def extract(chSentL, enSentL, waL, baseID, fwD, wpD):
expL = []
for k, chSent in enumerate(chSentL):
#print >> sys.stderr, k,
enSent = enSentL[k]
waSent = waL[k]
for i, ch in enumerate(chSent):
for j, en in enumerate(enSent):
ID = "ID" + str(baseID + k) + '--' + str(i) + '-' + str(j)
label = "False"
if str(i) + '-' + str(j) in waSent:
label = "True"
exp = Example(ID, label)
exp.featList = extractFeat(i, j, chSent, enSent, wpD, fwD)
expL.append(exp)
return expL
def extract(chSentL, enSentL, waL, baseID, fwD, wpD): expL = [] for k, chSent in enumerate(chSentL): #print >> sys.stderr, k, enSent = enSentL[k] waSent = waL[k] for i, ch in enumerate(chSent): for j, en in enumerate(enSent): ID = "ID" + str(baseID + k) + '--' + str(i) + '-' + str(j) label = "False" if str(i) + '-' + str(j) in waSent: label = "True" exp = Example(ID, label) exp.featList = extractFeat(i, j, chSent, enSent, wpD, fwD) expL.append(exp) return expL
def make(chF, enF, gwaF, waF, outF):
chSentL = [
line.split() for line in codecs.open(chF, 'r', 'utf-8').readlines()
]
enSentL = [
line.split() for line in codecs.open(enF, 'r', 'utf-8').readlines()
]
if gwaF == "None":
gwaL = [[] for i in xrange(len(chSentL))]
else:
gwaL = [line.split() for line in open(gwaF).readlines()]
waL = [line.split() for line in open(waF).readlines()]
print "len of chSentL, enSentL, gwaL, waL: ", len(chSentL), len(
enSentL), len(gwaL), len(waL)
fwD = loadFuncWordDict("ch_funcWordL.txt")
wpD = loadWordPairDict("cedict_hacept_train.dict")
#wpD = loadWordPairDict("hacept_train.dict")
expList = []
for k, chSent in enumerate(chSentL):
if k % 100 == 0: print k,
enSent = enSentL[k]
waSent = waL[k]
gwaSent = gwaL[k]
for wa in waSent:
ID = 'ID' + str(k) + '--' + wa
label = 'False'
if wa in gwaSent:
label = 'True'
exp = Example(ID, label)
i, j = int(wa.split('-')[0]), int(wa.split('-')[1])
exp.featList = extractFeat(i, j, chSent, enSent, wpD, fwD)
expList.append(exp)
outf = codecs.open(outF, 'w', 'utf-8')
for exp in expList:
outf.write(exp.__str__())
outf.close()
def make(chF, enF, gwaF, waF, outF):
chSentL = [line.split() for line in codecs.open(chF, 'r', 'utf-8').readlines()]
enSentL = [line.split() for line in codecs.open(enF, 'r', 'utf-8').readlines()]
if gwaF == "None":
gwaL = [[] for i in xrange(len(chSentL))]
else:
gwaL = [line.split() for line in open(gwaF).readlines()]
waL = [line.split() for line in open(waF).readlines()]
print "len of chSentL, enSentL, gwaL, waL: ", len(chSentL), len(enSentL), len(gwaL), len(waL)
fwD = loadFuncWordDict("ch_funcWordL.txt")
wpD = loadWordPairDict("cedict_hacept_train.dict")
#wpD = loadWordPairDict("hacept_train.dict")
expList = []
for k, chSent in enumerate(chSentL):
if k % 100 == 0: print k,
enSent = enSentL[k]
waSent = waL[k]
gwaSent = gwaL[k]
for wa in waSent:
ID = 'ID' + str(k) + '--' + wa
label = 'False'
if wa in gwaSent:
label = 'True'
exp = Example(ID, label)
i, j = int(wa.split('-')[0]), int(wa.split('-')[1])
exp.featList = extractFeat(i, j, chSent, enSent, wpD, fwD)
expList.append(exp)
outf = codecs.open(outF, 'w', 'utf-8')
for exp in expList:
outf.write(exp.__str__())
outf.close()
