def test_en_tagger_lemma_base_forms(NLP):
doc = NLP.make_doc("dive")
doc[0].morph = MorphAnalysis(NLP.vocab, "Number=Sing")
doc[0].pos_ = "NOUN"
NLP.get_pipe("lemmatizer")(doc)
assert doc[0].lemma_ == "dive"
doc = NLP.make_doc("diva")
doc[0].morph = MorphAnalysis(NLP.vocab, "Number=Plur")
doc[0].pos_ = "NOUN"
NLP.get_pipe("lemmatizer")(doc)
assert doc[0].lemma_ == "diva"
def test_en_tagger_lemma_nouns(NLP, text, lemmas, morphology):
# Cases like this are problematic - not clear what we should do to resolve
# ambiguity? ("axes", ["ax", "axes", "axis"])
# noun_index = lemmatizer.index["noun"]
# noun_exc = lemmatizer.exc["noun"]
doc = NLP.make_doc(text)
doc[0].morph = MorphAnalysis(NLP.vocab, morphology)
doc[0].pos_ = "NOUN"
NLP.get_pipe("lemmatizer")(doc)
assert doc[0].lemma_ == lemmas[0]
def test_en_tagger_lemma_verbs(NLP, text, lemmas):
doc = NLP.make_doc(text)
doc[0].morph = MorphAnalysis(NLP.vocab, "VerbForm=Inf")
doc[0].pos_ = "VERB"
NLP.get_pipe("lemmatizer")(doc)
assert doc[0].lemma_ == lemmas[0]
def test_en_tagger_lemma_issue781(NLP, word, lemmas):
doc = NLP.make_doc(word)
doc[0].pos_ = "NOUN"
doc[0].morph = MorphAnalysis(NLP.vocab, "Number=Plur")
NLP.get_pipe("lemmatizer")(doc)
assert doc[0].lemma_ == lemmas[0]
def test_en_tagger_lemma_base_form_verb(NLP):
doc = NLP.make_doc("saw")
doc[0].morph = MorphAnalysis(NLP.vocab, "VerbForm=Past")
doc[0].pos_ = "VERB"
NLP.get_pipe("lemmatizer")(doc)
assert doc[0].lemma_ == "see"