def testAllEndPointsShapesMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = 1000
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
_, end_points = pnasnet.build_pnasnet_mobile(inputs, num_classes)
endpoints_shapes = {
'Stem': [batch_size, 28, 28, 135],
'Cell_0': [batch_size, 28, 28, 270],
'Cell_1': [batch_size, 28, 28, 270],
'Cell_2': [batch_size, 28, 28, 270],
'Cell_3': [batch_size, 14, 14, 540],
'Cell_4': [batch_size, 14, 14, 540],
'Cell_5': [batch_size, 14, 14, 540],
'Cell_6': [batch_size, 7, 7, 1080],
'Cell_7': [batch_size, 7, 7, 1080],
'Cell_8': [batch_size, 7, 7, 1080],
'global_pool': [batch_size, 1080],
# Logits and predictions
'AuxLogits': [batch_size, num_classes],
'Predictions': [batch_size, num_classes],
'Logits': [batch_size, num_classes],
}
self.assertEqual(len(end_points), 14)
self.assertItemsEqual(endpoints_shapes.keys(), end_points.keys())
for endpoint_name in endpoints_shapes:
tf.logging.info('Endpoint name: {}'.format(endpoint_name))
expected_shape = endpoints_shapes[endpoint_name]
self.assertIn(endpoint_name, end_points)
self.assertListEqual(
end_points[endpoint_name].get_shape().as_list(),
expected_shape)
def testAllEndPointsShapesMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = 1000
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
_, end_points = pnasnet.build_pnasnet_mobile(inputs, num_classes)
endpoints_shapes = {
'Stem': [batch_size, 28, 28, 135],
'Cell_0': [batch_size, 28, 28, 270],
'Cell_1': [batch_size, 28, 28, 270],
'Cell_2': [batch_size, 28, 28, 270],
'Cell_3': [batch_size, 14, 14, 540],
'Cell_4': [batch_size, 14, 14, 540],
'Cell_5': [batch_size, 14, 14, 540],
'Cell_6': [batch_size, 7, 7, 1080],
'Cell_7': [batch_size, 7, 7, 1080],
'Cell_8': [batch_size, 7, 7, 1080],
'global_pool': [batch_size, 1080],
# Logits and predictions
'AuxLogits': [batch_size, num_classes],
'Predictions': [batch_size, num_classes],
'Logits': [batch_size, num_classes],
}
self.assertEqual(len(end_points), 14)
self.assertItemsEqual(endpoints_shapes.keys(), end_points.keys())
for endpoint_name in endpoints_shapes:
tf.logging.info('Endpoint name: {}'.format(endpoint_name))
expected_shape = endpoints_shapes[endpoint_name]
self.assertIn(endpoint_name, end_points)
self.assertListEqual(end_points[endpoint_name].get_shape().as_list(),
expected_shape)
def testBuildNonExistingLayerMobileModel(self):
"""Tests that the model is built correctly without unnecessary layers."""
inputs = tf.random_uniform((5, 224, 224, 3))
tf.train.create_global_step()
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
pnasnet.build_pnasnet_mobile(inputs, 1000)
vars_names = [x.op.name for x in tf.trainable_variables()]
self.assertIn('cell_stem_0/1x1/weights', vars_names)
self.assertNotIn('cell_stem_1/comb_iter_0/right/1x1/weights', vars_names)
def testBuildNonExistingLayerMobileModel(self):
"""Tests that the model is built correctly without unnecessary layers."""
inputs = tf.random_uniform((5, 224, 224, 3))
tf.train.create_global_step()
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
pnasnet.build_pnasnet_mobile(inputs, 1000)
vars_names = [x.op.name for x in tf.trainable_variables()]
self.assertIn('cell_stem_0/1x1/weights', vars_names)
self.assertNotIn('cell_stem_1/comb_iter_0/right/1x1/weights', vars_names)
def testBuildPreLogitsMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = None
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
net, end_points = pnasnet.build_pnasnet_mobile(inputs, num_classes)
self.assertFalse('AuxLogits' in end_points)
self.assertFalse('Predictions' in end_points)
self.assertTrue(net.op.name.startswith('final_layer/Mean'))
self.assertListEqual(net.get_shape().as_list(), [batch_size, 1080])
def testBuildPreLogitsMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = None
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
net, end_points = pnasnet.build_pnasnet_mobile(inputs, num_classes)
self.assertFalse('AuxLogits' in end_points)
self.assertFalse('Predictions' in end_points)
self.assertTrue(net.op.name.startswith('final_layer/Mean'))
self.assertListEqual(net.get_shape().as_list(), [batch_size, 1080])
def testOverrideHParamsMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = 1000
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
config = pnasnet.mobile_imagenet_config()
config.set_hparam('data_format', 'NCHW')
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
_, end_points = pnasnet.build_pnasnet_mobile(
inputs, num_classes, config=config)
self.assertListEqual(end_points['Stem'].shape.as_list(),
[batch_size, 135, 28, 28])
def testOverrideHParamsMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = 1000
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
config = pnasnet.mobile_imagenet_config()
config.set_hparam('data_format', 'NCHW')
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
_, end_points = pnasnet.build_pnasnet_mobile(
inputs, num_classes, config=config)
self.assertListEqual(end_points['Stem'].shape.as_list(),
[batch_size, 135, 28, 28])
def testNoAuxHeadMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = 1000
for use_aux_head in (True, False):
tf.reset_default_graph()
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
config = pnasnet.mobile_imagenet_config()
config.set_hparam('use_aux_head', int(use_aux_head))
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
_, end_points = pnasnet.build_pnasnet_mobile(
inputs, num_classes, config=config)
self.assertEqual('AuxLogits' in end_points, use_aux_head)
def testNoAuxHeadMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = 1000
for use_aux_head in (True, False):
tf.reset_default_graph()
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
config = pnasnet.mobile_imagenet_config()
config.set_hparam('use_aux_head', int(use_aux_head))
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
_, end_points = pnasnet.build_pnasnet_mobile(
inputs, num_classes, config=config)
self.assertEqual('AuxLogits' in end_points, use_aux_head)
def testUseBoundedAcitvationMobileModel(self):
batch_size = 1
height, width = 224, 224
num_classes = 1000
for use_bounded_activation in (True, False):
tf.reset_default_graph()
inputs = tf.random_uniform((batch_size, height, width, 3))
config = pnasnet.mobile_imagenet_config()
config.set_hparam('use_bounded_activation', use_bounded_activation)
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
_, _ = pnasnet.build_pnasnet_mobile(
inputs, num_classes, config=config)
for node in tf.get_default_graph().as_graph_def().node:
if node.op.startswith('Relu'):
self.assertEqual(node.op == 'Relu6', use_bounded_activation)
def pnasnet_mobile(inputs, is_training, opts):
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope(
weight_decay=opts.weight_decay,
batch_norm_decay=opts.batch_norm_decay,
batch_norm_epsilon=opts.batch_norm_epsilon)):
config = pnasnet.large_imagenet_config()
config.set_hparam('dense_dropout_keep_prob', opts.dropout_keep_prob)
config.set_hparam('use_aux_head', int(opts.create_aux_logits))
return pnasnet.build_pnasnet_mobile(
inputs,
num_classes=opts.num_classes,
is_training=is_training,
config=config)
def testUseBoundedAcitvationMobileModel(self):
batch_size = 1
height, width = 224, 224
num_classes = 1000
for use_bounded_activation in (True, False):
tf.reset_default_graph()
inputs = tf.random_uniform((batch_size, height, width, 3))
config = pnasnet.mobile_imagenet_config()
config.set_hparam('use_bounded_activation', use_bounded_activation)
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
_, _ = pnasnet.build_pnasnet_mobile(
inputs, num_classes, config=config)
for node in tf.get_default_graph().as_graph_def().node:
if node.op.startswith('Relu'):
self.assertEqual(node.op == 'Relu6', use_bounded_activation)
def testBuildLogitsMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = 1000
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
logits, end_points = pnasnet.build_pnasnet_mobile(inputs, num_classes)
auxlogits = end_points['AuxLogits']
predictions = end_points['Predictions']
self.assertListEqual(auxlogits.get_shape().as_list(),
[batch_size, num_classes])
self.assertListEqual(logits.get_shape().as_list(),
[batch_size, num_classes])
self.assertListEqual(predictions.get_shape().as_list(),
[batch_size, num_classes])
def testBuildLogitsMobileModel(self):
batch_size = 5
height, width = 224, 224
num_classes = 1000
inputs = tf.random_uniform((batch_size, height, width, 3))
tf.train.create_global_step()
with slim.arg_scope(pnasnet.pnasnet_mobile_arg_scope()):
logits, end_points = pnasnet.build_pnasnet_mobile(inputs, num_classes)
auxlogits = end_points['AuxLogits']
predictions = end_points['Predictions']
self.assertListEqual(auxlogits.get_shape().as_list(),
[batch_size, num_classes])
self.assertListEqual(logits.get_shape().as_list(),
[batch_size, num_classes])
self.assertListEqual(predictions.get_shape().as_list(),
[batch_size, num_classes])
