spaCy/spacy/lang/nb/lemmatizer/__init__.py
2019-08-20 17:36:34 +02:00

42 lines
1.2 KiB
Python

# coding: utf8
# structure copied from the English lemmatizer
from __future__ import unicode_literals
from pathlib import Path
from .lookup import LOOKUP
from ._adverbs_wordforms import ADVERBS_WORDFORMS
from ._lemma_rules import ADJECTIVE_RULES, NOUN_RULES, VERB_RULES, PUNCT_RULES
from ....util import load_language_data
ADJECTIVES = set()
ADVERBS = set()
NOUNS = set()
VERBS = set()
LEMMA_INDEX = {"adj": ADJECTIVES, "adv": ADVERBS, "noun": NOUNS, "verb": VERBS}
BASE_PATH = Path(__file__).parent
LEMMA_EXC = {
"adj": load_language_data(BASE_PATH / "_adjectives_wordforms.json"),
"adv": ADVERBS_WORDFORMS,
"noun": load_language_data(BASE_PATH / "_nouns_wordforms.json"),
"verb": load_language_data(BASE_PATH / "_verbs_wordforms.json"),
}
LEMMA_RULES = {
"adj": ADJECTIVE_RULES,
"noun": NOUN_RULES,
"verb": VERB_RULES,
"punct": PUNCT_RULES,
}
# Note on noun wordforms / lemmas:
# All wordforms are extracted from Norsk Ordbank in Norwegian Bokmål 2005, updated 20180627
# (CLARINO NB - Språkbanken), Nasjonalbiblioteket, Norway:
# https://www.nb.no/sprakbanken/show?serial=oai%3Anb.no%3Asbr-5&lang=en
# License:
# Creative_Commons-BY (CC-BY) (https://creativecommons.org/licenses/by/4.0/)