def testVariablesSetDeviceMobileModel(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()
# Force all Variables to reside on the device.
with tf.variable_scope('on_cpu'), tf.device('/cpu:0'):
with slim.arg_scope(nasnet.nasnet_mobile_arg_scope()):
nasnet.build_nasnet_mobile(inputs, num_classes)
with tf.variable_scope('on_gpu'), tf.device('/gpu:0'):
with slim.arg_scope(nasnet.nasnet_mobile_arg_scope()):
nasnet.build_nasnet_mobile(inputs, num_classes)
for v in tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES,
scope='on_cpu'):
self.assertDeviceEqual(v.device, '/cpu:0')
for v in tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES,
scope='on_gpu'):
self.assertDeviceEqual(v.device, '/gpu:0')
def MyNASNet(self, images, is_training):
arg_scope = nasnet.nasnet_mobile_arg_scope()
with slim.arg_scope(arg_scope):
logits, end_points = nasnet.build_nasnet_mobile(
images,
num_classes=self.num_classes + 1,
is_training=is_training)
global_step = tf.train.get_or_create_global_step()
return logits, end_points, global_step
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(nasnet.nasnet_mobile_arg_scope()):
net, end_points = nasnet.build_nasnet_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, 1056])
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 = nasnet.mobile_imagenet_config()
config.set_hparam('data_format', 'NCHW')
with slim.arg_scope(nasnet.nasnet_mobile_arg_scope()):
_, end_points = nasnet.build_nasnet_mobile(inputs,
num_classes,
config=config)
self.assertListEqual(end_points['Stem'].shape.as_list(),
[batch_size, 88, 28, 28])
def testEvaluationMobileModel(self):
batch_size = 2
height, width = 224, 224
num_classes = 1000
with self.test_session() as sess:
eval_inputs = tf.random_uniform((batch_size, height, width, 3))
with slim.arg_scope(nasnet.nasnet_mobile_arg_scope()):
logits, _ = nasnet.build_nasnet_mobile(eval_inputs,
num_classes,
is_training=False)
predictions = tf.argmax(logits, 1)
sess.run(tf.global_variables_initializer())
output = sess.run(predictions)
self.assertEquals(output.shape, (batch_size, ))
def testUnknownBatchSizeMobileModel(self):
batch_size = 1
height, width = 224, 224
num_classes = 1000
with self.test_session() as sess:
inputs = tf.placeholder(tf.float32, (None, height, width, 3))
with slim.arg_scope(nasnet.nasnet_mobile_arg_scope()):
logits, _ = nasnet.build_nasnet_mobile(inputs, num_classes)
self.assertListEqual(logits.get_shape().as_list(),
[None, num_classes])
images = tf.random_uniform((batch_size, height, width, 3))
sess.run(tf.global_variables_initializer())
output = sess.run(logits, {inputs: images.eval()})
self.assertEquals(output.shape, (batch_size, num_classes))
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 = nasnet.mobile_imagenet_config()
config.set_hparam('use_aux_head', int(use_aux_head))
with slim.arg_scope(nasnet.nasnet_mobile_arg_scope()):
_, end_points = nasnet.build_nasnet_mobile(inputs,
num_classes,
config=config)
self.assertEqual('AuxLogits' in end_points, use_aux_head)
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(nasnet.nasnet_mobile_arg_scope()):
logits, end_points = nasnet.build_nasnet_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 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(nasnet.nasnet_mobile_arg_scope()):
_, end_points = nasnet.build_nasnet_mobile(inputs, num_classes)
endpoints_shapes = {
'Stem': [batch_size, 28, 28, 88],
'Cell_0': [batch_size, 28, 28, 264],
'Cell_1': [batch_size, 28, 28, 264],
'Cell_2': [batch_size, 28, 28, 264],
'Cell_3': [batch_size, 28, 28, 264],
'Cell_4': [batch_size, 14, 14, 528],
'Cell_5': [batch_size, 14, 14, 528],
'Cell_6': [batch_size, 14, 14, 528],
'Cell_7': [batch_size, 14, 14, 528],
'Cell_8': [batch_size, 7, 7, 1056],
'Cell_9': [batch_size, 7, 7, 1056],
'Cell_10': [batch_size, 7, 7, 1056],
'Cell_11': [batch_size, 7, 7, 1056],
'Reduction_Cell_0': [batch_size, 14, 14, 352],
'Reduction_Cell_1': [batch_size, 7, 7, 704],
'global_pool': [batch_size, 1056],
# Logits and predictions
'AuxLogits': [batch_size, num_classes],
'Logits': [batch_size, num_classes],
'Predictions': [batch_size, num_classes]
}
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.assertTrue(endpoint_name in end_points)
self.assertListEqual(
end_points[endpoint_name].get_shape().as_list(),
expected_shape)
def pnasnet_mobile_arg_scope(weight_decay=4e-5,
batch_norm_decay=0.9997,
batch_norm_epsilon=0.001):
"""Default arg scope for the PNASNet Mobile ImageNet model."""
return nasnet.nasnet_mobile_arg_scope(weight_decay, batch_norm_decay,
batch_norm_epsilon)
import numpy as np
from slim.nets.nasnet import nasnet #轻量级 适合移动端
from slim.datasets import imagenet
slim = tf.contrib.slim
tf.reset_default_graph()
sample_images = ['hy.jpg', 'ps.jpg', 'filename3.jpg'] #导入测试样本
image_size = nasnet.build_nasnet_mobile.default_image_size #获得模型图片尺寸
input_imgs = tf.placeholder(tf.float32,
[None, image_size, image_size, 3]) #为根据模型尺寸定义输入占位符
#定义输出
x1 = 2 * (input_imgs / 255) - 1 #归一化图片
arg_scope = nasnet.nasnet_mobile_arg_scope() #获得模型的命名空间
with slim.arg_scope(arg_scope): #将图片放入模型
logits, end_points = nasnet.build_nasnet_mobile(x1,
num_classes=1001,
is_training=False)
prob = end_points['Predictions'] #获得结果的节点
y = tf.argmax(prob, axis=1) #按概率获得分类结果
checkpoint_file = 'nasnet-a_mobile_04_10_2017/model.ckpt' #定义模型地址
saver = tf.train.Saver()
with tf.Session() as sess:
saver.restore(sess, checkpoint_file) #载入模型
def preimg(img): #resize图片
ch = 3
if img.mode == 'RGBA':
