def predict(self):
"""Predicts result from the model."""
params, flags_obj, is_train = self.params, self.flags_obj, False
with tf.name_scope("model"):
model = transformer.create_model(params, is_train)
self._load_weights_if_possible(model, flags_obj.init_weight_path)
model.summary()
subtokenizer = tokenizer.Subtokenizer(flags_obj.vocab_file)
ds = data_pipeline.eval_input_fn(params)
ds = ds.map(lambda x, y: x).take(_SINGLE_SAMPLE)
ret = model.predict(ds)
val_outputs, _ = ret
length = len(val_outputs)
for i in range(length):
translate.translate_from_input(val_outputs[i], subtokenizer)
def predict(self):
"""Predicts result from the model."""
self.params['train'] = False
params = self.params
flags_obj = self.flags_obj
with tf.name_scope("model"):
model = transformer.create_model(params, is_train=False)
self._load_weights_if_possible(
model, tf.train.latest_checkpoint(self.flags_obj.model_dir))
model.summary()
subtokenizer = tokenizer.Subtokenizer(flags_obj.vocab_file)
print(params)
ds = data_pipeline.eval_input_fn(params)
ds = ds.map(lambda x, y: x).take(_SINGLE_SAMPLE)
import time
start = time.time()
ret = model.predict(ds)
val_outputs, _ = ret
length = len(val_outputs)
for i in range(length):
translate.translate_from_input(val_outputs[i], subtokenizer)
print('\n\n\n', time.time() - start)
