def _make_examples(self, sentences):
word_vocab = embeddings.get_word_vocab(self.config)
char_vocab = embeddings.get_char_vocab()
return [
example.Example(sentence, word_vocab, char_vocab)
for sentence in sentences
]
def _make_examples(self, sentences):
word_vocab = embeddings.get_word_vocab(self.config)
char_vocab = embeddings.get_char_vocab()
return [
example.Example(sentence, word_vocab, char_vocab)
for sentence in sentences
]
def _get_examples(self, split):
word_vocab = embeddings.get_word_vocab(self._config)
char_vocab = embeddings.get_char_vocab()
examples = [
SentenceClassificationExample(self._config, words, tag, word_vocab,
char_vocab, self.label_mapping,
self._task_name)
for words, tag in self.get_labeled_sentences(split)
]
return examples
def _get_examples(self, split):
word_vocab = embeddings.get_word_vocab(self._config)
word_vocab_vi = embeddings.get_word_vocab_vi(self._config)
char_vocab = embeddings.get_char_vocab()
examples = [
TranslationExample(
self._config, words_src, words_tgt, size_src, size_tgt,
word_vocab, char_vocab, self._task_name, word_vocab_vi, split)
for words_src, words_tgt, size_src, size_tgt in self.get_sentence_tuples(split)
]
return examples
def _get_examples(self, split):
word_vocab = embeddings.get_word_vocab(self._config)
char_vocab = embeddings.get_char_vocab()
examples = [
TaggingExample(
self._config, self._is_token_level, words, tags,
word_vocab, char_vocab, self.label_mapping, self._task_name)
for words, tags in self.get_labeled_sentences(split)]
if self._config.train_set_percent < 100:
utils.log('using reduced train set ({:}%)'.format(
self._config.train_set_percent))
random.shuffle(examples)
examples = examples[:int(len(examples) *
self._config.train_set_percent / 100.0)]
return examples
def _get_examples(self, split):
word_vocab = embeddings.get_word_vocab(self._config)
char_vocab = embeddings.get_char_vocab()
examples = [
TaggingExample(
self._config, self._is_token_level, words, tags,
word_vocab, char_vocab, self.label_mapping, self._task_name)
for words, tags in self.get_labeled_sentences(split)]
if self._config.train_set_percent < 100:
utils.log('using reduced train set ({:}%)'.format(
self._config.train_set_percent))
random.shuffle(examples)
examples = examples[:int(len(examples) *
self._config.train_set_percent / 100.0)]
return examples
def get_examples_translate(config, src, split):
words_src = src.strip().split()
size_src = len(words_src)
words_tgt = []
size_tgt = 1
word_vocab = embeddings.get_word_vocab(config)
word_vocab_vi = embeddings.get_word_vocab_vi(config)
char_vocab = embeddings.get_char_vocab()
examples = [
TranslationExample(
config, words_src, words_tgt, size_src, size_tgt,
word_vocab, char_vocab, 'translate', word_vocab_vi, split)
]
return examples
