def _annotate_all_tiers(text):
annotated = AnnoDoc(text)
anno_tiers = [
GeonameAnnotator(),
CountAnnotator(),
ResolvedKeywordAnnotator(),
DateAnnotator()
]
for tier in anno_tiers:
annotated.add_tiers(tier)
return annotated
def annotated_example():
dirname = os.path.dirname(__file__)
path = os.path.join(dirname, '..', 'data', 'fixtures', 'annotated_example.pkl')
if not os.path.isfile(path):
from epitator.annotator import AnnoDoc
from epitator.count_annotator import CountAnnotator
annotated = AnnoDoc('I am in Berlin. Here are 5 confirmed cases of influenza. '
'Still, less worse than those 100 confirmed and 200 suspected cases last year.')
annotated.add_tiers(CountAnnotator())
annotated = delete_non_epitator_name_entity_tiers(annotated)
os.makedirs(os.path.dirname(path), exist_ok=True)
with open(path, 'wb') as handel:
pickle.dump(annotated, handel)
else:
with open(path, 'rb') as handel:
annotated = pickle.load(handel)
return annotated
def __init__(self,
classifier,
dict_vectorizer,
cutoff_ratio=0.65,
keyword_array=None):
self.keyword_array = keyword_array
self.classifier = classifier
self.geoname_annotator = GeonameAnnotator()
self.count_annotator = CountAnnotator()
self.date_annotator = DateAnnotator()
self.keyword_annotator = KeywordAnnotator()
self.resolved_keyword_annotator = ResolvedKeywordAnnotator()
processing_pipeline = []
processing_pipeline.append(('link', LinkedKeywordAdder(keyword_array)))
processing_pipeline.append(('limit', LimitCounts(1)))
self.keyword_processor = Pipeline(processing_pipeline)
self.dict_vectorizer = dict_vectorizer
self.keywords = dict_vectorizer.get_feature_names()
self.keyword_extractor = KeywordExtractor(keyword_array)
self.cutoff_ratio = cutoff_ratio
def _annotate(self, text: str, entity: str) -> AnnoDoc:
tier = {"counts": CountAnnotator(), "dates": DateAnnotator()}
annotated = AnnoDoc(text)
annotated.add_tiers(tier[entity])
return annotated
def _annotate(text: str, to_optimize: str) -> AnnoDoc:
tier = {'counts': CountAnnotator(), 'dates': DateAnnotator()}
annotated = AnnoDoc(text)
annotated.add_tiers(tier[to_optimize])
return annotated
def setUp(self):
self.annotator = CountAnnotator()