from flair.models import SequenceTagger
from flair.data import Sentence, Token
import requests
from simstring.feature_extractor.character_ngram import CharacterNgramFeatureExtractor
from simstring.measure.jaccard import JaccardMeasure
from simstring.feature_extractor.word_ngram import WordNgramFeatureExtractor
from simstring.measure.cosine import CosineMeasure
from simstring.database.dict import DictDatabase
from simstring.searcher import Searcher
from flair.data_fetcher import NLPTaskDataFetcher
LOCATION_WIKIPEDIA_DISAMBIGUATION = "../wikidisambiguationpages.txt"
ag = Agdistis()
ag.agdistisApi = "http://localhost:8080/AGDISTIS"
def load_disambiguation():
db = DictDatabase(WordNgramFeatureExtractor(2))
with open(LOCATION_WIKIPEDIA_DISAMBIGUATION) as disambig_file:
for line in disambig_file:
r = line.replace("_(disambiguation)", "").replace("_", " ").lower()
db.add(r.strip())
return Searcher(db, JaccardMeasure())
def process_conll_doc(input_file_name, output_file_name, ner_model,
def __init__(self):
self.agdistis = Agdistis()
self.fox = Fox()
