def extract_and_get_intent(text):
# try:
# text = compound2unicode(text)
# if text.count(" ") == 0:
# text = text + "."
# for k,v in MAPPING_DICT.items():
# text = text.replace(k, v)
print('\nINPUT: ' + text)
extracted = extract(text)
extracted_dict = dict()
for extractEle in extracted:
key = extractEle['key']
value = extractEle['value']
if key not in extracted_dict.keys():
extracted_dict[key] = [value]
else:
if value not in extracted_dict[key]:
extracted_dict[key].append(value)
extracted_dict_keys = extracted_dict.keys()
candidate_intent_dict = get_candidate_intent([text])
for key in DEFINED_INTENTS:
if key in extracted_dict_keys and key in candidate_intent_dict.keys():
candidate_intent_dict.pop(key)
print('\n OUTPUT: INTENT:' + str(candidate_intent_dict))
print('\n OUTPUT: EXTRACT:' + str(extracted_dict))
return extracted, extracted_dict, candidate_intent_dict
def main():
infile = ''
template = ''
outfile = ''
csvFile = False
compFile = False
#argument handleing
try:
optList, args = getopt.getopt(
sys.argv[1:], 'hi:t:ovc',
["ifile=", "templatefile=", "ofile=", "csvfile=", "compoundfile="])
except getopt.GetoptError:
print(
'Usage: BRENDAExtract.py --ifile <inputfile> --templatefile <templatefile> --ofile <outputfile>'
)
sys.exit(2)
for opt, arg in optList:
if opt == '-h':
print(
'Usage: BRENDAExtract.py --ifile <inputfile> --templatefile <templatefile> --ofile <outputfile>'
)
sys.exit()
elif opt in ("-i", "--ifile"):
infile = arg
elif opt in ("-t", "--templatefile"):
template = arg
elif opt in ("-o", "--ofile"):
outfile = arg
elif opt in ("-v", "--csvfile"):
csvFile = arg
elif opt in ("-c", "--compoundfile"):
compFile = arg
q = extract()
input_file = open(infile).read()
input_json = json.loads(input_file)
template_file = open(template).read()
template_json = json.loads(template_file)
q.templateExtract(input_json, outfile, csvFile, compFile, template_json)
def summariser():
reqData = request.get_json()
print(reqData)
res = extract(reqData["url"])
data = res["content"]
t = res["title"]
title = t[0].text
summary = text_rank(data)
res = make_response(jsonify({
"Summary": summary,
"Title": str(title)
}), 200)
return res
def extract_data(train_filename, test_filename, extra_filename):
#extract datasets
train_folder = extract(train_filename)
test_folder = extract(test_filename)
extra_folder = extract(extra_filename)
return train_folder, test_folder, extra_folder