def CAM_net2D(x, nlabels, training, scope_reuse=False):
with tf.variable_scope('classifier') as scope:
if scope_reuse:
scope.reuse_variables()
init_filters = 32
conv1_1 = layers.conv2D_layer_bn(x, 'conv1_1', num_filters=init_filters, training=training)
pool1 = layers.maxpool2D_layer(conv1_1)
conv2_1 = layers.conv2D_layer_bn(pool1, 'conv2_1', num_filters=init_filters*2, training=training)
pool2 = layers.maxpool2D_layer(conv2_1)
conv3_1 = layers.conv2D_layer_bn(pool2, 'conv3_1', num_filters=init_filters*4, training=training)
conv3_2 = layers.conv2D_layer_bn(conv3_1, 'conv3_2', num_filters=init_filters*4, training=training)
conv4_1 = layers.conv2D_layer_bn(conv3_2, 'conv4_1', num_filters=init_filters*8, training=training)
conv4_2 = layers.conv2D_layer_bn(conv4_1, 'conv4_2', num_filters=init_filters*8, training=training)
conv5_1 = layers.conv2D_layer_bn(conv4_2, 'conv5_1', num_filters=init_filters*16, training=training)
conv5_2 = layers.conv2D_layer_bn(conv5_1, 'conv5_2', num_filters=init_filters*16, training=training)
convD_1 = layers.conv2D_layer_bn(conv5_2, 'feature_maps', num_filters=init_filters*16, training=training)
fm_averages = layers.averagepool2D_layer(convD_1, name='fm_averages')
logits = layers.dense_layer(fm_averages, 'weight_layer', hidden_units=nlabels, activation=tf.identity, add_bias=False)
return logits
def VGG16(x, nlabels, training, scope_reuse=False):
with tf.variable_scope('classifier') as scope:
if scope_reuse:
scope.reuse_variables()
init_filters = 64
conv1_1 = layers.conv2D_layer_bn(x, 'conv1_1', num_filters=init_filters, training=training)
conv1_2 = layers.conv2D_layer_bn(conv1_1, 'conv1_2', num_filters=init_filters, training=training)
pool1 = layers.maxpool2D_layer(conv1_2)
conv2_1 = layers.conv2D_layer_bn(pool1, 'conv2_1', num_filters=init_filters*2, training=training)
conv2_2 = layers.conv2D_layer_bn(conv2_1, 'conv2_2', num_filters=init_filters*2, training=training)
pool2 = layers.maxpool2D_layer(conv2_2)
conv3_1 = layers.conv2D_layer_bn(pool2, 'conv3_1', num_filters=init_filters*4, training=training)
conv3_2 = layers.conv2D_layer_bn(conv3_1, 'conv3_2', num_filters=init_filters*4, training=training)
conv3_3 = layers.conv2D_layer_bn(conv3_2, 'conv3_3', num_filters=init_filters*4, training=training)
pool3 = layers.maxpool2D_layer(conv3_3)
conv4_1 = layers.conv2D_layer_bn(pool3, 'conv4_1', num_filters=init_filters*8, training=training)
conv4_2 = layers.conv2D_layer_bn(conv4_1, 'conv4_2', num_filters=init_filters*8, training=training)
conv4_3 = layers.conv2D_layer_bn(conv4_2, 'conv4_3', num_filters=init_filters*8, training=training)
pool4 = layers.maxpool2D_layer(conv4_3)
conv5_1 = layers.conv2D_layer_bn(pool4, 'conv5_1', num_filters=init_filters*8, training=training)
conv5_2 = layers.conv2D_layer_bn(conv5_1, 'conv5_2', num_filters=init_filters*8, training=training)
conv5_3 = layers.conv2D_layer_bn(conv5_2, 'conv5_3', num_filters=init_filters*8, training=training)
pool5 = layers.maxpool2D_layer(conv5_3)
dense1 = layers.dense_layer_bn(pool5, 'dense1', hidden_units=init_filters*64, training=training)
dense2 = layers.dense_layer_bn(dense1, 'dense2', hidden_units=init_filters*64, training=training)
logits = layers.dense_layer_bn(dense2, 'dense3', hidden_units=nlabels, training=training, activation=tf.identity)
return logits
def rebuttalnet2D(x, nlabels, training, scope_reuse=False):
with tf.variable_scope('classifier') as scope:
if scope_reuse:
scope.reuse_variables()
init_filters = 32
conv1_1 = layers.conv2D_layer_bn(x, 'conv1_1', num_filters=init_filters, training=training)
pool1 = layers.maxpool2D_layer(conv1_1)
conv2_1 = layers.conv2D_layer_bn(pool1, 'conv2_1', num_filters=init_filters*2, training=training)
pool2 = layers.maxpool2D_layer(conv2_1)
conv3_1 = layers.conv2D_layer_bn(pool2, 'conv3_1', num_filters=init_filters*4, training=training)
conv3_2 = layers.conv2D_layer_bn(conv3_1, 'conv3_2', num_filters=init_filters*4, training=training)
pool3 = layers.maxpool2D_layer(conv3_2)
conv4_1 = layers.conv2D_layer_bn(pool3, 'conv4_1', num_filters=init_filters*8, training=training)
conv4_2 = layers.conv2D_layer_bn(conv4_1, 'conv4_2', num_filters=init_filters*8, training=training)
pool4 = layers.maxpool2D_layer(conv4_2)
conv5_1 = layers.conv2D_layer_bn(pool4, 'conv5_1', num_filters=init_filters*16, training=training)
conv5_2 = layers.conv2D_layer_bn(conv5_1, 'conv5_2', num_filters=init_filters*16, training=training)
convD_1 = layers.conv2D_layer_bn(conv5_2, 'convD_1', num_filters=init_filters*16, training=training)
avg_pool = layers.averagepool2D_layer(convD_1, name='avg_pool')
logits = layers.dense_layer_bn(avg_pool, 'dense2', hidden_units=nlabels, training=training, activation=tf.identity)
return logits
def C3D_fcn_16_2D_bn(x, training, scope_name='critic', scope_reuse=False):
with tf.variable_scope(scope_name) as scope:
if scope_reuse:
scope.reuse_variables()
conv1_1 = layers.conv2D_layer_bn(x, 'conv1_1', num_filters=16, training=training)
pool1 = layers.maxpool2D_layer(conv1_1)
conv2_1 = layers.conv2D_layer_bn(pool1, 'conv2_1', num_filters=32, training=training)
pool2 = layers.maxpool2D_layer(conv2_1)
conv3_1 = layers.conv2D_layer_bn(pool2, 'conv3_1', num_filters=64, training=training)
conv3_2 = layers.conv2D_layer_bn(conv3_1, 'conv3_2', num_filters=64, training=training)
pool3 = layers.maxpool2D_layer(conv3_2)
conv4_1 = layers.conv2D_layer_bn(pool3, 'conv4_1', num_filters=128, training=training)
conv4_2 = layers.conv2D_layer_bn(conv4_1, 'conv4_2', num_filters=128,training=training)
pool4 = layers.maxpool2D_layer(conv4_2)
conv5_1 = layers.conv2D_layer_bn(pool4, 'conv5_1', num_filters=256, training=training)
conv5_2 = layers.conv2D_layer_bn(conv5_1, 'conv5_2', num_filters=256, training=training)
convD_1 = layers.conv2D_layer_bn(conv5_2, 'convD_1', num_filters=256, training=training)
convD_2 = layers.conv2D_layer(convD_1,
'convD_2',
num_filters=1,
kernel_size=(1,1,1),
activation=tf.identity)
logits = layers.averagepool2D_layer(convD_2, name='diagnosis_avg')
return logits
def unet_16_2D_bn(x, training, scope_name='generator'):
n_ch_0 = 16
with tf.variable_scope(scope_name):
conv1_1 = layers.conv2D_layer_bn(x,
'conv1_1',
num_filters=n_ch_0,
training=training)
conv1_2 = layers.conv2D_layer_bn(conv1_1,
'conv1_2',
num_filters=n_ch_0,
training=training)
pool1 = layers.maxpool2D_layer(conv1_2)
conv2_1 = layers.conv2D_layer_bn(pool1,
'conv2_1',
num_filters=n_ch_0 * 2,
training=training)
conv2_2 = layers.conv2D_layer_bn(conv2_1,
'conv2_2',
num_filters=n_ch_0 * 2,
training=training)
pool2 = layers.maxpool2D_layer(conv2_2)
conv3_1 = layers.conv2D_layer_bn(pool2,
'conv3_1',
num_filters=n_ch_0 * 4,
training=training)
conv3_2 = layers.conv2D_layer_bn(conv3_1,
'conv3_2',
num_filters=n_ch_0 * 4,
training=training)
pool3 = layers.maxpool2D_layer(conv3_2)
conv4_1 = layers.conv2D_layer_bn(pool3,
'conv4_1',
num_filters=n_ch_0 * 8,
training=training)
conv4_2 = layers.conv2D_layer_bn(conv4_1,
'conv4_2',
num_filters=n_ch_0 * 8,
training=training)
upconv3 = layers.deconv2D_layer_bn(conv4_2,
name='upconv3',
num_filters=n_ch_0,
training=training)
concat3 = layers.crop_and_concat_layer([upconv3, conv3_2], axis=-1)
conv5_1 = layers.conv2D_layer_bn(concat3,
'conv5_1',
num_filters=n_ch_0 * 4,
training=training)
conv5_2 = layers.conv2D_layer_bn(conv5_1,
'conv5_2',
num_filters=n_ch_0 * 4,
training=training)
upconv2 = layers.deconv2D_layer_bn(conv5_2,
name='upconv2',
num_filters=n_ch_0,
training=training)
concat2 = layers.crop_and_concat_layer([upconv2, conv2_2], axis=-1)
conv6_1 = layers.conv2D_layer_bn(concat2,
'conv6_1',
num_filters=n_ch_0 * 2,
training=training)
conv6_2 = layers.conv2D_layer_bn(conv6_1,
'conv6_2',
num_filters=n_ch_0 * 2,
training=training)
upconv1 = layers.deconv2D_layer_bn(conv6_2,
name='upconv1',
num_filters=n_ch_0,
training=training)
concat1 = layers.crop_and_concat_layer([upconv1, conv1_2], axis=-1)
conv8_1 = layers.conv2D_layer_bn(concat1,
'conv8_1',
num_filters=n_ch_0,
training=training)
conv8_2 = layers.conv2D_layer(conv8_1,
'conv8_2',
num_filters=1,
activation=tf.identity)
return conv8_2