DEMorphy is a morphological analyzer for German language. DEMorphy provides gender, person, singular/plural etc. full inflection information as well as word lemma.
OS X & Linux, directly from Github:
$ git clone https://github.com/DuyguA/DEMorphy
$ cd DEMorphyDownload the dictionary file demorphy/data/words.dg and replace it under the corresponding directory again. Then you're ready to launch the setup script:
$ python setup.py installBasic usage:
$ python>>> from demorphy import Analyzer
>>> analyzer = Analyzer(char_subs_allowed=True)
>>> s = analyzer.analyze(u"gegangen")
>>> for anlyss in s:
print anlyss
{'CATEGORY': u'ADJ', 'PTB_TAG': u'JJ', 'STTS_TAG': u'ADJD', 'ADDITIONAL_ATTRIBUTES': u'<adv>', 'DEGREE': u'pos', 'LEMMA': u'gegangen'}
{'CATEGORY': u'ADJ', 'PTB_TAG': u'JJ', 'STTS_TAG': u'ADJD', 'ADDITIONAL_ATTRIBUTES': u'<pred>', 'DEGREE': u'pos', 'LEMMA': u'gegangen'}
{'CATEGORY': u'V', 'LEMMA': u'gehen', 'STTS_TAG': u'V', 'TENSE': u'ppast', 'PTB_TAG': u'V'}Usage with cache decorators:
>>> from demorphy import Analyzer
>>> from demorphy.cache import memoize, lrudecorator
>>> analyzer = Analyzer(char_subs_allowed=True)
>>> cache_size = 200 #you can arrange the size or unlimited cache. For German lang, we recommed 200 as cache size.
>>> cached = memoize if cache_size=="unlim" else (lrudecorator(cache_size) if cache_size else (lambda x: x))
>>> analyze = cached(analyzer.analyze)
>>> s = analyze(u"gegangen")
>>> for anlyss in s:
print anlyss
{'CATEGORY': u'ADJ', 'PTB_TAG': u'JJ', 'STTS_TAG': u'ADJD', 'ADDITIONAL_ATTRIBUTES': u'<adv>', 'DEGREE': u'pos', 'LEMMA': u'gegangen'}
{'CATEGORY': u'ADJ', 'PTB_TAG': u'JJ', 'STTS_TAG': u'ADJD', 'ADDITIONAL_ATTRIBUTES': u'<pred>', 'DEGREE': u'pos', 'LEMMA': u'gegangen'}
{'CATEGORY': u'V', 'LEMMA': u'gehen', 'STTS_TAG': u'V', 'TENSE': u'ppast', 'PTB_TAG': u'V'}Iterating over all the lexicon:
>>> from demorphy import Analyzer
>>> analyzer = Analyzer(char_subs_allowed=True)
>>> ix = analyzer.iter_lexicon_formatted()
>>> for i in ix:
print(i)One can iterate over the lexicon words with a given prefix. Following code will iterate over all the words that begins with "ge":
>>> ix = analyzer.iter_lexicon_formatted(prefix=u"ge")
>>> for i in ix:
print(i)Altinok, D.: DEMorphy, German Language Analyzer
Berlin, 2018
Links:
- Duygu Altinok