def build(self):
slab_input = Input(shape=self.input_shapes[0])
slice_input = Input(shape=self.input_shapes[1])
slab_path = CBR(slab_input, 32, (1, 1, 5), padding='valid')
slab_path = CBR(slab_path, 1, (1, 1, 1), padding='valid')
slab_path = Reshape(self.input_shapes[1])(slab_path)
combined = Concatenate(axis=2)([slice_input, slab_path])
block_A = CBR_2D(combined, 16, (3, 3), padding='valid')
block_A = CBR_2D(block_A, 32, (3, 3), padding='valid')
block_B = MaxPooling2D(pool_size=(2, 2))(block_A)
block_B = CBR_2D(block_B, 32, (3, 3), padding='valid')
block_B = CBR_2D(block_B, 64, (3, 3), padding='valid')
block_C = MaxPooling2D(pool_size=(2, 2))(block_B)
block_C = CBR_2D(block_C, 64, (3, 3), padding='valid')
block_C = CBR_2D(block_C, 128, (3, 3), padding='valid')
block_D = MaxPooling2D(pool_size=(2, 2))(block_C)
block_D = CBR_2D(block_D, 128, (3, 3), padding='valid')
block_D = CBR_2D(block_D, 64, (3, 3), padding='valid')
fc = Flatten()(block_D)
fc = Dropout(0.25)(fc)
fc = Dense(256, activation='relu')(fc)
discrim_output = Dense(1, activation='sigmoid')(fc)
self.output_name = discrim_output.name.split('/')[0]
self.model = keras.Model(inputs=[slab_input, slice_input], outputs=[discrim_output], name='network')
def __init__(self, in_ch, out_ch):
super().__init__()
#self.gpu_ids = gpu_ids
self.c0_0 = CBR(in_ch,
32,
bn=False,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c0_1 = CBR(out_ch,
32,
bn=False,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c1 = CBR(64,
128,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c2 = CBR(128,
256,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c3 = CBR(256,
512,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c4 = nn.Conv2d(512, 1, 3, 1, 1)
def build(self):
wmn_input = Input(shape=self.input_shape)
block_A = CBR(wmn_input, 16, (3, 3, 3))
block_A = CBR(block_A, 32, (3, 3, 3))
block_B = MaxPooling3D(pool_size=(2, 2, 1))(block_A)
block_B = CBR(block_B, 32, (3, 3, 3))
block_B = CBR(block_B, 64, (3, 3, 3))
block_C = MaxPooling3D(pool_size=(2, 2, 1))(block_B)
block_C = CBR(block_C, 64, (3, 3, 3))
block_C = CBR(block_C, 128, (3, 3, 3))
block_C = Dropout(rate=self.dropout_rate)(block_C)
block_C = Conv3DTranspose(filters=64, kernel_size=(3, 3, 3), strides=(2, 2, 1), padding='same')(block_C)
block_D = Concatenate()([block_C, block_B])
block_D = CBR(block_D, 64, (3, 3, 3))
block_D = CBR(block_D, 64, (3, 3, 3))
block_D = Conv3DTranspose(filters=32, kernel_size=(3, 3, 3), strides=(2, 2, 1), padding='same')(block_D)
block_E = Concatenate()([block_D, block_A])
block_E = CBR(block_E, 32, (3, 3, 3))
block_E = CBR(block_E, 32, (3, 3, 3))
if self.single_slice_out:
block_E = Conv3D(filters=32, kernel_size=(1, 1, 5))(block_E)
csfn_output = Conv3D(filters=1, kernel_size=(1, 1, 1))(block_E)
csfn_output = Reshape((256, 256, 1))(csfn_output)
else:
csfn_output = Conv3D(filters=1, kernel_size=(1, 1, 1))(block_E)
self.csfn_output_name = csfn_output.name.split('/')[0]
self.model = keras.Model(inputs=[wmn_input], outputs=[csfn_output], name='network')
def __init__(self, in_ch, out_ch):
super().__init__()
self.c0 = nn.Conv2d(in_ch, 64, 3, 1, 1)
self.c1 = CBR(64,
128,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c2 = CBR(128,
256,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c3 = CBR(256,
512,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c4 = CBR(512,
512,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c5 = CBR(512,
512,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c6 = CBR(512,
512,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.c7 = CBR(512,
512,
bn=True,
sample='down',
activation=nn.LeakyReLU(0.2, True),
dropout=False)
self.dc0 = CBR(512,
512,
bn=True,
sample='up',
activation=nn.ReLU(True),
dropout=True)
self.dc1 = CBR(1024,
512,
bn=True,
sample='up',
activation=nn.ReLU(True),
dropout=True)
self.dc2 = CBR(1024,
512,
bn=True,
sample='up',
activation=nn.ReLU(True),
dropout=True)
self.dc3 = CBR(1024,
512,
bn=True,
sample='up',
activation=nn.ReLU(True),
dropout=False)
self.dc4 = CBR(1024,
256,
bn=True,
sample='up',
activation=nn.ReLU(True),
dropout=False)
self.dc5 = CBR(512,
128,
bn=True,
sample='up',
activation=nn.ReLU(True),
dropout=False)
self.dc6 = CBR(256,
64,
bn=True,
sample='up',
activation=nn.ReLU(True),
dropout=False)
self.dc7 = nn.Conv2d(128, out_ch, 3, 1, 1)