OpenGrm NGram Library Suggestions

  1. Improve memory usage of ngramread - uses 14 gb for 1.5gb input (Brian)
  2. Methods for providing probability mass to OOV, e.g., use good-turing
  3. counting from cyclic (Cyril)
  4. shift VectorFst to MutableFst where possible (Cyril)
  5. create full purpose, efficient ngramrandgen (Michael) - done.
  6. change ngramrandcorput to output a far rather than text. - done (in ngramrandgen)
  7. support for printing pair LM in a way that facilitates building of transducer

-- MichaelRiley - 05 Nov 2010


This topic: GRM > WebHome > NGramLibrary > NGramSuggests
Topic revision: r15 - 2011-12-12 - BrianRoark
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback