--num_stopwords

Switch

--num_stopwords [num]

Description

Set the number of stop words to filter out of LDA topic estimation. Stop words are automatically set to be the top [num] most frequent words in the unigrams table.

Argument and Default Value

The number of stop words to use. Default: 50.

Details

See the DLATK LDA Interface tutorial for details.

Other Switches

Required Switches:

• -d, -g, -t, -f

• --estimate_lda_topics

• --lda_lexicon_name or --no_lda_lexicon

Optional Switches:

• --mallet_path

• --save_lda_files

• --lda_lexicon_name

• --no_lda_lexicon

• --num_topics

• --no_lda_stopping

• --lda_alpha

• --lda_beta

• --lda_iterations

Example Commands

See the DLATK LDA Interface tutorial for examples.