edit

lm.lua

lm.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.

Data options

  • <mode> (accepted: score, sample)
    'score' apply lm to input text, 'sample' samples output based on input text.
  • -src <string> (required)
    Source sequences to sample/score.
  • -output <string> (default: output.txt)
    Output file depend on <mode>.

LM options

  • -model <string> (required)
    Path to the serialized model file.
  • -batch_size <number> (default: 30)
    Batch size.
  • -max_length <number> (default: 100)
    Maximal length of sentences in sample mode.
  • -temperature <number> (default: 1)
    For sample mode, higher temperatures cause the model to take more chances and increase diversity of results, but at a cost of more mistakes.

Cuda options

  • -gpuid <table> (default: 0)
    List of GPU identifiers (1-indexed). CPU is used when set to 0.
  • -fallback_to_cpu [<boolean>] (default: false)
    If GPU can't be used, rollback on the CPU.
  • -fp16 [<boolean>] (default: false)
    Use half-precision float on GPU.
  • -no_nccl [<boolean>] (default: false)
    Disable usage of nccl in parallel mode.

Logger options

  • -log_file <string> (default: '')
    Output logs to a file under this path instead of stdout.
  • -disable_logs [<boolean>] (default: false)
    If set, output nothing.
  • -log_level <string> (accepted: DEBUG, INFO, WARNING, ERROR; default: INFO)
    Output logs at this level and above.

Other options

  • -time [<boolean>] (default: false)
    Measure average translation time.