required=True)
parser.add_argument("--editmodel",
"-e",
type=argparse.FileType('rb'),
required=True)
args = parser.parse_args()
s = SpellChecker(max_distance=2)
s.load_language_model(args.languagemodel)
s.load_channel_model(args.editmodel)
print(s.channel_model.prob("hello", "hello"))
print(s.channel_model.prob("hellp", "hello"))
print(s.channel_model.prob("hllp", "hello"))
print(s.check_text("they did not yb any menas"))
"""
>>> [['they'], ['did'], ['not'], ['by', 'b', 'ye', 'y', 'yo', 'ob', 'ya', 'ab'], ['any'],
>>> ['means', 'mens', 'mena', 'zenas', 'menan', 'mends']]
"""
print(s.autocorrect_line("they did not yb any menas"))
"""
>>> ['they', 'did', 'not', 'by', 'any', 'means']
"""
print(s.suggest_line("they did not yb any menas", max_suggestions=2))
"""
>>> ['they', 'did', 'not', ['by', 'b'], 'any', ['means', 'mens']]
"""
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("--languagemodel",
"-l",
type=argparse.FileType('rb'),
required=True)
parser.add_argument("--editmodel",
"-e",
type=argparse.FileType('rb'),
required=True)
parser.add_argument("--corpus",
"-c",
type=argparse.FileType('r'),
default=sys.stdin)
parser.add_argument("--icorpus", type=argparse.FileType('r'))
args = parser.parse_args()
s = SpellChecker(max_distance=2)
s.load_language_model(args.languagemodel)
s.load_channel_model(args.editmodel)
corpus = args.corpus.readlines()
icorpus = args.icorpus.readlines()
for i in range(len(corpus)):
if icorpus[i] != corpus[i]:
print("LINE: ", corpus[i])
corrected = s.autocorrect_line(corpus[i])
print("CORRECTED: ", corrected)
print("ISPELL: ", icorpus[i])
import argparse
import sys
from SpellCheck import SpellChecker
from LanguageModel import LanguageModel
from EditDistance import EditDistanceFinder
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("--languagemodel",
"-l",
type=argparse.FileType('rb'),
required=True)
parser.add_argument("--editmodel",
"-e",
type=argparse.FileType('rb'),
required=True)
parser.add_argument("--corpus",
"-c",
type=argparse.FileType('r'),
default=sys.stdin)
args = parser.parse_args()
s = SpellChecker(max_distance=2)
s.load_language_model(args.languagemodel)
s.load_channel_model(args.editmodel)
for line in args.corpus:
print("LINE: ", line)
corrected = s.autocorrect_line(line)
print("CORRECTED: ", corrected)
s.load_language_model(args.languagemodel)
s.load_channel_model(args.editmodel)
#print(s.channel_model.prob("hello", "hello"))
#print(s.channel_model.prob("hellp", "hello"))
#print(s.channel_model.prob("hllp", "hello"))
#print(s.check_text("they did not yb any menas"))
"""
>>> [['they'], ['did'], ['not'], ['by', 'b', 'ye', 'y', 'yo', 'ob', 'ya', 'ab'], ['any'],
>>> ['means', 'mens', 'mena', 'zenas', 'menan', 'mends']]
"""
sting = "they did not yb any menas"
print(sting)
print(s.autocorrect_line(sting))
"""
>>> ['they', 'did', 'not', 'by', 'any', 'means']
"""
sting = "I merely underlienes"
print(sting)
print(s.suggest_text(sting, 2))
sting = "pronounce the foreing names"
print(sting)
print(s.suggest_text(sting, 2))
sting = "One in hawiyei"
print(sting)
print(s.suggest_text(sting, 2))