def lenet5(images, nclass):
# conv 1
net = layers.conv_2d_layer(images, [5, 5, images.get_shape()[3], 20], nonlinearity=None,
padding='SAME', name='conv_1', with_biases=False)
biases = layers.variable_on_cpu('biases_1', net.get_shape()[3], tf.constant_initializer(0.0), dtype=tf.float32)
net = tf.nn.bias_add(net, biases)
net = tf.nn.relu(net)
# max pool
net = tf.nn.max_pool(net, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')
# conv 2
net = layers.conv_2d_layer(net, [5, 5, net.get_shape()[3], 50], nonlinearity=None,
padding='SAME', name='conv_2', with_biases=False)
biases = layers.variable_on_cpu('biases_2', net.get_shape()[3], tf.constant_initializer(0.0), dtype=tf.float32)
net = tf.nn.bias_add(net, biases)
net = tf.nn.relu(net)
# max pool
net = tf.nn.max_pool(net, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')
# reshape
net = tf.reshape(net, [-1, (net.get_shape()[1]*net.get_shape()[2]*net.get_shape()[3]).value])
# dense 1
net = layers.dense_layer(net, net.get_shape()[1], 500,
nonlinearity=None, name='dense_1')
net = tf.nn.relu(net)
# dense 2
net = layers.dense_layer(net, net.get_shape()[1], nclass,
nonlinearity=None, name='dense_2')
return net
def lenet5(images, nclass, wd, reuse):
# conv 1
net = layers.conv_2d_layer(images, [5, 5],
images.get_shape()[3].value,
20,
nonlinearity=tf.nn.relu,
wd=wd,
padding='SAME',
name='conv_1')
# max pool
net = tf.nn.max_pool(net,
ksize=[1, 2, 2, 1],
strides=[1, 2, 2, 1],
padding='VALID')
# conv 2
net = layers.conv_2d_layer(net, [5, 5],
net.get_shape()[3].value,
50,
nonlinearity=tf.nn.relu,
wd=wd,
padding='SAME',
name='conv_2')
# max pool
net = tf.nn.max_pool(net,
ksize=[1, 2, 2, 1],
strides=[1, 2, 2, 1],
padding='VALID')
# reshape
net = tf.reshape(net, [
-1,
(net.get_shape()[1] * net.get_shape()[2] * net.get_shape()[3]).value
])
# dense 1
net = tf.layers.dropout(net, 0.5, training=True)
net = layers.dense_layer(net,
net.get_shape()[1].value,
500,
nonlinearity=tf.nn.relu,
wd=wd,
name='dense_1')
net = tf.layers.dropout(net, 0.5, training=True)
# dense 2
net = layers.dense_layer(net,
net.get_shape()[1].value,
nclass,
nonlinearity=None,
wd=wd,
name='dense_2')
return net
def conv_bn_rectify(net, num_filters, wd, name, is_training, reuse):
with tf.variable_scope(name):
net = layers.conv_2d_layer(net, [3,3], net.get_shape()[3], num_filters, nonlinearity=None, wd=wd,
padding='SAME', name='conv', with_biases=False)
biases = layers._variable_on_cpu('biases', net.get_shape()[3], tf.constant_initializer(0.0), dtype=tf.float32)
net = tf.nn.bias_add(net, biases)
net = tf.contrib.layers.batch_norm(net, scope=tf.get_variable_scope(), decay=0.9, reuse=reuse, is_training=is_training)
net = tf.nn.relu(net)
return net
def lenet5(images, nclass, wd, reuse):
# conv 1
net = layers.conv_2d_layer(images, [5,5], images.get_shape()[3].value, 20, nonlinearity=tf.nn.relu, wd=wd,
padding='SAME', name='conv_1')
# max pool
net = tf.nn.max_pool(net, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')
# conv 2
net = layers.conv_2d_layer(net, [5,5], net.get_shape()[3].value, 50, nonlinearity=tf.nn.relu, wd=wd,
padding='SAME', name='conv_2')
# max pool
net = tf.nn.max_pool(net, ksize=[1,2,2,1], strides=[1,2,2,1], padding='VALID')
# reshape
net = tf.reshape(net, [-1, (net.get_shape()[1]*net.get_shape()[2]*net.get_shape()[3]).value])
# dense 1
net = layers.dense_layer(net, net.get_shape()[1].value, 500, nonlinearity=tf.nn.relu, wd=wd, name='dense_1')
# dense 2
net = layers.dense_layer(net, net.get_shape()[1].value, nclass, nonlinearity=None, wd=wd, name='dense_2')
return net
def conv_bn_rectify(net, num_filters, name, is_training, reuse, kl_weight=1.0):
with tf.variable_scope(name):
net = layers.conv_2d_layer(net, [3,3, net.get_shape()[3], num_filters], nonlinearity=None,
padding='SAME', name='conv', with_biases=False, reuse=reuse)
net = layers.sbp_dropout(net, 3, is_training, 'sbp', reuse, kl_weight=kl_weight)
biases = layers.variable_on_cpu('biases', net.get_shape()[3], tf.constant_initializer(0.0), dtype=tf.float32)
net = tf.nn.bias_add(net, biases)
net = tf.contrib.layers.batch_norm(net, scope=tf.get_variable_scope(), reuse=reuse, is_training=False,
center=True, scale=True)
net = tf.nn.relu(net)
return net