def test_discriminator_invalid_input(self):
try:
networks.discriminator(tf.zeros([28, 28, 3]))
except (ValueError, tf.errors.InvalidArgumentError):
# TF raises ValueError, while TF2 raises tf.errors.InvalidArgumentError.
return
except Exception as e: # pylint: disable=broad-except
self.assertTrue(False, msg='Unexpected exception: {}'.format(e))
self.assertTrue(False,
msg='Expected ValueError or InvalidArgumentError.')
def test_discriminator_graph(self):
# Check graph construction for a number of image size/depths and batch
# sizes.
for batch_size, patch_size in zip([3, 6], [70, 128]):
tf.reset_default_graph()
img = tf.ones([batch_size, patch_size, patch_size, 3])
disc_output = networks.discriminator(img)
self.assertEqual(2, disc_output.shape.ndims)
self.assertEqual(batch_size, disc_output.shape.as_list()[0])
def test_discriminator_run(self):
img_batch = tf.zeros([3, 70, 70, 3])
disc_output = networks.discriminator(img_batch)
with self.cached_session() as sess:
sess.run(tf.global_variables_initializer())
sess.run(disc_output)