def mobilenet(x, is_train=True, reuse=False):
with tf.variable_scope("mobilenet", reuse=reuse):
n = InputLayer(x)
n = conv_block(n, 32, strides=(2, 2), is_train=is_train, name="conv")
n = depthwise_conv_block(n, 64, is_train=is_train, name="depth1")
n = depthwise_conv_block(n, 128, strides=(2, 2), is_train=is_train, name="depth2")
n = depthwise_conv_block(n, 128, is_train=is_train, name="depth3")
n = depthwise_conv_block(n, 256, strides=(2, 2), is_train=is_train, name="depth4")
n = depthwise_conv_block(n, 256, is_train=is_train, name="depth5")
n = depthwise_conv_block(n, 512, strides=(2, 2), is_train=is_train, name="depth6")
n = depthwise_conv_block(n, 512, is_train=is_train, name="depth7")
n = depthwise_conv_block(n, 512, is_train=is_train, name="depth8")
n = depthwise_conv_block(n, 512, is_train=is_train, name="depth9")
n = depthwise_conv_block(n, 512, is_train=is_train, name="depth10")
n = depthwise_conv_block(n, 512, is_train=is_train, name="depth11")
n = depthwise_conv_block(n, 1024, strides=(2, 2), is_train=is_train, name="depth12")
n = depthwise_conv_block(n, 1024, is_train=is_train, name="depth13")
n = GlobalMeanPool2d(n)
# n = DropoutLayer(n, 1-1e-3, True, is_train, name='drop')
# n = DenseLayer(n, 1000, act=tf.identity, name='output') # equal
n = ReshapeLayer(n, [-1, 1, 1, 1024])
n = Conv2d(n, 1000, (1, 1), (1, 1), name='out')
n = FlattenLayer(n)
return n
def get_E_x2za(x_shape=(None, flags.img_size_h, flags.img_size_w, flags.c_dim),
name="Encoder_x2za"):
# ref: Multimodal Unsupervised Image-to-Image Translation
# input: (batch_size_train, 256, 256, 3)
# output: vector (batch_size_train, za_dim)
w_init = tf.random_normal_initializer(stddev=0.02)
ni = Input(x_shape)
n = Conv2d(64, (7, 7), (1, 1), act=tf.nn.relu, W_init=w_init)(ni)
n = Conv2d(128, (4, 4), (2, 2), act=tf.nn.relu, W_init=w_init)(n)
n = Conv2d(256, (4, 4), (2, 2), act=tf.nn.relu, W_init=w_init)(n)
n = Conv2d(256, (4, 4), (2, 2), act=tf.nn.relu, W_init=w_init)(n)
n = GlobalMeanPool2d()(n)
n = Flatten()(n)
n = Dense(flags.za_dim)(n)
M = Model(inputs=ni, outputs=n, name=name)
return M
def mobilenetv1(self, x, end_with='out', is_train=False, reuse=None):
with tf.variable_scope("mobilenetv1", reuse=reuse):
n = InputLayer(x)
n = self.conv_block(n, 32, strides=(2, 2), is_train=is_train, name="conv")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 64, is_train=is_train, name="depth1")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 128, strides=(2, 2), is_train=is_train, name="depth2")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 128, is_train=is_train, name="depth3")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 256, strides=(2, 2), is_train=is_train, name="depth4")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 256, is_train=is_train, name="depth5")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 512, strides=(2, 2), is_train=is_train, name="depth6")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 512, is_train=is_train, name="depth7")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 512, is_train=is_train, name="depth8")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 512, is_train=is_train, name="depth9")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 512, is_train=is_train, name="depth10")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 512, is_train=is_train, name="depth11")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 1024, strides=(2, 2), is_train=is_train, name="depth12")
if end_with in n.outputs.name: return n
n = self.depthwise_conv_block(n, 1024, is_train=is_train, name="depth13")
if end_with in n.outputs.name: return n
n = GlobalMeanPool2d(n, name='globalmeanpool')
if end_with in n.outputs.name: return n
# n = DropoutLayer(n, 1-1e-3, True, is_train, name='drop')
# n = DenseLayer(n, 1000, act=tf.identity, name='output') # equal
n = ReshapeLayer(n, [-1, 1, 1, 1024], name='reshape')
if end_with in n.outputs.name: return n
n = Conv2d(n, 1000, (1, 1), (1, 1), name='out')
n = FlattenLayer(n, name='flatten')
if end_with == 'out': return n
raise Exception("end_with : conv, depth1, depth2 ... depth13, globalmeanpool, out")
def get_Ea(x_shape=(None, flags.img_size_h, flags.img_size_w, flags.c_dim),
name=None): # input: (1, 256, 256, 3)
# ref: DRIT source code (Pytorch Implementation)
w_init = tf.random_normal_initializer(stddev=0.02)
ndf = 64
ni = Input(x_shape)
nn = Conv2d(ndf, (7, 7), (1, 1), padding='VALID', W_init=w_init,
act=None)(ni)
## Basic Blocks * 3
for i in range(1, 4):
## Basic Block
# conv part
n = Lambda(lambda x: tf.nn.leaky_relu(x, 0.2))(nn) # leaky relu (0.2)
n = Conv2d(ndf * i, (3, 3), (1, 1),
padding='VALID',
W_init=w_init,
act=lrelu)(n) # conv3x3
n = Conv2d(ndf * (i + 1), (3, 3), (1, 1),
padding='VALID',
W_init=w_init,
act=None)(n) # conv3x3 in convMeanpool
n = MeanPool2d((2, 2), (2, 2))(n) # meanPool2d in convMeanpool
# shortcut part
ns = MeanPool2d((2, 2), (2, 2))(nn)
ns = Conv2d(ndf * (i + 1), (3, 3), (1, 1),
padding='VALID',
W_init=w_init)(ns)
nn = Elementwise(tf.add)([n, ns])
n = GlobalMeanPool2d()(nn)
n_mu = Dense(n_units=flags.za_dim, W_init=w_init, name="mean_linear")(n)
n_log_var = Dense(n_units=flags.za_dim, W_init=w_init,
name="var_linear")(n)
# Sampling using reparametrization trick
def sample(input):
n_mu = input[0]
n_log_var = input[1]
epsilon = tf.random.truncated_normal(n_mu.shape)
stddev = tf.exp(n_log_var)
out = n_mu + stddev * epsilon
return out
no = Lambda(sample)([n_mu, n_log_var])
M = Model(inputs=ni, outputs=[no, n_mu, n_log_var], name=name)
return M
def ShuffleNetV1(self, inputlayer, name):
inputlayer = InputLayer(inputlayer, name='input')#32*32*2
#print(inputlayer.outputs.get_shape())
x = Conv2d(inputlayer, 24, (3, 3), strides=(2, 2), padding='SAME', act=tf.nn.relu, name=name+'_Con2d')###24
x = MaxPool2d(x, filter_size=(3, 3), strides=(2, 2), padding='SAME', name=name+'_MaxPool')
x = self.stage(x, n_filter=384, filter_size=(3, 3), groups=8, repeat=4, stage=2, name=name+'_stage1')
#print("stage1 finished!!!!!!!!!!!!!!!!")
x = self.stage(x, n_filter=768, filter_size=(3, 3), groups=8, repeat=8, stage=3, name=name+'_stage2')
#print("stage2 finished!!!!!!!!!!!!!!!!")
x = self.stage(x, n_filter=1536, filter_size=(3, 3), groups=8, repeat=4, stage=4, name=name+'_stage3')
#print("stage3 finished!!!!!!!!!!!!!!!!")
print("stage3", x.outputs.get_shape())
print(x.count_params())
#x = GlobalMaxPool2d(x, name=name+'_GlobalMaxPool')
#print("GMP", x.outputs.get_shape())
#print(x.count_params())
x = GlobalMeanPool2d(x, name=name+'_GlobalMaxPool')
print("GAP", x.outputs.get_shape())
print(x.count_params())
x = DenseLayer(x, name=name+'_Dense')
print("DENSE", x.outputs.get_shape())
print(x.count_params())
return x
def MobileNetV1(pretrained=False, end_with='out', name=None):
"""Pre-trained MobileNetV1 model (static mode). Input shape [?, 224, 224, 3], value range [0, 1].
Parameters
----------
pretrained : boolean
Whether to load pretrained weights. Default False.
end_with : str
The end point of the model [conv, depth1, depth2 ... depth13, globalmeanpool, out]. Default ``out`` i.e. the whole model.
name : None or str
Name for this model.
Examples
---------
Classify ImageNet classes, see `tutorial_models_mobilenetv1.py <https://github.com/tensorlayer/tensorlayer/blob/master/example/tutorial_models_mobilenetv1.py>`__
>>> # get the whole model with pretrained weights
>>> mobilenetv1 = tl.models.MobileNetV1(pretrained=True)
>>> # use for inferencing
>>> output = mobilenetv1(img1, is_train=False)
>>> prob = tf.nn.softmax(output)[0].numpy()
Extract features and Train a classifier with 100 classes
>>> # get model without the last layer
>>> cnn = tl.models.MobileNetV1(pretrained=True, end_with='reshape').as_layer()
>>> # add one more layer and build new model
>>> ni = Input([None, 224, 224, 3], name="inputs")
>>> nn = cnn(ni)
>>> nn = Conv2d(100, (1, 1), (1, 1), name='out')(nn)
>>> nn = Flatten(name='flatten')(nn)
>>> model = tl.models.Model(inputs=ni, outputs=nn)
>>> # train your own classifier (only update the last layer)
>>> train_params = model.get_layer('out').trainable_weights
Returns
-------
static MobileNetV1.
"""
ni = Input([None, 224, 224, 3], name="input")
for i in range(len(layer_names)):
if i == 0:
n = conv_block(ni,
n_filters[i],
strides=(2, 2),
name=layer_names[i])
elif layer_names[i] in ['depth2', 'depth4', 'depth6', 'depth12']:
n = depthwise_conv_block(n,
n_filters[i],
strides=(2, 2),
name=layer_names[i])
elif layer_names[i] == 'globalmeanpool':
n = GlobalMeanPool2d(name='globalmeanpool')(n)
elif layer_names[i] == 'reshape':
n = Reshape([-1, 1, 1, 1024], name='reshape')(n)
elif layer_names[i] == 'out':
n = Conv2d(1000, (1, 1), (1, 1), name='out')(n)
n = Flatten(name='flatten')(n)
else:
n = depthwise_conv_block(n, n_filters[i], name=layer_names[i])
if layer_names[i] == end_with:
break
network = Model(inputs=ni, outputs=n, name=name)
if pretrained:
restore_params(network)
return network
def ResNet50(pretrained=False, end_with='fc1000', n_classes=1000, name=None):
"""Pre-trained MobileNetV1 model (static mode). Input shape [?, 224, 224, 3].
To use pretrained model, input should be in BGR format and subtracted from ImageNet mean [103.939, 116.779, 123.68].
Parameters
----------
pretrained : boolean
Whether to load pretrained weights. Default False.
end_with : str
The end point of the model [conv, depth1, depth2 ... depth13, globalmeanpool, out].
Default ``out`` i.e. the whole model.
n_classes : int
Number of classes in final prediction.
name : None or str
Name for this model.
Examples
---------
Classify ImageNet classes, see `tutorial_models_resnet50.py`
>>> # get the whole model with pretrained weights
>>> resnet = tl.models.ResNet50(pretrained=True)
>>> # use for inferencing
>>> output = resnet(img1, is_train=False)
>>> prob = tf.nn.softmax(output)[0].numpy()
Extract the features before fc layer
>>> resnet = tl.models.ResNet50(pretrained=True, end_with='5c')
>>> output = resnet(img1, is_train=False)
Returns
-------
ResNet50 model.
"""
ni = Input([None, 224, 224, 3], name="input")
n = Conv2d(64, (7, 7),
strides=(2, 2),
padding='SAME',
W_init=tf.initializers.he_normal(),
name='conv1')(ni)
n = BatchNorm(name='bn_conv1', act='relu')(n)
n = MaxPool2d((3, 3), strides=(2, 2), name='max_pool1')(n)
for i, block_name in enumerate(block_names):
if len(block_name) == 2:
stage = int(block_name[0])
block = block_name[1]
if block == 'a':
strides = (1, 1) if stage == 2 else (2, 2)
n = conv_block(n,
3,
block_filters[stage - 2],
stage=stage,
block=block,
strides=strides)
else:
n = identity_block(n,
3,
block_filters[stage - 2],
stage=stage,
block=block)
elif block_name == 'avg_pool':
n = GlobalMeanPool2d(name='avg_pool')(n)
elif block_name == 'fc1000':
n = Dense(n_classes, name='fc1000')(n)
if block_name == end_with:
break
network = Model(inputs=ni, outputs=n, name=name)
if pretrained:
restore_params(network)
return network
def squeezenetv1(cls, x, end_with='output', is_train=False, reuse=None):
with tf.compat.v1.variable_scope("squeezenetv1", reuse=reuse):
with tf.compat.v1.variable_scope("input"):
n = InputLayer(x)
# n = Conv2d(n, 96, (7,7),(2,2),tf.nn.relu,'SAME',name='conv1')
n = Conv2d(n,
64, (3, 3), (2, 2),
tf.nn.relu,
'SAME',
name='conv1')
n = MaxPool2d(n, (3, 3), (2, 2), 'VALID', name='max')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("fire2"):
n = Conv2d(n,
16, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
64, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
64, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("fire3"):
n = Conv2d(n,
16, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
64, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
64, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
n = MaxPool2d(n, (3, 3), (2, 2), 'VALID', name='max')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("fire4"):
n = Conv2d(n,
32, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
128, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
128, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("fire5"):
n = Conv2d(n,
32, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
128, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
128, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
n = MaxPool2d(n, (3, 3), (2, 2), 'VALID', name='max')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("fire6"):
n = Conv2d(n,
48, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
192, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
192, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("fire7"):
n = Conv2d(n,
48, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
192, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
192, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("fire8"):
n = Conv2d(n,
64, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
256, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
256, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("fire9"):
n = Conv2d(n,
64, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
256, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
256, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
if end_with in n.outputs.name:
return n
with tf.compat.v1.variable_scope("output"):
n = DropoutLayer(n,
keep=0.5,
is_fix=True,
is_train=is_train,
name='drop1')
n = Conv2d(n,
1000, (1, 1), (1, 1),
padding='VALID',
name='conv10') # 13, 13, 1000
n = GlobalMeanPool2d(n)
if end_with in n.outputs.name:
return n
raise Exception("end_with : input, fire2, fire3 ... fire9, output")
def SqueezeNetV1(pretrained=False, end_with='out', name=None):
"""Pre-trained SqueezeNetV1 model (static mode). Input shape [?, 224, 224, 3], value range [0, 1].
Parameters
------------
pretrained : boolean
Whether to load pretrained weights. Default False.
end_with : str
The end point of the model [conv1, maxpool1, fire2, fire3, fire4, ..., out]. Default ``out`` i.e. the whole model.
name : None or str
Name for this model.
Examples
---------
Classify ImageNet classes, see `tutorial_models_squeezenetv1.py <https://github.com/tensorlayer/tensorlayer/blob/master/example/tutorial_models_squeezenetv1.py>`__
>>> # get the whole model
>>> squeezenet = tl.models.SqueezeNetV1(pretrained=True)
>>> # use for inferencing
>>> output = squeezenet(img1, is_train=False)
>>> prob = tf.nn.softmax(output)[0].numpy()
Extract features and Train a classifier with 100 classes
>>> # get model without the last layer
>>> cnn = tl.models.SqueezeNetV1(pretrained=True, end_with='drop1').as_layer()
>>> # add one more layer and build new model
>>> ni = Input([None, 224, 224, 3], name="inputs")
>>> nn = cnn(ni)
>>> nn = Conv2d(100, (1, 1), (1, 1), padding='VALID', name='conv10')(nn)
>>> nn = GlobalMeanPool2d(name='globalmeanpool')(nn)
>>> model = tl.models.Model(inputs=ni, outputs=nn)
>>> # train your own classifier (only update the last layer)
>>> train_params = model.get_layer('conv10').trainable_weights
Returns
-------
static SqueezeNetV1.
"""
ni = Input([None, 224, 224, 3], name="input")
n = Lambda(lambda x: x * 255, name='scale')(ni)
for i in range(len(layer_names)):
if layer_names[i] == 'conv1':
n = Conv2d(64, (3, 3), (2, 2), tf.nn.relu, 'SAME', name='conv1')(n)
elif layer_names[i] == 'maxpool1':
n = MaxPool2d((3, 3), (2, 2), 'VALID', name='maxpool1')(n)
elif layer_names[i] == 'drop1':
n = Dropout(keep=0.5, name='drop1')(n)
elif layer_names[i] == 'out':
n = Conv2d(1000, (1, 1), (1, 1), padding='VALID', name='conv10')(n) # 13, 13, 1000
n = GlobalMeanPool2d(name='globalmeanpool')(n)
elif layer_names[i] in ['fire3', 'fire5']:
n = fire_block(n, n_filters[i - 2], max_pool=True, name=layer_names[i])
else:
n = fire_block(n, n_filters[i - 2], max_pool=False, name=layer_names[i])
if layer_names[i] == end_with:
break
network = Model(inputs=ni, outputs=n, name=name)
if pretrained:
restore_params(network)
return network
def squeezenet(x, is_train=True, reuse=False):
# model from: https://github.com/wohlert/keras-squeezenet
# https://github.com/DT42/squeezenet_demo/blob/master/model.py
with tf.variable_scope("squeezenet", reuse=reuse):
with tf.variable_scope("input"):
n = InputLayer(x)
# n = Conv2d(n, 96, (7,7),(2,2),tf.nn.relu,'SAME',name='conv1')
n = Conv2d(n, 64, (3, 3), (2, 2), tf.nn.relu, 'SAME', name='conv1')
n = MaxPool2d(n, (3, 3), (2, 2), 'VALID', name='max')
with tf.variable_scope("fire2"):
n = Conv2d(n,
16, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
64, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
64, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
with tf.variable_scope("fire3"):
n = Conv2d(n,
16, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
64, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
64, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
n = MaxPool2d(n, (3, 3), (2, 2), 'VALID', name='max')
with tf.variable_scope("fire4"):
n = Conv2d(n,
32, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
128, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
128, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
with tf.variable_scope("fire5"):
n = Conv2d(n,
32, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
128, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
128, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
n = MaxPool2d(n, (3, 3), (2, 2), 'VALID', name='max')
with tf.variable_scope("fire6"):
n = Conv2d(n,
48, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
192, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
192, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
with tf.variable_scope("fire7"):
n = Conv2d(n,
48, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
192, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
192, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
with tf.variable_scope("fire8"):
n = Conv2d(n,
64, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
256, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
256, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
with tf.variable_scope("fire9"):
n = Conv2d(n,
64, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='squeeze1x1')
n1 = Conv2d(n,
256, (1, 1), (1, 1),
tf.nn.relu,
'SAME',
name='expand1x1')
n2 = Conv2d(n,
256, (3, 3), (1, 1),
tf.nn.relu,
'SAME',
name='expand3x3')
n = ConcatLayer([n1, n2], -1, name='concat')
with tf.variable_scope("output"):
n = DropoutLayer(n,
keep=0.5,
is_fix=True,
is_train=is_train,
name='drop1')
n = Conv2d(n, 1000, (1, 1), (1, 1), padding='VALID',
name='conv10') # 13, 13, 1000
n = GlobalMeanPool2d(n)
return n