0

I want to use topic modeling and found MALLET suitable for me.
I successfully created my first demo using some 0.1 million Documents.Now as per my requirements i have to deal with 10 million documents for which am not able to processed further.Is it possible to add new documents to existing topic model or something like create two models and merge into single model and get the output by merging all the models because mallet is not able to handle such large documents in one go for which am thinking to batch the models and get the output by merging all the documents
Say for example i will make 100 batch of 0.1 million documents and run mallet on each batch and at last get result by merging all 100 batches

Thanks

Hardik Dobariya
  • 339
  • 2
  • 4
  • 20

1 Answers1

0

I dont think so this is possible with Mallet. I dont think once you have the model created, you can incrementally add new documents to the trained model and have it re-trained.

I will wait for someone to either support or refute my answer.

London guy
  • 27,522
  • 44
  • 121
  • 179