillic, Ethiopic, Tagbanwa, Gurmukhi, Ogham, Khmer, Arabic, Oriya, Hiragana, Mongolian, Kangxi, Syriac, Gujarati, Braille, Bengali, Tamil, Bopomofo, Tibetan)
    Segment all letters from indicated alphabet.
  • -segment_alphabet_change [<boolean>] (default: false)
    Segment if alphabet change between 2 letters.
  • -bpe_model <string> (default: '')
    Apply Byte Pair Encoding if the BPE model path is given. If the option is used, BPE related options will be overridden/set automatically if the BPE model specified by -bpe_model is learnt using learn_bpe.lua.
  • -EOT_marker <string> (default: </w>)
    Marker used to mark the end of token.
  • -BOT_marker <string> (default: <w>)
    Marker used to mark the beginning of token.
  • -bpe_case_insensitive [<boolean>] (default: false)
    Apply BPE internally in lowercase, but still output the truecase units. This option will be overridden/set automatically if the BPE model specified by -bpe_model is learnt using learn_bpe.lua.
  • -bpe_mode <string> (accepted: suffix, prefix, both, none; default: suffix)
    Define the BPE mode. This option will be overridden/set automatically if the BPE model specified by -bpe_model is learnt using learn_bpe.lua. prefix: append -BOT_marker to the begining of each word to learn prefix-oriented pair statistics; suffix: append -EOT_marker to the end of each word to learn suffix-oriented pair statistics, as in the original Python script; both: suffix and prefix; none: no suffix nor prefix.

Other options

  • -nparallel <number> (default: 1)
    Number of parallel thread to run the tokenization
  • -batchsize <number> (default: 1000)
    Size of each parallel batch - you should not change except if low memory