Hiya everyone,
I'm trying to retrain the af-nl pos-tagger since i'm attempting to fix
some things by using .tsx rules, but it isn't going very well. I get
this error:
pim-oneiric@pim-K53SV:~/source/apertium-af-nl$ apertium-tagger -t 1
dev/apertium-af-nl.af.exp af-tagger-data/af.smaller.crp
apertium-af-nl.af.tsx af-nl.prob
Calculating ambiguity classes...
64 states and 65 ambiguity classes
Kupiec's initialization of transition and emission probabilities...
Error: A new ambiguity class was found. I cannot continue.
Word 'van' not found in the dictionary.
New ambiguity class: {NOUN,ADV,PR}
Take a look at the dictionary and at the training corpus. Then, retrain.
However, 'van' is in the dictionary:
pim-oneiric@pim-K53SV:~/source/apertium-af-nl$ cat
dev/apertium-af-nl.af.exp | grep ^van:
van:van<pr>
van:van<adv>
van:van<n><sg>
van:>:van<n><sg>
van:>:van<n><sg>
van:<:van<n><sg>
van:<:van<n><sg>
Relevant files:
dev/apertium-af-nl.af.exp can be created with "lt-expand
apertium-af-nl.af.dix | grep -v '<cmp>' | grep -v '<compound' >
dev/apertium-af-nl.exp"
first 500 lines are at http://ilazki.thinkgeek.co.uk/~aureianimus/af.smaller.exp
af-tagger-data/af.smaller.crp is at
http://ilazki.thinkgeek.co.uk/~aureianimus/af.smaller.crp
apertium-af-nl.af.tsx and af-nl.prob are in svn
https://apertium.svn.sourceforge.net/svnroot/apertium/staging/apertium-af-nl/
Any ideas will be greatly apreciated.
Regards,
Pim (/aureianimus)
------------------------------------------------------------------------------
Cloud Computing - Latest Buzzword or a Glimpse of the Future?
This paper surveys cloud computing today: What are the benefits?
Why are businesses embracing it? What are its payoffs and pitfalls?
http://www.accelacomm.com/jaw/sdnl/114/51425149/
_______________________________________________
Apertium-stuff mailing list
Apertium-stuff@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
