def __init__(self):
self.gen = Generator()
self.cleanser = TextCleanser()
gen = self.gen
cln = self.cleanser
self.cleanse_methods = {
gen.IBM_SIM: cln.heuristic_cleanse,
gen.SSK_SIM: cln.ssk_cleanse,
gen.PHONETIC_ED_SIM: cln.phonetic_ED_cleanse
}
def __init__(self):
self.cleanser = TextCleanser()
cln = self.cleanser
self.cleanse_methods = {
'IBM': cln.ibm_cleanse,
'SSK': cln.ssk_cleanse,
'PHONETIC_ED': cln.phonetic_cleanse
}
self.gold_sent_clean = []
self.gold_word_pairs = []
self.gold_sent_pairs = []
class CleanserWebService():
def __init__(self):
self.tc = TextCleanser()
@expose
def clean(self, text):
cleantext, error, replacements = self.tc.ssk_cleanse(text)
if error == "":
return cleantext
else:
# an error occurred
return error
class CleanserWebService():
def __init__(self):
self.tc = TextCleanser()
@expose
def clean(self, text):
cleantext, error, replacements = self.tc.ssk_cleanse(text)
if error=="":
return cleantext
else:
# an error occurred
return error
def __init__(self):
self.tc = TextCleanser()
as input on stdin and outputs normalised strings on stdout.
Author: Stephan Gouws
Contact: [email protected]
"""
from cleanser import TextCleanser
import json
import codecs
import getopt
import sys, time
from random import choice
if __name__ == '__main__':
# print "Noisy text cleanser"
clnsr = TextCleanser()
text = sys.stdin.readline()
while (text):
if len(text) <= 1:
break
# cleantext,error,replacements = clnsr.heuristic_cleanse(text, gen_off_by_ones=False)
# to use a phonetic edit-distance based similarity function, use the
# method below:
# cleantext,error,replacements = clnsr.phonetic_ED_cleanse(text, gen_off_by_ones=False)
# to use SSK-based cleanser, use
cleantext, error, replacements = clnsr.ssk_cleanse(
text, gen_off_by_ones=False)
if error == "ERROR":
sys.stderr.write("ERROR")
def __init__(self):
self.tc = TextCleanser()
as input on stdin and outputs normalised strings on stdout.
Author: Stephan Gouws
Contact: [email protected]
"""
from cleanser import TextCleanser
import json
import codecs
import getopt
import sys, time
from random import choice
if __name__ == '__main__':
# print "Noisy text cleanser"
clnsr = TextCleanser()
text=sys.stdin.readline()
while (text):
if len(text)<=1:
break
# cleantext,error,replacements = clnsr.heuristic_cleanse(text, gen_off_by_ones=False)
# to use a phonetic edit-distance based similarity function, use the
# method below:
# cleantext,error,replacements = clnsr.phonetic_ED_cleanse(text, gen_off_by_ones=False)
# to use SSK-based cleanser, use
cleantext,error,replacements = clnsr.ssk_cleanse(text, gen_off_by_ones=False)
if error=="ERROR":
sys.stderr.write("ERROR")
continue
