def __init__(self,
inputs,
filters,
convs,
weight_decay=0.,
weights=None,
trainable=True):
blocks = []
# Convolutional blocks
for i, (fltr, conv) in enumerate(zip(filters, convs)):
block_name = 'block{}'.format(i + 1)
block = vgg_conv(filters=fltr,
convs=conv,
padding=False,
weight_decay=weight_decay,
block_name=block_name)
blocks.append(block)
# Fully Convolutional block
fc_block = vgg_fc(filters=4096, weight_decay=weight_decay)
blocks.append(fc_block)
super(VGGEncoder, self).__init__(inputs=inputs,
blocks=blocks,
weights=weights,
trainable=trainable)
def test_vgg_conv():
if K.image_data_format() == 'channels_first':
x = Input(shape=(3, 224, 224))
y1_shape = (None, 64, 112, 112)
y2_shape = (None, 128, 56, 56)
else:
x = Input(shape=(224, 224, 3))
y1_shape = (None, 112, 112, 64)
y2_shape = (None, 56, 56, 128)
block1 = vgg_conv(filters=64, convs=2, block_name='block1')
y = block1(x)
assert K.int_shape(y) == y1_shape
block2 = vgg_conv(filters=128, convs=2, block_name='block2')
y = block2(y)
assert K.int_shape(y) == y2_shape
def test_vgg_conv():
if K.image_data_format() == 'channels_first':
x = Input(shape=(3, 224, 224))
y1_shape = (None, 64, 112, 112)
y2_shape = (None, 128, 56, 56)
else:
x = Input(shape=(224, 224, 3))
y1_shape = (None, 112, 112, 64)
y2_shape = (None, 56, 56, 128)
block1 = vgg_conv(filters=64, convs=2, block_name='block1')
y = block1(x)
assert K.int_shape(y) == y1_shape
block2 = vgg_conv(filters=128, convs=2, block_name='block2')
y = block2(y)
assert K.int_shape(y) == y2_shape
def __init__(self, inputs, filters, convs, weight_decay=0.,
weights=None, trainable=True):
blocks = []
# Convolutional blocks
for i, (fltr, conv) in enumerate(zip(filters, convs)):
block_name = 'block{}'.format(i + 1)
block = vgg_conv(filters=fltr, convs=conv, padding=False,
weight_decay=weight_decay,
block_name=block_name)
blocks.append(block)
# Fully Convolutional block
fc_block = vgg_fc(filters=4096, weight_decay=weight_decay)
blocks.append(fc_block)
super(VGGEncoder, self).__init__(inputs=inputs, blocks=blocks,
weights=weights, trainable=trainable)