def evaluate(self):
logger = logging.getLogger("brc")
logger.info("====== evaluating ======")
logger.info('Load data_set and vocab...')
with open(os.path.join(self.config.get_filepath().vocab_dir, 'vocab.data'), 'rb') as fin:
vocab = pickle.load(fin)
assert len(self.dev_files) > 0, 'No dev files are provided.'
dataloader = Propress(self.config.get_default_params().max_p_num,
self.config.get_default_params().max_p_len,
self.config.get_default_params().max_q_len,
self.config.get_default_params().max_ch_len,
dev_files=self.dev_files)
logger.info('Converting text into ids...')
dataloader.convert_to_ids(vocab)
logger.info('Restoring the model...')
model = Model(vocab, trainable=False)
model.restore(self.config.get_filepath().model_dir, self.algo)
logger.info('Evaluating the model on dev set...')
dev_batches = dataloader.next_batch('dev',
self.config.get_default_params().batch_size,
vocab.get_id_byword(vocab.pad_token),
vocab.get_id_bychar(vocab.pad_token),
shuffle=False)
dev_loss, dev_bleu_rouge, summ = model.evaluate(
dev_batches, 'dev', result_dir=self.config.get_filepath().output_dir, result_prefix='dev.predicted')
logger.info('Loss on dev set: {}'.format(dev_loss))
logger.info('Result on dev set: {}'.format(dev_bleu_rouge))
logger.info('Predicted answers are saved to {}'.format(os.path.join(self.config.get_filepath().output_dir)))
def predict(self):
logger = logging.getLogger("brc")
logger.info('Load data_set and vocab...')
with open(os.path.join(self.config.get_filepath().vocab_dir, 'vocab.data'), 'rb') as fin:
vocab = pickle.load(fin)
assert len(self.test_files) > 0, 'No test files are provided.'
dataloader = Propress(self.config.get_default_params().max_p_num,
self.config.get_default_params().max_p_len,
self.config.get_default_params().max_q_len,
self.config.get_default_params().max_ch_len,
test_files=self.test_files)
logger.info('Converting text into ids...')
dataloader.convert_to_ids(vocab)
logger.info('Restoring the model...')
model = Model(vocab, trainable=False)
model.restore(self.config.get_filepath().model_dir, self.algo)
logger.info('Predicting answers for test set...')
test_batches = dataloader.next_batch('test',
self.config.get_default_params().batch_size,
vocab.get_word_id(vocab.pad_token),
vocab.get_char_id(vocab.pad_token),
shuffle=False)
model.evaluate(test_batches, 'test', result_dir=self.config.get_filepath().output_dir, result_prefix='test.predicted')
def evaluate(args):
logger = logging.getLogger("brc")
logger.info("====== evaluating ======")
logger.info('Load data_set and vocab...')
with open(os.path.join(args.vocab_dir, 'vocab.data'), 'rb') as fin:
vocab = pickle.load(fin)
assert len(args.dev_files) > 0, 'No dev files are provided.'
dataloader = Propress(args.max_p_num,
args.max_p_len,
args.max_q_len,
args.max_ch_len,
dev_files=args.dev_files)
logger.info('Converting text into ids...')
dataloader.convert_to_ids(vocab)
logger.info('Restoring the model...')
model = Model(vocab, args)
model.restore(args.model_dir, args.algo)
logger.info('Evaluating the model on dev set...')
dev_batches = dataloader.next_batch('dev',
args.batch_size,
vocab.get_word_id(vocab.pad_token),
vocab.get_char_id(vocab.pad_token),
shuffle=False)
dev_loss, dev_bleu_rouge = model.evaluate(dev_batches,
result_dir=args.result_dir,
result_prefix='dev.predicted')
logger.info('Loss on dev set: {}'.format(dev_loss))
logger.info('Result on dev set: {}'.format(dev_bleu_rouge))
logger.info('Predicted answers are saved to {}'.format(
os.path.join(args.result_dir)))