def generator(self, z):
with tf.variable_scope("generator") as scope:
self.z_, self.h0_w, self.h0_b = linear(z, self.gf_dim*8*4*4, 'g_h0_lin', with_w=True)
# TODO: Nicer iteration pattern here. #readability
hs = [None]
hs[0] = tf.reshape(self.z_, [-1, 4, 4, self.gf_dim * 8])
hs[0] = tf.nn.relu(self.g_bns[0](hs[0], self.training_bool))
i = 1 # Iteration number.
depth_mul = 8 # Depth decreases as spatial component increases.
size = 8 # Size increases as depth decreases.
# 4 convolutional layers as well..for 64x64
while size < self.image_size:
hs.append(None)
name = 'g_h{}'.format(i)
hs[i], _, _ = conv2d_transpose(hs[i-1],
[self.batch_size, size, size, self.gf_dim*depth_mul], name=name, with_w=True)
hs[i] = tf.nn.relu(self.g_bns[i](hs[i], self.training_bool))
i += 1
depth_mul //= 2
size *= 2
hs.append(None)
name = 'g_h{}'.format(i)
hs[i], _, _ = conv2d_transpose(hs[i - 1],
[self.batch_size, size, size, 3], name=name, with_w=True)
return tf.nn.tanh(hs[i])
def generator_context(z):
n = 32
with arg_scope([conv2d, conv2d_transpose], batch_norm_params=batch_norm_params, stddev=0.02):
z = z*2-1
d = 8
z = fc(z, num_units_out=d*d*32, batch_norm_params=batch_norm_params)
c = z.get_shape()[1].value / (d*d)
z = tf.reshape(z, (-1, d, d, c))
o = conv2d_transpose(z, n, (3, 3), stride=(2, 2))
o = conv2d_transpose(o, n, (3, 3), stride=(2, 2))
o = conv2d(o, num_filters_out=n, kernel_size=(3, 3), stride=1)
o = conv2d(o, num_filters_out=4, kernel_size=(3, 3), stride=1)
attended = o
return attended
def generator(self, images):
with tf.variable_scope("generator"):
g_h0 = tf.nn.relu(conv2d(images, 16, name='g_encode_0'))
g_h1 = tf.nn.relu(conv2d(g_h0, 32, name='g_encode_1'))
g_h2 = tf.nn.relu(conv2d(g_h1, 64, name='g_encode_2'))
g_flat = tf.reshape(g_h2, [self.batch_size, -1])
g_encode = linear(g_flat, 128, 'g_encode')
g_decode = linear(g_encode, 512 * 4 * 4, 'g_h0')
g_h3 = tf.nn.relu(tf.reshape(g_decode, [self.batch_size, 4, 4, 512]))
g_h4 = tf.nn.relu(conv2d_transpose(g_h3, [self.batch_size, 8, 8, 256], name='g_h1'))
g_h5 = tf.nn.relu(conv2d_transpose(g_h4, [self.batch_size, 16, 16, 128], name='g_h2'))
g_h6 = tf.nn.relu(conv2d_transpose(g_h5, [self.batch_size, 32, 32, 64], name='g_h3'))
g_h7 = conv2d_transpose(g_h6, [self.batch_size, 64, 64, 3], name='g_h4')
return tf.nn.tanh(g_h7)
def generator(z):
n = 32
with arg_scope([conv2d, conv2d_transpose], batch_norm_params=batch_norm_params, stddev=0.02):
z = z*2-1
d = 8
z = fc(z, num_units_out=d*d*32, batch_norm_params=batch_norm_params)
c = z.get_shape()[1].value / (d*d)
z = tf.reshape(z, (-1, d, d, c))
o = conv2d_transpose(z, n, (3, 3), stride=(2, 2))
o = conv2d_transpose(o, n, (3, 3), stride=(2, 2))
o = conv2d(o, num_filters_out=n*2, kernel_size=(3, 3), stride=1)
o = conv2d(o, num_filters_out=1, kernel_size=(3, 3), stride=1, padding="VALID", batch_norm_params=None)
out = o[:, 1:29, 1:29, :]
return out
def _generator_dcgan(inpt, gf_dim, is_training):
with tf.variable_scope('Generator'):
z, h0_w, h0_b = ops.linear(inpt, gf_dim * 8 * 4 * 4, scope='g0_dcgan',
with_w=True)
hs = [None]
hs[0] = tf.reshape(z, [-1, 4, 4, gf_dim * 8])
# batch norm parameter
decay = 0.9
epsilon = 1e-5
# do batch norm
bn = tf.contrib.layers.batch_norm(hs[0], decay=decay,
updates_collections=None,
epsilon=epsilon,
center=True, scale=True,
is_training=is_training,
scope='g_bn0')
hs[0] = tf.nn.relu(bn)
i = 1 # Iteration number.
depth_mul = 8 # Depth decreases as spatial component increases.
size = 8 # Size increases as depth decreases.
while size < gf_dim:
hs.append(None)
name = 'g{}_dcgan'.format(i)
hs[i], _, _ = ops.conv2d_transpose(hs[i - 1],
[batch_size, size, size,
gf_dim * depth_mul], scope=name,
with_w=True)
bn = tf.contrib.layers.batch_norm(hs[i], decay=decay,
updates_collections=None,
epsilon=epsilon,
center=True, scale=True,
is_training=is_training,
scope='g_bn{}'.format(i))
hs[i] = tf.nn.relu(bn)
i += 1
depth_mul //= 2
size *= 2
hs.append(None)
name = 'g{}_dcgan'.format(i)
hs[i], _, _ = ops.conv2d_transpose(hs[i - 1],
[batch_size, size, size, 1], scope=name,
with_w=True)
return tf.nn.tanh(hs[i])
def generator(self, images):
with tf.variable_scope("generator"):
g_h0 = tf.nn.relu(conv2d(images, 64, 5, 5, 1, 1, name='g_h0'))
g_h1 = tf.nn.relu(self.g_bns[0](conv2d(g_h0, 128, 3, 3, 2, 2, name='g_h1')))
g_h2 = tf.nn.relu(self.g_bns[1](conv2d(g_h1, 128, 3, 3, 1, 1, name='g_h2')))
g_h3 = tf.nn.relu(self.g_bns[2](conv2d(g_h2, 256, 3, 3, 2, 2, name='g_h3')))
g_h4 = tf.nn.relu(self.g_bns[3](conv2d(g_h3, 256, 3, 3, 1, 1, name='g_h4')))
g_h5 = tf.nn.relu(self.g_bns[4](conv2d(g_h4, 256, 3, 3, 1, 1, name='g_h5')))
g_h6 = tf.nn.relu(self.g_bns[5](dilated_conv2d(g_h5, 256, 3, 3, 2, name='g_h6')))
g_h7 = tf.nn.relu(self.g_bns[6](dilated_conv2d(g_h6, 256, 3, 3, 4, name='g_h7')))
g_h8 = tf.nn.relu(self.g_bns[7](dilated_conv2d(g_h7, 256, 3, 3, 8, name='g_h8')))
g_h9 = tf.nn.relu(self.g_bns[8](dilated_conv2d(g_h8, 256, 3, 3, 16, name='g_h9')))
g_h10 = tf.nn.relu(self.g_bns[9](conv2d(g_h9, 256, 3, 3, 1, 1, name='g_h10')))
g_h11 = tf.nn.relu(self.g_bns[10](conv2d(g_h10, 256, 3, 3, 1, 1, name='g_h11')))
g_h12 = tf.nn.relu(self.g_bns[11](conv2d_transpose(
g_h11, [self.batch_size, int(self.image_size/2), int(self.image_size/2), 128], 4, 4, 2, 2, name='g_h12')))
g_h13 = tf.nn.relu(self.g_bns[12](conv2d(g_h12, 128, 3, 3, 1, 1, name='g_h13')))
g_h14 = tf.nn.relu(self.g_bns[13](conv2d_transpose(
g_h13, [self.batch_size, self.image_size, self.image_size, 64], 4, 4, 2, 2, name='g_h14')))
g_h15 = tf.nn.relu(self.g_bns[14](conv2d(g_h14, 32, 3, 3, 1, 1, name='g_h15')))
g_h16 = tf.nn.sigmoid(conv2d(g_h15, 3, 3, 3, 1, 1, name='g_h16'))
return g_h16
def generator(self, images):
with tf.variable_scope("generator"):
g_h0 = tf.nn.relu(conv2d(images, 16, name='g_encode_0'))
g_h1 = tf.nn.relu(conv2d(g_h0, 32, name='g_encode_1'))
g_h2 = tf.nn.relu(conv2d(g_h1, 64, name='g_encode_2'))
g_flat = tf.reshape(g_h2, [self.batch_size, -1])
g_encode = linear(g_flat, 128, 'g_encode')
g_decode = linear(g_encode, 512 * 4 * 4, 'g_h0')
g_h3 = tf.nn.relu(
tf.reshape(g_decode, [self.batch_size, 4, 4, 512]))
g_h4 = tf.nn.relu(
conv2d_transpose(g_h3, [self.batch_size, 8, 8, 256],
name='g_h1'))
g_h5 = tf.nn.relu(
conv2d_transpose(g_h4, [self.batch_size, 16, 16, 128],
name='g_h2'))
g_h6 = tf.nn.relu(
conv2d_transpose(g_h5, [self.batch_size, 32, 32, 64],
name='g_h3'))
g_h7 = conv2d_transpose(g_h6, [self.batch_size, 64, 64, 3],
name='g_h4')
return tf.nn.tanh(g_h7)
def generator(self, images):
with tf.variable_scope("generator"):
g_h0 = tf.nn.relu(conv2d(images, 64, 5, 5, 1, 1, name='g_h0'))
g_h1 = tf.nn.relu(self.g_bns[0](conv2d(g_h0, 128, 3, 3, 2, 2, name='g_h1')))
g_h2 = tf.nn.relu(self.g_bns[1](conv2d(g_h1, 128, 3, 3, 1, 1, name='g_h2')))
g_h3 = tf.nn.relu(self.g_bns[2](conv2d(g_h2, 256, 3, 3, 2, 2, name='g_h3')))
g_h4 = tf.nn.relu(self.g_bns[3](conv2d(g_h3, 256, 3, 3, 1, 1, name='g_h4')))
g_h5 = tf.nn.relu(self.g_bns[4](conv2d(g_h4, 256, 3, 3, 1, 1, name='g_h5')))
g_h6 = tf.nn.relu(self.g_bns[5](dilated_conv2d(g_h5, 256, 3, 3, 2, name='g_h6')))
g_h7 = tf.nn.relu(self.g_bns[6](dilated_conv2d(g_h6, 256, 3, 3, 4, name='g_h7')))
g_h8 = tf.nn.relu(self.g_bns[7](dilated_conv2d(g_h7, 256, 3, 3, 8, name='g_h8')))
g_h9 = tf.nn.relu(self.g_bns[8](dilated_conv2d(g_h8, 256, 3, 3, 16, name='g_h9')))
g_h10 = tf.nn.relu(self.g_bns[9](conv2d(g_h9, 256, 3, 3, 1, 1, name='g_h10')))
g_h11 = tf.nn.relu(self.g_bns[10](conv2d(g_h10, 256, 3, 3, 1, 1, name='g_h11')))
g_h12 = tf.nn.relu(self.g_bns[11](conv2d_transpose(
g_h11, [self.batch_size, int(self.image_size/2), int(self.image_size/2), 128], 4, 4, 2, 2, name='g_h12')))
g_h13 = tf.nn.relu(self.g_bns[12](conv2d(g_h12, 128, 3, 3, 1, 1, name='g_h13')))
g_h14 = tf.nn.relu(self.g_bns[13](conv2d_transpose(
g_h13, [self.batch_size, self.image_size, self.image_size, 64], 4, 4, 2, 2, name='g_h14')))
g_h15 = tf.nn.relu(self.g_bns[14](conv2d(g_h14, 32, 3, 3, 1, 1, name='g_h15')))
g_h16 = tf.nn.sigmoid(conv2d(g_h15, 3, 3, 3, 1, 1, name='g_h16'))
return g_h16
def decode(fm, training):
fm = tf.reshape(fm, [-1, 2, 2, 256])
network = conv2d_transpose('dcnv1', fm, 128, training, bn=False)
network = conv2d_transpose('dcnv2', network, 64, training, bn=False)
network = conv2d_transpose('dcnv3', network, 32, training, bn=False)
network = conv2d_transpose('dcnv4', network, 16, training, bn=False)
network = conv2d_transpose('dcnv5', network, 8, training, bn=False)
network = conv2d_transpose('dcnv6', network, 4, training, bn=False)
network = conv2d_transpose('dcnv7',
network,
3,
training,
activation_fn=tf.nn.sigmoid,
bn=False)
return network
def inference1(img_l_batch,
img_l_gra_batch,
theme_ab_batch,
theme_mask_batch,
local_ab_batch,
local_mask_batch,
is_training=True,
scope_name='UserGuide'):
"""
:param img_l_batch: l channel of input image
:param img_l_gra_batch: sobel edge map of l channel of input image
:param theme_ab_batch: ab channel of input color theme
:param theme_mask_batch: theme mask
:param local_ab_batch: ab channel of input local points
:param local_mask_batch: local points mask
:param is_training: bool, indicate usage of model (training or testing)
:param scope_name: model name
:return: ab channel of output image
"""
assert img_l_batch.get_shape()[-1] == 1
assert img_l_gra_batch.get_shape(
)[-1] == 1 # horizontal and vertical direction
assert theme_ab_batch.get_shape()[-1] == 2
assert theme_mask_batch.get_shape()[-1] == 1
assert local_ab_batch.get_shape()[-1] == 2
assert local_mask_batch.get_shape()[-1] == 1
ngf = 64
theme_batch = tf.concat([theme_ab_batch, theme_mask_batch], axis=3)
local_batch = tf.concat([local_ab_batch, local_mask_batch], axis=3)
print('Image Inputs:', img_l_batch)
print('Theme Inputs:', theme_batch)
print('Points Inputs:', local_batch)
print()
with tf.variable_scope(scope_name, reuse=tf.AUTO_REUSE):
theme_batch = tf.reshape(theme_batch,
[img_l_batch.get_shape()[0], 1, 1, -1])
glob_conv1 = ops.conv2d(theme_batch,
ngf * 8,
1,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='glob_conv1')
glob_conv2 = ops.conv2d(glob_conv1,
ngf * 8,
1,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='glob_conv2')
glob_conv3 = ops.conv2d(glob_conv2,
ngf * 8,
1,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='glob_conv3')
glob_conv4 = ops.conv2d(glob_conv3,
ngf * 8,
1,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='glob_conv4')
print('ThemeBlock', glob_conv4)
ab_conv1_1 = ops.conv2d(local_batch,
ngf,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='ab_conv1_1')
bw_conv1_1 = ops.conv2d(img_l_batch,
ngf,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='bw_conv1_1')
gra_conv1_1 = ops.conv2d(img_l_gra_batch,
ngf,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='gra_conv1_1')
print('LocalBlock', gra_conv1_1)
conv1_1 = ab_conv1_1 + bw_conv1_1 + gra_conv1_1 # TODO: Merge Local Points and Gradient Maps
conv1_1 = ops.conv2d(conv1_1,
ngf,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv1_1')
conv1_2 = ops.conv2d(conv1_1,
ngf,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv1_2')
conv1_2_ss = ops.depth_wise_conv2d(conv1_2,
1,
1,
2,
activation_fn=None,
scope_name='conv1_2_ss')
print('ConvBlock 1', conv1_2_ss)
conv2_1 = ops.conv2d(conv1_2_ss,
ngf * 2,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv2_1')
conv2_2 = ops.conv2d(conv2_1,
ngf * 2,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv2_2')
conv2_2_ss = ops.depth_wise_conv2d(conv2_2,
1,
1,
2,
activation_fn=None,
scope_name='conv2_2_ss')
print('ConvBlock 2', conv2_2_ss)
conv3_1 = ops.conv2d(conv2_2_ss,
ngf * 4,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv3_1')
conv3_2 = ops.conv2d(conv3_1,
ngf * 4,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv3_2')
conv3_3 = ops.conv2d(conv3_2,
ngf * 4,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv3_3')
conv3_3_ss = ops.depth_wise_conv2d(conv3_3,
1,
1,
2,
activation_fn=None,
scope_name='conv3_3_ss')
print('ConvBlock 3', conv3_3_ss)
conv4_1 = ops.conv2d(conv3_3_ss,
ngf * 8,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv4_1')
conv4_2 = ops.conv2d(conv4_1,
ngf * 8,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv4_2')
conv4_3 = ops.conv2d(conv4_2,
ngf * 8,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv4_3')
print('ConvBlock 4', conv4_3)
conv4_3 = conv4_3 + glob_conv4 # TODO: Merge Color Theme
conv5_1 = ops.conv2d(conv4_3,
ngf * 8,
3,
1,
dilation=2,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv5_1')
conv5_2 = ops.conv2d(conv5_1,
ngf * 8,
3,
1,
dilation=2,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv5_2')
conv5_3 = ops.conv2d(conv5_2,
ngf * 8,
3,
1,
dilation=2,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv5_3')
print('ConvBlock 5', conv5_3)
conv6_1 = ops.conv2d(conv5_3,
ngf * 8,
3,
1,
dilation=2,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv6_1')
conv6_2 = ops.conv2d(conv6_1,
ngf * 8,
3,
1,
dilation=2,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv6_2')
conv6_3 = ops.conv2d(conv6_2,
ngf * 8,
3,
1,
dilation=2,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv6_3')
print('ConvBlock 6', conv6_3)
conv7_1 = ops.conv2d(conv6_3,
ngf * 8,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv7_1')
conv7_2 = ops.conv2d(conv7_1,
ngf * 8,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv7_2')
conv7_3 = ops.conv2d(conv7_2,
ngf * 8,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv7_3')
print('ConvBlock 7', conv7_3)
conv3_3_short = ops.conv2d(conv3_3,
ngf * 4,
3,
1,
activation_fn=None,
is_training=is_training,
scope_name='conv3_3_short')
conv8_1 = ops.conv2d_transpose(conv7_3,
ngf * 4,
4,
2,
activation_fn=None,
is_training=is_training,
scope_name='conv8_1')
conv8_1_comb = tf.nn.relu(conv3_3_short + conv8_1)
conv8_2 = ops.conv2d(conv8_1_comb,
ngf * 4,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=None,
is_training=is_training,
scope_name='conv8_2')
conv8_3 = ops.conv2d(conv8_2,
ngf * 4,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv8_3')
print('ConvBlock 8', conv8_3)
conv2_2_short = ops.conv2d(conv2_2,
ngf * 2,
3,
1,
activation_fn=None,
is_training=is_training,
scope_name='conv2_2_short')
conv9_1 = ops.conv2d_transpose(conv8_3,
ngf * 2,
4,
2,
activation_fn=None,
is_training=is_training,
scope_name='conv9_1')
conv9_1_comb = tf.nn.relu(conv2_2_short + conv9_1)
conv9_2 = ops.conv2d(conv9_1_comb,
ngf * 2,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv9_2')
print('ConvBlock 9', conv9_2)
conv1_2_short = ops.conv2d(conv1_2,
ngf * 2,
3,
1,
activation_fn=None,
is_training=is_training,
scope_name='conv1_2_short')
conv10_1 = ops.conv2d_transpose(conv9_2,
ngf * 2,
4,
2,
activation_fn=None,
is_training=is_training,
scope_name='conv10_1')
conv10_1_comb = tf.nn.relu(conv1_2_short + conv10_1)
conv10_2 = ops.conv2d(conv10_1_comb,
ngf * 2,
3,
1,
activation_fn=tf.nn.relu,
norm_fn=tf.layers.batch_normalization,
is_training=is_training,
scope_name='conv10_2')
print('ConvBlock 10', conv10_2)
conv10_ab = ops.conv2d(conv10_2,
2,
1,
1,
activation_fn=tf.nn.tanh,
norm_fn=None,
is_training=is_training,
scope_name='conv10_ab')
print('OutputBlock', conv10_ab, end='\n\n')
return conv10_ab
def generator(z, labels):
with tf.variable_scope("generator") as scope:
# concat z and labels
z_labels = tf.concat([z, labels], 1)
# project z and reshape
oh, ow = FLAGS.output_height, FLAGS.output_width
z_labels_ = ops.fc(z_labels, 1024 * oh / 32 * ow / 32, scope="project")
z_labels_ = tf.reshape(z_labels_, [-1, oh / 32, ow / 32, 1024])
# batch norm
norm0 = ops.batch_norm(z_labels_,
scope="batch_norm0",
is_training=True)
# ReLU
h0 = tf.nn.relu(norm0)
# conv1
conv1 = ops.conv2d_transpose(h0,
[FLAGS.batch_size, oh / 16, ow / 16, 512],
scope="conv_tranpose1")
# batch norm
norm1 = ops.batch_norm(conv1,
scope="batch_norm1",
is_training=FLAGS.is_train)
# ReLU
h1 = tf.nn.relu(norm1)
# conv2
conv2 = ops.conv2d_transpose(h1,
[FLAGS.batch_size, oh / 8, ow / 8, 256],
scope="conv_tranpose2")
# batch norm
norm2 = ops.batch_norm(conv2,
scope="batch_norm2",
is_training=FLAGS.is_train)
# ReLU
h2 = tf.nn.relu(norm2)
# conv3
conv3 = ops.conv2d_transpose(h2,
[FLAGS.batch_size, oh / 4, ow / 4, 128],
scope="conv_tranpose3")
# batch norm
norm3 = ops.batch_norm(conv3,
scope="batch_norm3",
is_training=FLAGS.is_train)
# ReLU
h3 = tf.nn.relu(norm3)
# conv4
conv4 = ops.conv2d_transpose(h3,
[FLAGS.batch_size, oh / 2, ow / 2, 64],
scope="conv_tranpose4")
# batch norm
norm4 = ops.batch_norm(conv4,
scope="batch_norm4",
is_training=FLAGS.is_train)
# ReLU
h4 = tf.nn.relu(norm4)
conv5 = ops.conv2d_transpose(
h4, [FLAGS.batch_size, oh, ow, FLAGS.input_channels],
scope="conv_tranpose5")
# tanh
h5 = tf.nn.tanh(conv5)
h5 = tf.map_fn(
lambda i: tf.image.resize_images(
i, [FLAGS.input_height, FLAGS.input_width]), h5)
return h5
'''
delta = numpy.add.reduce(weighted, 3)
image = partial_add(image, delta, [i, j])
if padding > 0:
return image[padding:-padding,padding:-padding]
else:
return image
# value [-1, 14, 14, 64]
# fil [5, 5, 32, 64]
p = np.array([1.0, 2.0, 3.0])
arr = np.array([[p,p],[p,p]])
a_value = np.array([arr, arr, arr, arr])
l = np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
a_fil = np.array([[l]])
value = tf.constant(a_value, dtype=tf.float32)
fil = tf.constant(a_fil, dtype=tf.float32)
#value = np.array([[[[1]],[[2]]],[[[3]],[[4]]]])
#fil = np.array([[[[2]]]])
#result = dconv(value, fil, padding=0)
result = dconv(a_value, a_fil, padding=0) # nn_ops.deconv2d(value, fil, [-1,2,2,2], [1,1,1,1])
print(result)
#result = ops.partial_add(tf.constant([[1,1,1],[1,1,1],[1,1,1]],dtype=tf.float32), tf.constant([[1,1],[1,1]], dtype=tf.float32), [0,0])
result = ops.conv2d_transpose(value, fil, padding=0)
print(result)
print(result.eval())
def generator(self, images):
"""
This is the completion network
:param images: 4D-tensor, shape is [batch, height, width, channels]
:return: 4D-tensor, shape is [batch, height, width, channels], completed images tensor
"""
with tf.variable_scope("generator"):
g_h0 = tf.nn.relu(conv2d(images, 64, 5, 5, 1, 1, name='g_h0'))
g_h1 = tf.nn.relu(self.g_bns[0](conv2d(g_h0,
128,
3,
3,
2,
2,
name='g_h1')))
g_h2 = tf.nn.relu(self.g_bns[1](conv2d(g_h1,
128,
3,
3,
1,
1,
name='g_h2')))
g_h3 = tf.nn.relu(self.g_bns[2](conv2d(g_h2,
256,
3,
3,
2,
2,
name='g_h3')))
g_h4 = tf.nn.relu(self.g_bns[3](conv2d(g_h3,
256,
3,
3,
1,
1,
name='g_h4')))
g_h5 = tf.nn.relu(self.g_bns[4](conv2d(g_h4,
256,
3,
3,
1,
1,
name='g_h5')))
g_h6 = tf.nn.relu(self.g_bns[5](dilated_conv2d(g_h5,
256,
3,
3,
2,
name='g_h6')))
g_h7 = tf.nn.relu(self.g_bns[6](dilated_conv2d(g_h6,
256,
3,
3,
4,
name='g_h7')))
g_h8 = tf.nn.relu(self.g_bns[7](dilated_conv2d(g_h7,
256,
3,
3,
8,
name='g_h8')))
g_h9 = tf.nn.relu(self.g_bns[8](dilated_conv2d(g_h8,
256,
3,
3,
16,
name='g_h9')))
g_h10 = tf.nn.relu(self.g_bns[9](conv2d(g_h9,
256,
3,
3,
1,
1,
name='g_h10')))
g_h11 = tf.nn.relu(self.g_bns[10](conv2d(g_h10,
256,
3,
3,
1,
1,
name='g_h11')))
g_h12 = tf.nn.relu(self.g_bns[11](conv2d_transpose(g_h11, [
self.batch_size,
int(self.image_size / 2),
int(self.image_size / 2), 128
],
4,
4,
2,
2,
name='g_h12')))
g_h13 = tf.nn.relu(self.g_bns[12](conv2d(g_h12,
128,
3,
3,
1,
1,
name='g_h13')))
g_h14 = tf.nn.relu(self.g_bns[13](conv2d_transpose(
g_h13, [self.batch_size, self.image_size, self.image_size, 64],
4,
4,
2,
2,
name='g_h14')))
g_h15 = tf.nn.relu(self.g_bns[14](conv2d(g_h14,
32,
3,
3,
1,
1,
name='g_h15')))
g_h16 = tf.nn.sigmoid(conv2d(g_h15, 3, 3, 3, 1, 1, name='g_h16'))
return g_h16
def create(self,
inputs,
kernel_size=None,
seed=None,
reuse_variables=None):
output = inputs
with tf.variable_scope(self.name, reuse=reuse_variables):
layers = []
# encoder branch
for index, kernel in enumerate(self.encoder_kernels):
name = 'conv' + str(index)
output = conv2d(inputs=output,
name=name,
kernel_size=kernel_size,
filters=kernel[0],
strides=kernel[1],
activation=tf.nn.leaky_relu,
seed=seed)
# save contracting path layers to be used for skip connections
layers.append(output)
if kernel[2] > 0:
keep_prob = 1.0 - kernel[2] if self.training else 1.0
output = tf.nn.dropout(output,
keep_prob=keep_prob,
name='dropout_' + name,
seed=seed)
# decoder branch
for index, kernel in enumerate(self.decoder_kernels):
name = 'deconv' + str(index)
output = conv2d_transpose(inputs=output,
name=name,
kernel_size=kernel_size,
filters=kernel[0],
strides=kernel[1],
activation=tf.nn.relu,
seed=seed)
if kernel[2] > 0:
keep_prob = 1.0 - kernel[2] if self.training else 1.0
output = tf.nn.dropout(output,
keep_prob=keep_prob,
name='dropout_' + name,
seed=seed)
# concat the layer from the contracting path with the output of the current layer
# concat only the channels (axis=3)
#将紧缩路径中的图层与当前图层的输出连接
#仅连接通道（轴= 3）
output = tf.concat([layers[len(layers) - index - 2], output],
axis=3)
output = conv2d(
inputs=output,
name='conv_last',
filters=self.output_channels, # number of output chanels
kernel_size=1, # last layer kernel size = 1
strides=1, # last layer stride = 1
bnorm=False, # do not use batch-norm for the last layer
activation=tf.nn.
tanh, # tanh activation function for the output
seed=seed)
self.var_list = tf.get_collection(tf.GraphKeys.TRAINABLE_VARIABLES,
self.name)
return output
def generator(z, labels):
with tf.variable_scope("generator") as scope:
# labels to one_hot
labels_one_hot = tf.one_hot(labels, FLAGS.n_classes)
# concat z and labels
#z_labels = tf.concat(1, [z, labels_one_hot])
z_labels = tf.concat( [z, labels_one_hot], 1)
# project z and reshape
oh, ow = FLAGS.output_height, FLAGS.output_width
z_labels_ = ops.fc(z_labels, 512 * oh / 16 * ow / 16, scope="project")
z_labels_ = tf.reshape(z_labels_, [-1, oh / 16, ow / 16, 512])
# batch norm
norm0 = ops.batch_norm(
z_labels_, scope="batch_norm0", is_training=True)
# ReLU
h0 = tf.nn.relu(norm0)
# conv1
conv1 = ops.conv2d_transpose(
h0, [FLAGS.batch_size, oh / 8, ow / 8, 256],
scope="conv_tranpose1")
# batch norm
norm1 = ops.batch_norm(conv1, scope="batch_norm1", is_training=True)
# ReLU
h1 = tf.nn.relu(norm1)
# conv2
conv2 = ops.conv2d_transpose(
h1, [FLAGS.batch_size, oh / 4, ow / 4, 128],
scope="conv_tranpose2")
# batch norm
norm2 = ops.batch_norm(conv2, scope="batch_norm2", is_training=True)
# ReLU
h2 = tf.nn.relu(norm2)
# conv3
conv3 = ops.conv2d_transpose(
h2, [FLAGS.batch_size, oh / 2, ow / 2, 64], scope="conv_tranpose3")
# batch norm
norm3 = ops.batch_norm(conv3, scope="batch_norm3", is_training=True)
# ReLU
h3 = tf.nn.relu(norm3)
# conv4
conv4 = ops.conv2d_transpose(
h3, [FLAGS.batch_size, oh, ow, FLAGS.input_channels],
scope="conv_tranpose4")
# tanh
h4 = tf.nn.tanh(conv4)
return h4