In order to train the Catalan part-of-speech tagger yo need to place here
some corpora:
* For supervised training:
- A file called 'ca.tagged' should contain the hand-tagged corpus.
- A file called 'ca.tagged.txt' should contain the same corpus that
'ca.tagged' contains, but in RAW format.
* For unsupervised training:
- A file called 'ca.crp.txt' should contain large corpus in RAW format