def process_noroot(self, sSentence, wfile):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if head.name == "ROOT":
head = head.children[0]
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
# print head
if self.m_bBinarize:
outh = CBinarizedTreeNode()
# print outh
self.build_binarized_node(outh, head)
wfile.write(outh.utf8print()+"\n")
#print outh.utf8print()
else:
outh = fidtree.CTreeNode()
self.build_node(outh, head)
wfile.write(outh.utf8print()+"\n")
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
lTokens = []
head_child = self.find_head(head, lTokens)
if self.m_bLabeled:
head_child.label = 'ROOT'
for tokn in lTokens:
if self.m_bLabeled:
print "\t".join([
gb2utf(tokn.token), tokn.pos,
str(tokn.link), tokn.label
])
else:
print "\t".join(
[gb2utf(tokn.token), tokn.pos,
str(tokn.link)])
print # empty line
def process_noroot(self, sSentence, wfile):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if head.name == "ROOT":
head = head.children[0]
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
# print head
if self.m_bBinarize:
outh = CBinarizedTreeNode()
# print outh
self.build_binarized_node(outh, head)
wfile.write(outh.utf8print()+"\n")
#print outh.utf8print()
else:
outh = fidtree.CTreeNode()
self.build_node(outh, head)
wfile.write(outh.utf8print()+"\n")
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
outh=self.count_flatness(head)
print float(outh[1])/float(outh[0])
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
outh=self.count_flatness(head)
print float(outh[1])/float(outh[0])
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
outh = self.count_flatness(head)
if outh[0] == 0: #avoid division by zero if there is no constituent.
print 0.0
else:
print float(outh[1]) / float(outh[0])
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
outh = self.count_flatness(head)
if outh[0] == 0: # avoid division by zero if there is no constituent.
print 0.0
else:
print float(outh[1]) / float(outh[0])
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
lTokens = []
self.find_head(head, lTokens)
for tokn in lTokens:
print "\t".join([tokn.token, tokn.pos, str(tokn.link)])
print # empty line
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
# print head
if self.m_bBinarize:
outh = CBinarizedTreeNode()
# print outh
self.build_binarized_node(outh, head)
print outh
else:
outh = fidtree.CTreeNode()
self.build_node(outh, head)
print outh
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
# print head
if self.m_bBinarize:
outh = CBinarizedTreeNode()
# print outh
self.build_binarized_node(outh, head)
print outh
else:
outh = fidtree.CTreeNode()
self.build_node(outh, head)
print outh
def process(self, sSentence):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
lTokens = []
head_child = self.find_head(head, lTokens)
if self.m_bLabeled:
head_child.label = 'ROOT'
for tokn in lTokens:
if self.m_bLabeled:
print "\t".join([gb2utf(tokn.token), tokn.pos, str(tokn.link), tokn.label])
else:
print "\t".join([gb2utf(tokn.token), tokn.pos, str(tokn.link)])
print # empty line
def process(self, sSentence, wfile):
# don't process empty sentences
if sSentence.strip() == "":
return
# find the cfg node
head = fidtree.parse_object(sSentence)
if type(head) == list:
lHead = head
else:
lHead = [head]
# output the dep node
for head in lHead:
lTokens = []
head_child = self.find_head(head, lTokens)
if self.m_bLabeled:
head_child.label = 'ROOT'
index = 1
for tokn in lTokens:
if self.m_bLabeled:
wfile.write("\t".join([str(index), tokn.token, "_", tokn.pos, tokn.pos, "_", str(tokn.link+1), tokn.label, "_", "_"])+"\n")
else:
wfile.write("\t".join([str(index), tokn.token, "_", tokn.pos, tokn.pos, "_", str(tokn.link+1), "_", "_", "_"])+"\n")
index += 1
wfile.write('\n')