def test_person_summary():
s = Summarizer()
people = [
{
"gender": "F",
"image": "https://example.com/image1",
"party": [{"name": "Democratic"}, {"name": "Democratic", "end_date": "1990"}],
},
{
"gender": "F",
"image": "https://example.com/image2",
"party": [{"name": "Democratic"}, {"name": "Working Families"}],
"extras": {"religion": "Zoroastrian"},
"contact_details": [{"fax": "123-456-7890", "note": "Capitol Office"}],
"other_identifiers": [{"scheme": "fake", "identifier": "abc"}],
"ids": {"twitter": "fake"},
},
{
"gender": "M",
"image": "https://example.com/image3",
"party": [{"name": "Republican"}],
"contact_details": [{"phone": "123-456-7890", "note": "Capitol Office"}],
"other_identifiers": [{"scheme": "fake", "identifier": "123"}],
},
]
for p in people:
s.summarize(p)
assert s.parties == {"Republican": 1, "Democratic": 2, "Working Families": 1}
assert s.contact_counts == {"Capitol Office phone": 1, "Capitol Office fax": 1}
assert s.id_counts == {"fake": 2, "twitter": 1}
assert s.optional_fields == {"gender": 3, "image": 3}
assert s.extra_counts == {"religion": 1}
def test_summarize_multipe(self):
args = '{"output_ids": [], \
"ppg_ids": ["DEF", "LTFL", "LP61"], \
"problem_objective_ids": ["FGH"], \
"goal_ids": ["EFG"], \
"operation_ids": ["BEN", "LISA"], \
"report_type": "Mid Year Report", \
"year": 2013 }'
s = Summarizer(env = 'test', args = args)
summary = s.summarize()
assert summary.strip() == 'the quick brown fox. the second quick brown fox.'
def test_summarize_multipe(self):
args = '{"output_ids": [], \
"ppg_ids": ["DEF", "LTFL", "LP61"], \
"problem_objective_ids": ["FGH"], \
"goal_ids": ["EFG"], \
"operation_ids": ["BEN", "LISA"], \
"report_type": "Mid Year Report", \
"year": 2013 }'
s = Summarizer(env='test', args=args)
summary = s.summarize()
assert summary.strip(
) == 'the quick brown fox. the second quick brown fox.'
def test_summarize_large(self):
args = '{"output_ids": [], \
"ppg_ids": ["DEF", "LTFL", "LP61"], \
"problem_objective_ids": ["FGH"], \
"goal_ids": ["EFG"], \
"operation_ids": ["BEN", "LISA", "JEFF"], \
"report_type": "Mid Year Report", \
"year": 2013 }'
max_chars = 500
s = Summarizer(env = 'test', args = args, max_chars = max_chars)
summary = s.summarize()
assert len(summary) <= max_chars
def test_summarize_large(self):
args = '{"output_ids": [], \
"ppg_ids": ["DEF", "LTFL", "LP61"], \
"problem_objective_ids": ["FGH"], \
"goal_ids": ["EFG"], \
"operation_ids": ["BEN", "LISA", "JEFF"], \
"report_type": "Mid Year Report", \
"year": 2013 }'
max_chars = 500
s = Summarizer(env='test', args=args, max_chars=max_chars)
summary = s.summarize()
assert len(summary) <= max_chars
import pickle
from pathlib import Path
from summa import summarizer
from summarize import Summarizer
if __name__ == '__main__':
DATA_PATH = Path('data')
reviewtext = pd.read_pickle(DATA_PATH/'df_processed_reviews.p')
idx2sent = pickle.load(open(DATA_PATH/'idx2sent.p', 'rb'))
sent2vec = pickle.load(open(DATA_PATH/'sent2vec.p', 'rb'))
idx = 198
my_summarizer = Summarizer(reviewtext, idx2sent, sent2vec)
review, my_summary = my_summarizer.summarize(idx, n=100)
summary = summarizer.summarize(review, ratio=0.2)
