3

I want to do topic modelling , So, I ran the below command :- 

bin\mallet train-topics --input web.mallet --output-state output-file.gz

It tells me :- Topic modeling currently only supports feature sequences: use --keep-sequence option when importing data.

When I add --keep-sequence my command becomes :-

bin\mallet train-topics --input web.mallet --output-state output-file.gz --keep-sequence

This gives below error :-

Exception in thread "main" java.lang.IllegalArgumentException: Unrecognized option 4: --keep-sequence
        at cc.mallet.util.CommandOption$List.process(CommandOption.java:345)
        at cc.mallet.util.CommandOption.process(CommandOption.java:146)
        at cc.mallet.topics.tui.Vectors2Topics.main(Vectors2Topics.java:201)

What should I do? I am running this first time. Please help.

shahrukh
  • 73
  • 5

1 Answers1

1

The train-topics command is for running a model on a previously imported Mallet data file. The --keep-sequence option is used for the data import phase, using the import-file or import-dir command.

David Mimno
  • 1,836
  • 7
  • 7