def __call__(self, x, is_training, reuse):
# return only logits
h = x
with tf.variable_scope(self.name_scope, reuse=reuse):
for i, (in_dim, out_dim) in enumerate(
zip(self.layer_list, self.layer_list[1:-1])):
h = linear_layer(h, in_dim, out_dim, i)
#h = batch_norm(h, i, is_training=is_training)
h = lrelu(h)
ret = linear_layer(h, self.layer_list[-2], self.layer_list[-1],
'output')
return ret
def __call__(self, x, is_training, reuse):
h = x
with tf.variable_scope(self.name_scope, reuse=reuse):
for i, (in_dim, out_dim) in enumerate(
zip(self.layer_list, self.layer_list[1:-1])):
h = linear_layer(h, in_dim, out_dim, i)
h = batch_norm(h, i, is_training=is_training)
h = lrelu(h)
mu = linear_layer(h, self.layer_list[-2], self.layer_list[-1],
'mu')
log_sigma = linear_layer(h, self.layer_list[-2],
self.layer_list[-1], 'log_sigma')
return mu, log_sigma
def set_model(self, z, labels, batch_size, is_training, reuse=False):
# reshape z
with tf.variable_scope(self.name_scope_reshape, reuse=reuse):
h = linear_layer(z, self.z_dim,
self.in_dim * self.in_dim * self.layer_chanels[0],
'reshape')
h = batch_norm(h, 'reshape', is_training)
h = lrelu(h)
h_z = tf.reshape(h,
[-1, self.in_dim, self.in_dim, self.layer_chanels[0]])
# reshape labels
with tf.variable_scope(self.name_scope_label, reuse=reuse):
h = linear_layer(z, self.z_dim,
self.in_dim * self.in_dim * self.layer_chanels[0],
'label')
h = batch_norm(h, 'label', is_training)
h = lrelu(h)
# concat
h_label = tf.reshape(
h, [-1, self.in_dim, self.in_dim, self.layer_chanels[0]])
h = tf.concat([h_z, h_label], 3)
# deconvolution
layer_num = len(self.layer_chanels) - 1
with tf.variable_scope(self.name_scope_deconv, reuse=reuse):
for i, (in_chan, out_chan) in enumerate(
zip(self.layer_chanels, self.layer_chanels[1:])):
deconved = deconv_layer(inputs=h,
out_shape=[
batch_size,
self.in_dim * 2**(i + 1),
self.in_dim * 2**(i + 1), out_chan
],
filter_width=5,
filter_hight=5,
stride=2,
l_id=i)
if i == layer_num - 1:
h = tf.nn.tanh(deconved)
else:
bn_deconved = batch_norm(deconved, i, is_training)
#h = tf.nn.relu(bn_deconved)
h = lrelu(bn_deconved)
return h
def set_model(self, figs, is_training, reuse=False):
# return only logits
h = figs
# convolution
with tf.variable_scope(self.name_scope_conv, reuse=reuse):
for i, (in_chan, out_chan) in enumerate(
zip(self.layer_chanels, self.layer_chanels[1:])):
if i == 0:
conved = conv_layer(inputs=h,
out_num=out_chan,
filter_width=5,
filter_hight=5,
stride=1,
l_id=i)
h = tf.nn.relu(conved)
#h = lrelu(conved)
else:
conved = conv_layer(inputs=h,
out_num=out_chan,
filter_width=5,
filter_hight=5,
stride=2,
l_id=i)
bn_conved = batch_norm(conved, i, is_training)
h = tf.nn.relu(bn_conved)
#h = lrelu(bn_conved)
feature_image = h
# full connect
dim = get_dim(h)
h = tf.reshape(h, [-1, dim])
with tf.variable_scope(self.name_scope_fc, reuse=reuse):
h = linear_layer(h, dim, self.fc_dim, 'fc')
h = batch_norm(h, 'fc', is_training)
h = tf.nn.relu(h)
h = linear_layer(h, self.fc_dim, self.class_num, 'fc2')
return h, feature_image
def set_model(self, figs, is_training, reuse = False):
u'''
return only logits.
'''
h = figs
# convolution
with tf.variable_scope(self.name_scope, reuse = reuse):
for i, (in_dim, out_dim) in enumerate(zip(self.layer_list, self.layer_list[1:])):
h = linear_layer(h, in_dim, out_dim, i)
h = batch_norm(h, i, is_training)
h = tf.nn.relu(h)
dim = get_dim(h)
mu = linear_layer(h, dim, self.z_dim, 'mu')
log_sigma = linear_layer(h, dim, self.z_dim, 'log_sigma')
return mu, log_sigma
def set_model(self, z, y, is_training, reuse = False):
# return only logits
h = tf.concat([z, y], 1)
with tf.variable_scope(self.name_scope, reuse = reuse):
for i, (in_dim, out_dim) in enumerate(zip(self.layer_list, self.layer_list[1:])):
ret = linear_layer(h, in_dim, out_dim, i)
h = batch_norm(ret, i, is_training)
h = tf.nn.relu(h)
return ret
def set_model(self, figs, labels, is_training, reuse=False):
fig_shape = figs.get_shape().as_list()
height, width = fig_shape[1:3]
class_num = get_dim(labels)
with tf.variable_scope(self.name_scope_label, reuse=reuse):
tmp = linear_layer(labels, class_num, height * width, 'reshape')
tmp = tf.reshape(tmp, [-1, height, width, 1])
h = tf.concat((figs, tmp), 3)
# convolution
with tf.variable_scope(self.name_scope_conv, reuse=reuse):
for i, (in_chan, out_chan) in enumerate(
zip(self.layer_chanels, self.layer_chanels[1:])):
conved = conv_layer(inputs=h,
out_num=out_chan,
filter_width=5,
filter_hight=5,
stride=2,
l_id=i)
if i == 0:
h = tf.nn.relu(conved)
#h = lrelu(conved)
else:
bn_conved = batch_norm(conved, i, is_training)
h = tf.nn.relu(bn_conved)
#h = lrelu(bn_conved)
# full connect
dim = get_dim(h)
h = tf.reshape(h, [-1, dim])
with tf.variable_scope(self.name_scope_fc, reuse=reuse):
h = linear_layer(h, dim, self.fc_dim, 'fc')
h = batch_norm(h, 'en_fc_bn', is_training)
h = tf.nn.relu(h)
mu = linear_layer(h, self.fc_dim, self.z_dim, 'mu')
log_sigma = linear_layer(h, self.fc_dim, self.z_dim, 'sigma')
return mu, log_sigma
def set_model(self, z, is_training, reuse = False):
u'''
return only logits. not sigmoid(logits).
'''
h = z
with tf.variable_scope(self.name_scope, reuse = reuse):
for i, (in_dim, out_dim) in enumerate(zip(self.layer_list, self.layer_list[1:])):
ret = linear_layer(h, in_dim, out_dim, i)
h = batch_norm(ret, i, is_training)
h = tf.nn.relu(h)
return ret
