def pretrained_vgg16_2(image, num_blocks, tower_setup, freeze_c2=True):
assert len(num_blocks) == 3
l = tf.pad(image, [[0, 0], [2, 3], [2, 3], [0, 0]])
l = Conv(name='conv0', inputs=[l], n_features=64, tower_setup=tower_setup, filter_size=(7, 7),
activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
l = Conv(name='conv1', inputs=[l], n_features=64, tower_setup=tower_setup, filter_size=(7, 7),
activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
l = tf.pad(l, [[0, 0], [0, 1], [0, 1], [0, 0]])
l = max_pool(l, shape=[2, 2], strides=[2, 2], padding='VALID')
l = Conv(name='conv2', inputs=[l], n_features=128, tower_setup=tower_setup, filter_size=(7, 7),
activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
l = Conv(name='conv3', inputs=[l], n_features=128, tower_setup=tower_setup, filter_size=(7, 7),
activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
l = tf.pad(l, [[0, 0], [0, 1], [0, 1], [0, 0]])
l = max_pool(l, shape=[2, 2], strides=[2, 2], padding='VALID')
l = Conv(name='conv4', inputs=[l], n_features=256, tower_setup=tower_setup, filter_size=(7, 7),
activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
l = Conv(name='conv5', inputs=[l], n_features=256, tower_setup=tower_setup, filter_size=(7, 7),
activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
l = Conv(name='conv6', inputs=[l], n_features=256, tower_setup=tower_setup, filter_size=(7, 7),
activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
# l = tf.pad(l, [[0, 0], [0, 1], [0, 1], [0, 0]])
# l = max_pool(l, shape=[2, 2], strides=[2, 2], padding='VALID')
# l = Conv(name='conv7', inputs=[l], n_features=512, tower_setup=tower_setup, filter_size=(7, 7),
# activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
# l = Conv(name='conv8', inputs=[l], n_features=512, tower_setup=tower_setup, filter_size=(7, 7),
# activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
# l = Conv(name='conv9', inputs=[l], n_features=512, tower_setup=tower_setup, filter_size=(7, 7),
# activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
# l = tf.pad(l, [[0, 0], [0, 1], [0, 1], [0, 0]])
# l = max_pool(l, shape=[2, 2], strides=[2, 2], padding='VALID')
# l = Conv(name='conv10', inputs=[l], n_features=512, tower_setup=tower_setup, filter_size=(7, 7),
# activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
# l = Conv(name='conv11', inputs=[l], n_features=512, tower_setup=tower_setup, filter_size=(7, 7),
# activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
# l = Conv(name='conv12', inputs=[l], n_features=512, tower_setup=tower_setup, filter_size=(7, 7),
# activation='relu', padding='SAME', batch_norm=True, old_order=True).outputs[0]
l = tf.pad(l, [[0, 0], [0, 1], [0, 1], [0, 0]])
l = max_pool(l, shape=[2, 2], strides=[2, 2], padding='VALID')
c2 = resnet_group(l, 'group0', 64, num_blocks[0], 1, tower_setup)
if freeze_c2:
c2 = tf.stop_gradient(c2)
c3 = resnet_group(c2, 'group1', 128, num_blocks[1], 2, tower_setup)
#if tower_setup.is_training:
# tf.add_to_collection('checkpoints', c3)
c4 = resnet_group(c3, 'group2', 256, num_blocks[2], 2, tower_setup)
return [c2, c3, c4]
def __init__(self, name, inputs, pool_size=(2, 2), pool_strides=None): super().__init__() assert len(inputs) == 1 pool_size = list(pool_size) inp = inputs[0] out = max_pool(inp, pool_size, pool_strides) self.outputs = [out]
def __init__(self, name, inputs, n_features, tower_setup, filter_size=(3, 3), old_order=False,
strides=(1, 1), dilation=None, pool_size=(1, 1), pool_strides=None, activation="relu", dropout=0.0,
batch_norm=False, bias=False, batch_norm_decay=Layer.BATCH_NORM_DECAY_DEFAULT, l2=Layer.L2_DEFAULT,
padding="SAME", W_initializer=None, b_initializer=None, freeze_batchnorm_override=None):
super(Conv, self).__init__()
# mind the order of dropout, conv, activation and batchnorm!
# batchnorm -> activation -> dropout -> conv -> pool
# if old_order: dropout -> conv -> batchnorm -> activation -> pool (used for example in tensorpack)
curr, n_features_inp = prepare_input(inputs)
filter_size = list(filter_size)
strides = list(strides)
pool_size = list(pool_size)
if pool_strides is None:
pool_strides = pool_size
with tf.variable_scope(name):
if isinstance(W_initializer, list):
W_initializer = tf.constant_initializer(W_initializer)
W = self.create_weight_variable("W", filter_size + [n_features_inp, n_features], l2, tower_setup,
initializer=W_initializer)
b = None
if bias:
if b_initializer is not None:
b_initializer = tf.constant_initializer(b_initializer)
b = self.create_bias_variable("b", [n_features], tower_setup, initializer=b_initializer)
else:
b = self.create_bias_variable("b", [n_features], tower_setup)
if old_order:
curr = apply_dropout(curr, dropout)
if dilation is None:
curr = conv2d(curr, W, strides, padding=padding)
else:
curr = conv2d_dilated(curr, W, dilation, padding=padding)
if bias:
curr += b
if batch_norm:
curr = self.create_and_apply_batch_norm(curr, n_features, batch_norm_decay, tower_setup, freeze_batchnorm_override=freeze_batchnorm_override)
curr = get_activation(activation)(curr)
else:
if batch_norm:
curr = self.create_and_apply_batch_norm(curr, n_features_inp, batch_norm_decay, tower_setup, freeze_batchnorm_override=freeze_batchnorm_override)
curr = get_activation(activation)(curr)
curr = apply_dropout(curr, dropout)
if dilation is None:
curr = conv2d(curr, W, strides, padding=padding)
else:
curr = conv2d_dilated(curr, W, dilation, padding=padding)
if bias:
curr += b
if pool_size != [1, 1]:
curr = max_pool(curr, pool_size, pool_strides)
self.outputs = [curr]
def pretrained_xception_done(image, num_blocks, tower_setup, freeze_c2=True):
assert len(num_blocks) == 3
l = tf.pad(image, [[0, 0], [2, 3], [2, 3], [0, 0]])
l = Conv(name='conv0',
inputs=[l],
n_features=32,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv1',
inputs=[l],
n_features=64,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv2',
inputs=[l],
n_features=128,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv3',
inputs=[l],
n_features=128,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = max_pool(l, shape=[3, 3], strides=[2, 2], padding='SAME')
l = Conv(name='conv4',
inputs=[l],
n_features=256,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv5',
inputs=[l],
n_features=256,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = max_pool(l, shape=[3, 3], strides=[2, 2], padding='SAME')
l = Conv(name='conv6',
inputs=[l],
n_features=728,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv7',
inputs=[l],
n_features=728,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = max_pool(l, shape=[3, 3], strides=[2, 2], padding='SAME')
l = Conv(name='conv8',
inputs=[l],
n_features=728,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv9',
inputs=[l],
n_features=728,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv10',
inputs=[l],
n_features=728,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv11',
inputs=[l],
n_features=728,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv12',
inputs=[l],
n_features=1024,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = max_pool(l, shape=[3, 3], strides=[2, 2], padding='SAME')
l = Conv(name='conv13',
inputs=[l],
n_features=1536,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = Conv(name='conv14',
inputs=[l],
n_features=2048,
tower_setup=tower_setup,
filter_size=(3, 3),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
l = max_pool(l, shape=[3, 3], strides=[2, 2], padding='SAME')
l = Conv(name='conv15',
inputs=[l],
n_features=2048,
tower_setup=tower_setup,
filter_size=(1, 1),
activation='relu',
padding='VALID',
batch_norm=True,
old_order=True).outputs[0]
c2 = resnet_group(l, 'group0', 64, num_blocks[0], 1, tower_setup)
if freeze_c2:
c2 = tf.stop_gradient(c2)
c3 = resnet_group(c2, 'group1', 128, num_blocks[1], 2, tower_setup)
c4 = resnet_group(c3, 'group2', 256, num_blocks[2], 2, tower_setup)
return [c2, c3, c4]