def tower_func(image, label):
assert not self.training
image = self.image_preprocess(image)
image = tf.transpose(image, [0, 3, 1, 2])
image, target_label = attacker.attack(image, label, self.get_logits)
logits = self.get_logits(image)
ImageNetModel.compute_loss_and_error(logits, label) # compute top-1 and top-5
AdvImageNetModel.compute_attack_success(logits, target_label)
def build_graph(self, image, label, indices):
"""
The default tower function.
"""
image = self.image_preprocess(image)
assert self.data_format == 'NCHW'
image = tf.transpose(image, [0, 3, 1, 2])
with tf.variable_scope(tf.get_variable_scope(), reuse=tf.AUTO_REUSE):
# BatchNorm always comes with trouble. We use the testing mode of it during attack.
with freeze_collection([tf.GraphKeys.UPDATE_OPS]), argscope(BatchNorm, training=False):
image, target_label = self.attacker.attack(image, label, self.get_logits)
image = tf.stop_gradient(image, name='adv_training_sample')
logits = self.get_logits(image)
loss = ImageNetModel.compute_loss_and_error(
logits, label, label_smoothing=self.label_smoothing)
AdvImageNetModel.compute_attack_success(logits, target_label)
if not self.training:
return
wd_loss = regularize_cost(self.weight_decay_pattern,
tf.contrib.layers.l2_regularizer(self.weight_decay),
name='l2_regularize_loss')
add_moving_summary(loss, wd_loss)
total_cost = tf.add_n([loss, wd_loss], name='cost')
if self.loss_scale != 1.:
logger.info("Scaling the total loss by {} ...".format(self.loss_scale))
return total_cost * self.loss_scale
else:
return total_cost
def tower_func(image, label):
assert not self.training
image_orig = self.image_preprocess(image)
image_orig = tf.transpose(image_orig, [0, 3, 1, 2])
if hasattr(self, 'palatte'):
image_adv, target_label = attacker.attack(
image_orig, label, self.get_logits_raw, self.palatte)
else:
image_adv, target_label = attacker.attack(
image_orig, label, self.get_logits)
logits = self.get_logits(image_adv)
if save:
from adv_tf_record import save_adv
save_func = lambda image_orig, image_adv, label, target_label, logits: \
save_adv(image_orig, image_adv, label, target_label, logits, data_obj)
image_orig, image_adv, label, target_label, logits = \
tf.py_func(save_func, [image_orig, image_adv, label, target_label, logits],
[image_orig.dtype, image_adv.dtype, label.dtype, target_label.dtype, logits.dtype])
ImageNetModel.compute_loss_and_error(
logits, label) # compute top-1 and top-5
AdvImageNetModel.compute_attack_success(logits, target_label)