def testSequentialEncoder(self, transition_layer_fn):
inputs = tf.zeros([3, 5, 10])
encoder = encoders.SequentialEncoder(
[DenseEncoder(1, 20), DenseEncoder(3, 20)],
transition_layer_fn=transition_layer_fn)
outputs, states, _ = encoder(inputs)
self.assertEqual(len(states), 4)
outputs = self.evaluate(outputs)
self.assertAllEqual(outputs.shape, [3, 5, 20])
def testSequentialEncoder(self):
sequence_length = [17, 21, 20]
inputs = _build_dummy_sequences(sequence_length)
encoder = encoders.SequentialEncoder([
encoders.UnidirectionalRNNEncoder(1, 20),
encoders.PyramidalRNNEncoder(3, 10, reduction_factor=2)])
_, _, encoded_length = encoder.encode(
inputs, sequence_length=sequence_length)
with self.test_session() as sess:
sess.run(tf.global_variables_initializer())
encoded_length = sess.run(encoded_length)
self.assertAllEqual([4, 5, 5], encoded_length)
def testSequentialEncoder(self, transition_layer_fn):
inputs = tf.zeros([3, 5, 10])
encoder = encoders.SequentialEncoder(
[DenseEncoder(1, 20), DenseEncoder(3, 20)],
transition_layer_fn=transition_layer_fn)
outputs, states, _ = encoder.encode(inputs)
self.assertEqual(len(states), 4)
if not compat.is_tf2():
with self.test_session() as sess:
sess.run(tf.global_variables_initializer())
outputs = self.evaluate(outputs)
self.assertAllEqual(outputs.shape, [3, 5, 20])
def testSequentialEncoder(self):
sequence_length = [17, 21, 20]
inputs = _build_dummy_sequences(sequence_length)
encoder = encoders.SequentialEncoder([
encoders.UnidirectionalRNNEncoder(1, 20),
encoders.PyramidalRNNEncoder(3, 10, reduction_factor=2)
])
_, state, encoded_length = encoder.encode(
inputs, sequence_length=sequence_length)
self.assertEqual(4, len(state))
for s in state:
self.assertIsInstance(s, tf.contrib.rnn.LSTMStateTuple)
with self.test_session() as sess:
sess.run(tf.global_variables_initializer())
encoded_length = sess.run(encoded_length)
self.assertAllEqual([4, 5, 5], encoded_length)
def testSequentialEncoderWithTooManyTransitionLayers(self):
with self.assertRaises(ValueError):
_ = encoders.SequentialEncoder(
[DenseEncoder(1, 20), DenseEncoder(3, 20)],
transition_layer_fn=[tf.identity, tf.identity],
)
