def test_basic_unigram_counts(self):
z = Corpus("data/ex.unicode")
for letter in z.next_tamil_letter():
# if ( LINUX ): print(letter)
pass
# LetterModels
q = Unigram("data/ex.unicode")
q.frequency_model()
self.assertEqual(q.letter[u"ஷை"] + q.letter[u"சி"], q.letter[u"ந"])
del z, q
def test_basic_unigram_counts(self):
z = Corpus("data/ex.unicode")
for letter in z.next_tamil_letter():
if ( LINUX ): print(letter)
q = LetterModels.Unigram( "data/ex.unicode" )
q.frequency_model( )
if not PYTHON3:
if ( LINUX ): print(unicode(q))
else:
if ( LINUX ): print( q )
assert( q.letter[u"ஷை"] + q.letter[u"சி"] == q.letter[u"ந"] )
del z, q
def test_basic_unigram_counts(self):
z = Corpus("data/ex.unicode")
for letter in z.next_tamil_letter():
if (LINUX): print(letter)
q = LetterModels.Unigram("data/ex.unicode")
q.frequency_model()
if not PYTHON3:
if (LINUX): print(unicode(q))
else:
if (LINUX): print(q)
assert (q.letter[u"ஷை"] + q.letter[u"சி"] == q.letter[u"ந"])
del z, q
def test_basic_unigram_counts(self):
z = Corpus("data/ex.unicode")
for letter in z.next_tamil_letter():
# if ( LINUX ): print(letter)
pass
# LetterModels
q = Unigram("data/ex.unicode")
q.frequency_model()
if not PYTHON3:
# if ( LINUX ): print(unicode(q))
pass
else:
# if ( LINUX ): print( q )
pass
self.assertEqual(q.letter[u"ஷை"] + q.letter[u"சி"], q.letter[u"ந"])
del z, q
def run(parent,outputfile):
x=None
for filename in glob.glob(os.path.join(parent,"*.word")):
if not x:
x = Unigram(filename)
else:
x.corpus = Corpus(filename) #update file
x.frequency_model()
x.save(outputfile)
proc_stats(get_prob(x.letter), outputfile)
return