def hpv_with_par_sent_subsent(d2v_identifier, space_separated_words):
labeled_sentences = []
for paragraph_index, paragraph_text in enumerate(
space_separated_words.split('paragraphend')):
paragraph_identifier = '{}_PARA{}'.format(d2v_identifier,
paragraph_index)
for sentence_index, sentence_text in enumerate(
paragraph_text.split(' sentenceend')):
sentence_identifier = '{}_SENT{}'.format(paragraph_identifier,
sentence_index)
for sub_sentence_index, sub_sentence_text in enumerate(
sentence_text.split(' subsentenceend')):
sub_sentence_identifier = '{}_SSENT{}'.format(
sentence_identifier, sub_sentence_index)
words = [
s for s in sub_sentence_text.split(' ')
if len(s.strip(' ')) > 0
]
tags = [
d2v_identifier, paragraph_identifier, sentence_identifier,
sub_sentence_identifier
]
labeled_sentences.append(create_tagged_document(tags, words))
return labeled_sentences
def no_hpv(d2v_identifier, space_separated_words):
space_separated_words = space_separated_words.replace(' paragraphend', '')
space_separated_words = space_separated_words.replace(' sentenceend', '')
space_separated_words = space_separated_words.replace(
' subsentenceend', '')
return [
create_tagged_document([d2v_identifier],
space_separated_words.split(' '))
]
def hpv_with_par(d2v_identifier, space_separated_words):
labeled_sentences = []
for paragraph_index, paragraph_text in enumerate(space_separated_words.split('paragraphend')):
paragraph_identifier = '{}_PARA{}'.format(d2v_identifier, paragraph_index)
paragraph_text = paragraph_text.replace(' subsentenceend', '')
paragraph_text = paragraph_text.replace(' sentenceend', '')
words = [s for s in paragraph_text.split(' ') if len(s.strip(' ')) > 0]
tags = [d2v_identifier, paragraph_identifier]
labeled_sentences.append(create_tagged_document(tags, words))
return labeled_sentences
def hpv_with_par_sent_subsentnv(d2v_identifier, space_separated_words):
labeled_sentences = []
for paragraph_index, paragraph_text in enumerate(space_separated_words.split('paragraphend')):
paragraph_identifier = '{}_PARA{}'.format(d2v_identifier, paragraph_index)
for sentence_index, sentence_text in enumerate(paragraph_text.split(' sentenceend')):
sentence_identifier = '{}_SENT{}'.format(paragraph_identifier, sentence_index)
for sub_sentence_index, sub_sentence_text in enumerate(sentence_text.split(' subsentenceend')):
words = [s for s in sub_sentence_text.split(' ') if len(s.strip(' ')) > 0]
tags = [d2v_identifier, paragraph_identifier, sentence_identifier]
labeled_sentences.append(create_tagged_document(tags, words))
return labeled_sentences
def hpv_with_par_sent_and_topics(d2v_identifier, space_separated_words, best_topic, second_best_topic):
labeled_sentences = []
for paragraph_index, paragraph_text in enumerate(space_separated_words.split('paragraphend')):
paragraph_identifier = '{}_PARA{}'.format(d2v_identifier, paragraph_index)
for sentence_index, sentence_text in enumerate(paragraph_text.split(' sentenceend')):
sentence_identifier = '{}_SENT{}'.format(paragraph_identifier, sentence_index)
sentence_text = sentence_text.replace(' subsentenceend', '')
words = [s for s in sentence_text.split(' ') if len(s.strip(' ')) > 0]
tags = [d2v_identifier, paragraph_identifier, sentence_identifier, best_topic_label(best_topic),
second_best_topic_label(second_best_topic)]
labeled_sentences.append(create_tagged_document(tags, words))
return labeled_sentences
def hpv_with_topics(d2v_identifier, space_separated_words, best_topic,
second_best_topic):
space_separated_words = space_separated_words.replace(' paragraphend', '')
space_separated_words = space_separated_words.replace(' sentenceend', '')
space_separated_words = space_separated_words.replace(
' subsentenceend', '')
tags = [
d2v_identifier,
best_topic_label(best_topic),
second_best_topic_label(second_best_topic)
]
return [create_tagged_document(tags, space_separated_words.split(' '))]
def hpv_with_par_and_topics(d2v_identifier, space_separated_words, best_topic,
second_best_topic):
labeled_sentences = []
for paragraph_index, paragraph_text in enumerate(
space_separated_words.split('paragraphend')):
paragraph_identifier = '{}_PARA{}'.format(d2v_identifier,
paragraph_index)
paragraph_text = paragraph_text.replace(' subsentenceend', '')
paragraph_text = paragraph_text.replace(' sentenceend', '')
words = [s for s in paragraph_text.split(' ') if len(s.strip(' ')) > 0]
tags = [
d2v_identifier, paragraph_identifier,
best_topic_label(best_topic),
second_best_topic_label(second_best_topic)
]
labeled_sentences.append(create_tagged_document(tags, words))
return labeled_sentences
def hpv_with_topics(d2v_identifier, space_separated_words, best_topic, second_best_topic):
space_separated_words = space_separated_words.replace(' paragraphend', '')
space_separated_words = space_separated_words.replace(' sentenceend', '')
space_separated_words = space_separated_words.replace(' subsentenceend', '')
tags = [d2v_identifier, best_topic_label(best_topic), second_best_topic_label(second_best_topic)]
return [create_tagged_document(tags, space_separated_words.split(' '))]
def no_hpv(d2v_identifier, space_separated_words):
space_separated_words = space_separated_words.replace(' paragraphend', '')
space_separated_words = space_separated_words.replace(' sentenceend', '')
space_separated_words = space_separated_words.replace(' subsentenceend', '')
return [create_tagged_document([d2v_identifier], space_separated_words.split(' '))]
