def _discriminator_fn(x):
"""Builds discriminator network."""
return networks.discriminator(x,
progress,
_num_filters_fn,
resolution_schedule,
num_blocks=num_blocks,
kernel_size=kernel_size)
def test_discriminator_grad_norm_progress(self):
if tf.executing_eagerly():
# tf.placeholder() is not compatible with eager execution.
return
stable_stage_num_images = 2
transition_stage_num_images = 3
current_image_id_ph = tf.compat.v1.placeholder(tf.int32, [])
progress = networks.compute_progress(
current_image_id_ph,
stable_stage_num_images,
transition_stage_num_images,
num_blocks=3)
x = tf.random.normal([2, 16, 16, 3])
logits, _ = networks.discriminator(
x, progress, _num_filters_stub,
networks.ResolutionSchedule(
start_resolutions=(4, 4), scale_base=2, num_resolutions=3))
fake_loss = tf.reduce_sum(input_tensor=tf.square(logits))
grad_norms = [
_get_grad_norm(
fake_loss,
tf.compat.v1.trainable_variables('.*/progressive_gan_block_1/.*')),
_get_grad_norm(
fake_loss,
tf.compat.v1.trainable_variables('.*/progressive_gan_block_2/.*')),
_get_grad_norm(
fake_loss,
tf.compat.v1.trainable_variables('.*/progressive_gan_block_3/.*'))
]
grad_norms_output = None
with self.cached_session(use_gpu=True) as sess:
sess.run(tf.compat.v1.global_variables_initializer())
grad_norms_output = np.array([
sess.run(grad_norms, feed_dict={current_image_id_ph: i})
for i in range(15) # total num of images
])
# The gradient of block_1 is always on.
self.assertEqual(
np.argmax(grad_norms_output[:, 0] > 0), 0,
'gradient norms {} for block 1 is not always on'.format(
grad_norms_output[:, 0]))
# The gradient of block_2 is on after 1 stable stage.
self.assertEqual(
np.argmax(grad_norms_output[:, 1] > 0), 3,
'gradient norms {} for block 2 is not on at step 3'.format(
grad_norms_output[:, 1]))
# The gradient of block_3 is on after 2 stable stage + 1 transition stage.
self.assertEqual(
np.argmax(grad_norms_output[:, 2] > 0), 8,
'gradient norms {} for block 3 is not on at step 8'.format(
grad_norms_output[:, 2]))