def getSummarizeFeed():
threadId = request.args['key']
cur = mysql.connection.cursor()
query = "SELECT * FROM mails1 WHERE thread_no = " + threadId + ";"
result1 = cur.execute(query)
mails = cur.fetchall()
# print("query : " + query)
cur.close()
subject = ""
sents = []
subject = mails[0]['subject']
for mail in mails:
sents.append(mail['content'])
# print(sents)
ob = summarizer.Summarizer(subject, sents)
data = ob.generate_summary()
return data
import summarizer summary = summarizer.Summarizer() summary.generate_summaries()
import summarizer
import summary_generator as sg
import rouge
# Fissare le variabili
body = "rouge-data/body/"
len = "rouge-data/len.txt"
destination = "rouge-data/all_systems_new/"
result = "lda_result-bis.txt"
rouge_script = "rouge-data/ROUGE-1.5.5.pl"
data_rouge = "rouge-data/data/"
summary_dest = "rouge-data/all_systems_n_bis/"
gold = "rouge-data/models/"
# Preparare il summarizer
s = summarizer.Summarizer(model_path="C:/enwiki_20161220_skip_300.bin")
# Preparare il loop
num_topic = [2, 3, 4, 5, 6, 7, 8, 9]
num_words = [2, 3, 4, 5, 6, 7, 8, 9]
print "Start summarizing..."
for topic in num_topic:
for word in num_words:
new_dir = 'topic_' + str(topic) + "_word_" + str(word)
print new_dir
destination_path = destination + new_dir
generator = sg.SummaryGenerator(body_dir_path=body,
target_length_path=len,
destination_path=destination_path)
model_path = 'C:/Users/Peppo/Desktop/w2vm/enwiki_20161220_skip_300.bin'
summary_destination_root = 'C:/grid-search/'
script_path = 'C:/Users/Peppo/Desktop/w2vm/rouge4MultiLing/rouge/ROUGE-1.5.5.pl'
data_path = 'C:/Users/Peppo/Desktop/w2vm/rouge4MultiLing/rouge/data'
gold_standard_path = 'C:/training/summary/'
results_path = 'grid-search-results.txt'
training_body_path = 'C:/training/body/'
training_length_path = 'C:/training/length.txt'
# Set your ranges
tfidf_values = [0.2, 0.25, 0.3, 0.35]
redundancy_values = [0.8, 0.85, 0.9, 0.95]
# Init my summarizer
print 'Loading model...'
s = s.Summarizer(model_path=model_path)
print 'done\n'
# Generate summaries for all parameters combinations
for tfidf_value in tfidf_values:
print 'Summarization with tdidf ' + str(tfidf_value) + '...'
for redundancy_value in redundancy_values:
# Update summarizer's configuration
tfidf_str = "".join(str(tfidf_value).split("."))
redundancy_str = "".join(str(redundancy_value).split("."))
# Generate new name for the destination directory
new_dir = 'tfidf_' + str(tfidf_str) + "_redundancy_" + str(
redundancy_str)
destination_path = summary_destination_root + new_dir
# coding=utf-8
from rest_framework import generics
from rest_framework import mixins
import json
from django.http import JsonResponse
import summarizer
s = summarizer.Summarizer(
model_path=
"C:/Users/Gianni Mastroscianni/Desktop/Magistrale/Accesso Intelligente all'Informazione ed Elaborazione del Linguaggio Naturale/Progetto/word2vec_models/enwiki_20161220_skip_300.bin"
)
class Summary(generics.GenericAPIView, mixins.CreateModelMixin):
def post(self, request):
# load json
data = json.loads(request.body)
list = []
redundancy_threshold = data['redundancy_threshold']
tfidf = data['tfidf_threshold']
summary_length = data['summary_length']
query_based_token = data['query_based_token']
# print (query_based_token, "query token")
s.set_tfidf_threshold(tfidf)
s.set_redundancy_threshold(redundancy_threshold)
import summarizer SUMMARY_LENGTH = 7 input_file = "input_file.txt" # "reason_deep_learning_csv.txt" s = summarizer.Summarizer() s.set_factors(10, 10, 10) summary = s.generate_summary(input_file, SUMMARY_LENGTH) print summary
import newsProvider
import summarizer
if __name__ == '__main__':
newsProvider = newsProvider.NewsProvider("Apple stock", 5)
news = newsProvider.getNews()
summary = {}
for source, text in news.items():
summarize = summarizer.Summarizer(text, source,5)
print(source, summarize.summarize())
roseType: 'radius',
animationType: 'scale',
animationEasing: 'elasticOut'
}
]
});
''')
return indent(doc.getvalue())
while True:
print('----------------------------------------')
with open(input("please choose input file:")) as file:
reviews = '\n\n'.join(item['review'] for item in json.load(file))
print('parsing text ...')
s = summarizer.Summarizer(reviews).summary()
for i, f in enumerate(s):
print(
str(i + 1) + '.', '[' + f.lemma + ']',
str(len(f.positive)) + '/' + str(len(f.negative)))
print('\tpositive:')
for n in f.positive[:3]:
print('\t', n.sent)
print('\tnegative:')
for n in f.negative[:3]:
print('\t', n.sent)
with open('summary.html', 'w') as file:
def generate_summary():
summary = summarizer.Summarizer()
summary.generate_summaries()
return ""