def getSeqs(p1, p2, words):
# p1 = p1.split()
# p2 = p2.split()
X1, X2 = [], []
for i in p1:
X1.append(ppdb_utils.lookupIDX(words, i))
for i in p2:
X2.append(ppdb_utils.lookupIDX(words, i))
return X1, X2
def getSeqs(p1,p2,words):
p1 = p1.split()
p2 = p2.split()
X1 = []
X2 = []
for i in p1:
X1.append(ppdb_utils.lookupIDX(words,i))
for i in p2:
X2.append(ppdb_utils.lookupIDX(words,i))
return X1, X2
def getSeqs2(p1, p2, words):
p1 = p1.split()
p2 = p2.split()
p11 = [i.split('_')[0] for i in p1]
p22 = [i.split('_')[0] for i in p2]
sp11 = [i.split('_')[1] for i in p1]
sp22 = [i.split('_')[1] for i in p2]
X1 = []
X2 = []
SX1 = []
SX2 = []
for i in p11:
X1.append(ppdb_utils.lookupIDX(words, i))
for i in p22:
X2.append(ppdb_utils.lookupIDX(words, i))
for i in sp11:
SX1.append(i)
for i in sp22:
SX2.append(i)
return X1, X2, SX1, SX2
def getSeqs2(p1, p2, words):
# p1 = p1.split()
# p2 = p2.split()
X1 = []
X2 = []
for i in p1:
tmp = []
if len(i) == 0:
X1.append(tmp)
else:
for j in i:
tmp.append(ppdb_utils.lookupIDX(words, j))
X1.append(tmp)
for i in p2:
tmp = []
if len(i) == 0:
X2.append(tmp)
else:
for j in i:
tmp.append(ppdb_utils.lookupIDX(words, j))
X2.append(tmp)
return X1, X2
def populate_embeddings(self, words, chars):
phrase = self.phrase.lower()
#print phrase
arr = phrase.split()
word = []
char = []
for i in arr:
if len(i.split('_')) > 1:
word.append(i.split('_')[0])
char.append(i.split('_')[1:])
else:
word.append(i)
char.append([])
for i in word:
self.embeddings.append(ppdb_utils.lookupIDX(words,i))
for i in char:
tmp = []
if len(i) == 0:
self.char_embeddings.append([])
else:
for j in i:
tmp.append(ppdb_utils.lookupIDX(chars, j))
self.char_embeddings.append(tmp)
def getSeq(p1,words):
p1 = p1.split()
X1 = []
for i in p1:
X1.append(ppdb_utils.lookupIDX(words,i))
return X1
def populate_embeddings(self, words, pos_vocab):
for i in self.phrase:
self.embeddings.append(ppdb_utils.lookupIDX(words,i))
for i in self.pos:
self.pos_embeddings.append(ppdb_utils.lookupIDX(pos_vocab,i))
def populate_embeddings(self, words):
#phrase = self.phrase.lower()
arr = self.phrase.split()
for i in arr:
self.embeddings.append(ppdb_utils.lookupIDX(words, i))
def getSeq(p1, words):
p1 = p1.split()
X1 = []
for i in p1:
X1.append(ppdb_utils.lookupIDX(words, i))
return X1
def populate_embeddings(self, words):
phrase = self.phrase.lower()
arr = phrase.split()
for i in arr:
self.embeddings.append(ppdb_utils.lookupIDX(words,i))
