-1

I have two datasets: dataset A and dataset B. I want to use CRF++ (mirror) to train a conditional random field (CRF) on dataset A, then train the CRF on dataset B. Is it possible to achieve that with CRF++?

I do not want to train the CRF on two datasets at the same time.

Franck Dernoncourt
  • 77,520
  • 72
  • 342
  • 501
  • Your question makes no sense. Do you want two different models? If you don't want two different models, why do you think combining your datasets is inappropriate? – polm23 Jun 23 '17 at 04:48
  • @polm23 combining isn't good because I want a model that is specialized for one data set – Franck Dernoncourt Jun 23 '17 at 04:54
  • Then just train two different models. `crf_learn templateA trainA modelA` then `crf_learn templateB trainB modelB`. – polm23 Jun 23 '17 at 05:14

1 Answers1

0

I think its absolutely possible to train 2 individual models ( one on dataset A and another on dataset B ).

In case you want only 1 model , then you can combine both datasets and train 1 model.

If I was unable to answer your question, then I'm not sure what you are trying to do. Could you please elaborate your doubt?

user2238884
  • 572
  • 5
  • 12
  • Thanks. What they want to do is: train a conditional random field (CRF) on dataset A, then train the CRF on dataset B. Training the CRF on both data sets at once would achieve a different result. – Franck Dernoncourt May 08 '17 at 17:43