def export(self, name, force=False, add_lemmes=False):
"""
Export tokens in a file.
force for export in the valid extension, otherwise it use the pending.
"""
self.content = ""
for sample in self.samples:
for token in sample:
lemme = ""
if add_lemmes:
# Add lemme only if different from original
if token.lemme != token.original:
lemme = u"/%s" % token.lemme
self.content += u"%s/%s%s " % (unicode(
token.original), token.tag, lemme)
self.content += u"\n" # Carriage return on each sample, for human reading
# Define extention
ext = self.PENDING_EXT
if force:
if add_lemmes:
ext = self.LEXICON_EXT
else:
ext = self.VALID_EXT
save_to_file(os.path.join(self.PATH, "%s%s" % (name, ext)),
self.content)
def export(self, name, force=False, add_lemmes=False):
"""
Export tokens in a file.
force for export in the valid extension, otherwise it use the pending.
"""
self.content = ""
for sample in self.samples:
for token in sample:
lemme = ""
if add_lemmes:
# Add lemme only if different from original
if token.lemme != token.original:
lemme = u"/%s" % token.lemme
self.content += u"%s/%s%s " % (unicode(token.original), token.tag, lemme)
self.content += u"\n" # Carriage return on each sample, for human reading
# Define extention
ext = self.PENDING_EXT
if force:
if add_lemmes:
ext = self.LEXICON_EXT
else:
ext = self.VALID_EXT
save_to_file(os.path.join(self.PATH, "%s%s" % (name, ext)), self.content)
def reset_triggers(self):
"""
For full training, we need to remove previous triggers.
"""
save_to_file("corpus/triggers.trg", "")
def reset_triggers(self):
"""
For full training, we need to remove previous triggers.
"""
save_to_file("corpus/triggers.trg", "")