soaxelbrooke/phrase

Issues

• Add start and end indexes for detected phrases

  #23 opened 3 years ago by soaxelbrooke
  0

• Duplicated tokens when using MAX_NGRAM > 10

  #22 opened 5 years ago by soaxelbrooke
  1

• Try Different PMI Estimate

  #21 opened 5 years ago by soaxelbrooke
  0

• Saturates memory on large corpus

  #18 opened 6 years ago by jjhbw
  3

• Ignore numbers and emails

  #15 opened 6 years ago by soaxelbrooke
  0

• Make label names more permissive

  #17 opened 6 years ago by soaxelbrooke
  1

• Add /count API endpoint

  #14 opened 6 years ago by soaxelbrooke
  2

• Add fit-transform style CLI option

  #13 opened 6 years ago by soaxelbrooke
  0

• Performance Regression

  #12 opened 6 years ago by soaxelbrooke
  1

• Accept multiple labels per document for counting and labeling

  #3 opened 6 years ago by soaxelbrooke
  3

• Counting halts for ~30 minutes during counting

  #10 opened 6 years ago by soaxelbrooke
  0

• Calculate stems during export

  #11 opened 6 years ago by soaxelbrooke
  1

• Make thread per label being counted

  #8 opened 6 years ago by soaxelbrooke
  0

• Add stemming and canonical form mapping

  #4 opened 6 years ago by soaxelbrooke
  0

• Keep ngram counts in sorted order

  #9 opened 6 years ago by soaxelbrooke
  0

• Add "transform" subcommand

  #5 opened 6 years ago by soaxelbrooke
  0

• Implement vocab pruning

  #7 opened 6 years ago by soaxelbrooke
  0

• Add environment variables for parameters

  #6 opened 6 years ago by soaxelbrooke
  0

• `phrase` doesn't print usage

  #1 opened 6 years ago by soaxelbrooke
  0

• Doesn't create data dir if it doesn't exist (fails instead)

  #2 opened 6 years ago by soaxelbrooke
  0