def testConvGRU(self):
x = np.random.rand(5, 7, 3, 11)
y = common_layers.conv_gru(tf.constant(x, dtype=tf.float32), (1, 3), 11)
z = common_layers.conv_gru(
tf.constant(x, dtype=tf.float32), (1, 3), 11, padding="LEFT")
self.evaluate(tf.global_variables_initializer())
res1 = self.evaluate(y)
res2 = self.evaluate(z)
self.assertEqual(res1.shape, (5, 7, 3, 11))
self.assertEqual(res2.shape, (5, 7, 3, 11))
def testConvGRU(self):
x = np.random.rand(5, 7, 3, 11)
y = common_layers.conv_gru(tf.constant(x, dtype=tf.float32), (1, 3), 11)
z = common_layers.conv_gru(
tf.constant(x, dtype=tf.float32), (1, 3), 11, padding="LEFT")
self.evaluate(tf.global_variables_initializer())
res1 = self.evaluate(y)
res2 = self.evaluate(z)
self.assertEqual(res1.shape, (5, 7, 3, 11))
self.assertEqual(res2.shape, (5, 7, 3, 11))
def step(state, inp): # pylint: disable=missing-docstring
x = tf.nn.dropout(state, 1.0 - hparams.dropout)
for layer in range(hparams.num_hidden_layers):
x = common_layers.conv_gru(
x, (hparams.kernel_height, hparams.kernel_width),
hparams.hidden_size,
name="cgru_%d" % layer)
# Padding input is zeroed-out in the modality, we check this by summing.
padding_inp = tf.less(tf.reduce_sum(tf.abs(inp), axis=[1, 2]), 0.00001)
new_state = tf.where(padding_inp, state, x) # No-op where inp is padding.
return new_state
def step(state, inp): # pylint: disable=missing-docstring
x = tf.nn.dropout(state, 1.0 - hparams.dropout)
for layer in xrange(hparams.num_hidden_layers):
x = common_layers.conv_gru(
x, (hparams.kernel_height, hparams.kernel_width),
hparams.hidden_size,
name="cgru_%d" % layer)
# Padding input is zeroed-out in the modality, we check this by summing.
padding_inp = tf.less(tf.reduce_sum(tf.abs(inp), axis=[1, 2]), 0.00001)
new_state = tf.where(padding_inp, state, x) # No-op where inp is padding.
return new_state
