Mercurial > hg > python
diff hmm/tinySup.py @ 3:26d9c0308fcf
updated/added from ecclerig version
| author | Henry S. Thompson <ht@inf.ed.ac.uk> |
|---|---|
| date | Mon, 09 Mar 2020 17:35:28 +0000 |
| parents | e07789816ca5 |
| children |
line wrap: on
line diff
--- a/hmm/tinySup.py Mon Mar 09 16:48:09 2020 +0000 +++ b/hmm/tinySup.py Mon Mar 09 17:35:28 2020 +0000 @@ -12,9 +12,9 @@ [('<s>','<s>'),('run','V'),('the','D'),('sheep','N'),('</s>','</s>')]] taglists=[('<s>',[('<s>',1),('the',0),('sheep',0),('run',0),('</s>',0)]), - ('D',[('the',1),('sheep',0),('run',0),('<s>',0),('</s>',0)]), - ('N',[('the',0),('sheep',.5),('run',.5),('<s>',0),('</s>',0)]), - ('V',[('the',0),('sheep',.5),('run',.5),('<s>',0),('</s>',0)]), + ('D',[('the',.8),('sheep',.1),('run',.1),('<s>',0),('</s>',0)]), + ('N',[('the',.2),('sheep',.4),('run',.4),('<s>',0),('</s>',0)]), + ('V',[('the',.2),('sheep',.4),('run',.4),('<s>',0),('</s>',0)]), ('</s>',[('<s>',0),('the',0),('sheep',0),('run',0),('</s>',1)])] tagdict=dict((k,MLEProbDist(FreqDist(dict(v)))) for k,v in taglists) @@ -48,7 +48,7 @@ # Note that contrary to naive reading of the documentation, # train_unsupervised expects a sequence of sequences of word/tag pairs, # it just ignores the tags -nnm=nm.train_unsupervised(sents,model=model,max_iterations=10,updateOutputs=False) +nnm=nm.train_unsupervised(sents,model=model,max_iterations=15,updateOutputs=False) for tag in tagset: if tag=='</s>':