OriMorph is a morphological analyzer and lemmatizer for Odia (Oriya). It only analyzes nouns right now.
OriMorph requires that the Foma binaries (specifically, flookup
) be in your path. It expects the following files to be in the current directory:
fsinter.py
-- Python interface toflookup
orimorph.py
-- high-level interface toori.fst
ori.fst
-- a compiled Foma FST. This can actually be anywhere but you need to supply the path.
Usage:
>>> import orimorph
>>> m = orimorph.OriMorph('ori.fst')
>>> m.analyses(['ଓଡ଼ିଶାର', 'ଅନୁଭୂତିଗୁଡିକୁ', 'ଆକାଶରେ'])
[['ଓଡ଼ିଶା', 'SG', 'GEN'], ['ଅନୁଭୂତିଗୁଡି', 'SG', 'ACC'], ['ଆକାଶ', 'SG', 'INSTR']]
>>> m.lemmas(['ଓଡ଼ିଶାର', 'ଅନୁଭୂତିଗୁଡିକୁ', 'ଆକାଶରେ'])
['ଓଡ଼ିଶା', 'ଅନୁଭୂତିଗୁଡି', 'ଆକାଶ']