def testCreateSquareMaxPool(self):
height, width = 3, 3
with self.test_session():
images = tf.random_uniform((5, height, width, 3), seed=1)
output = ops.max_pool(images, 3)
self.assertEquals(output.op.name, 'MaxPool/MaxPool')
self.assertListEqual(output.get_shape().as_list(), [5, 1, 1, 3])
def inception_v3(inputs,
dropout_keep_prob=0.8,
num_classes=1000,
is_training=True,
restore_logits=True,
scope=''):
"""Latest Inception from http://arxiv.org/abs/1512.00567.
"Rethinking the Inception Architecture for Computer Vision"
Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens,
Zbigniew Wojna
Args:
inputs: a tensor of size [batch_size, height, width, channels].
dropout_keep_prob: dropout keep_prob.
num_classes: number of predicted classes.
is_training: whether is training or not.
restore_logits: whether or not the logits layers should be restored.
Useful for fine-tuning a model with different num_classes.
scope: Optional scope for op_scope.
Returns:
a list containing 'logits', 'aux_logits' Tensors.
"""
# end_points will collect relevant activations for external use, for example
# summaries or losses.
end_points = {}
with tf.op_scope([inputs], scope, 'inception_v3'):
with scopes.arg_scope(
[ops.conv2d, ops.fc, ops.batch_norm, ops.dropout],
is_training=is_training):
with scopes.arg_scope([ops.conv2d, ops.max_pool, ops.avg_pool],
stride=1,
padding='VALID'):
# 299 x 299 x 3
end_points['conv0'] = ops.conv2d(inputs,
32, [3, 3],
stride=2,
scope='conv0')
# 149 x 149 x 32
end_points['conv1'] = ops.conv2d(end_points['conv0'],
32, [3, 3],
scope='conv1')
# 147 x 147 x 32
end_points['conv2'] = ops.conv2d(end_points['conv1'],
64, [3, 3],
padding='SAME',
scope='conv2')
# 147 x 147 x 64
end_points['pool1'] = ops.max_pool(end_points['conv2'], [3, 3],
stride=2,
scope='pool1')
# 73 x 73 x 64
end_points['conv3'] = ops.conv2d(end_points['pool1'],
80, [1, 1],
scope='conv3')
# 73 x 73 x 80.
end_points['conv4'] = ops.conv2d(end_points['conv3'],
192, [3, 3],
scope='conv4')
# 71 x 71 x 192.
end_points['pool2'] = ops.max_pool(end_points['conv4'], [3, 3],
stride=2,
scope='pool2')
# 35 x 35 x 192.
net = end_points['pool2']
# Inception blocks
with scopes.arg_scope([ops.conv2d, ops.max_pool, ops.avg_pool],
stride=1,
padding='SAME'):
# mixed: 35 x 35 x 256.
with tf.variable_scope('mixed_35x35x256a'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 64, [1, 1])
with tf.variable_scope('branch5x5'):
branch5x5 = ops.conv2d(net, 48, [1, 1])
branch5x5 = ops.conv2d(branch5x5, 64, [5, 5])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 64, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 32, [1, 1])
net = tf.concat(
3, [branch1x1, branch5x5, branch3x3dbl, branch_pool])
end_points['mixed_35x35x256a'] = net
# mixed_1: 35 x 35 x 288.
with tf.variable_scope('mixed_35x35x288a'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 64, [1, 1])
with tf.variable_scope('branch5x5'):
branch5x5 = ops.conv2d(net, 48, [1, 1])
branch5x5 = ops.conv2d(branch5x5, 64, [5, 5])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 64, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 64, [1, 1])
net = tf.concat(
3, [branch1x1, branch5x5, branch3x3dbl, branch_pool])
end_points['mixed_35x35x288a'] = net
# mixed_2: 35 x 35 x 288.
with tf.variable_scope('mixed_35x35x288b'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 64, [1, 1])
with tf.variable_scope('branch5x5'):
branch5x5 = ops.conv2d(net, 48, [1, 1])
branch5x5 = ops.conv2d(branch5x5, 64, [5, 5])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 64, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 64, [1, 1])
net = tf.concat(
3, [branch1x1, branch5x5, branch3x3dbl, branch_pool])
end_points['mixed_35x35x288b'] = net
# mixed_3: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768a'):
with tf.variable_scope('branch3x3'):
branch3x3 = ops.conv2d(net,
384, [3, 3],
stride=2,
padding='VALID')
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 64, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
branch3x3dbl = ops.conv2d(branch3x3dbl,
96, [3, 3],
stride=2,
padding='VALID')
with tf.variable_scope('branch_pool'):
branch_pool = ops.max_pool(net, [3, 3],
stride=2,
padding='VALID')
net = tf.concat(3, [branch3x3, branch3x3dbl, branch_pool])
end_points['mixed_17x17x768a'] = net
# mixed4: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768b'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 192, [1, 1])
with tf.variable_scope('branch7x7'):
branch7x7 = ops.conv2d(net, 128, [1, 1])
branch7x7 = ops.conv2d(branch7x7, 128, [1, 7])
branch7x7 = ops.conv2d(branch7x7, 192, [7, 1])
with tf.variable_scope('branch7x7dbl'):
branch7x7dbl = ops.conv2d(net, 128, [1, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 128, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 128, [1, 7])
branch7x7dbl = ops.conv2d(branch7x7dbl, 128, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(
3, [branch1x1, branch7x7, branch7x7dbl, branch_pool])
end_points['mixed_17x17x768b'] = net
# mixed_5: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768c'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 192, [1, 1])
with tf.variable_scope('branch7x7'):
branch7x7 = ops.conv2d(net, 160, [1, 1])
branch7x7 = ops.conv2d(branch7x7, 160, [1, 7])
branch7x7 = ops.conv2d(branch7x7, 192, [7, 1])
with tf.variable_scope('branch7x7dbl'):
branch7x7dbl = ops.conv2d(net, 160, [1, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [1, 7])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(
3, [branch1x1, branch7x7, branch7x7dbl, branch_pool])
end_points['mixed_17x17x768c'] = net
# mixed_6: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768d'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 192, [1, 1])
with tf.variable_scope('branch7x7'):
branch7x7 = ops.conv2d(net, 160, [1, 1])
branch7x7 = ops.conv2d(branch7x7, 160, [1, 7])
branch7x7 = ops.conv2d(branch7x7, 192, [7, 1])
with tf.variable_scope('branch7x7dbl'):
branch7x7dbl = ops.conv2d(net, 160, [1, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [1, 7])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(
3, [branch1x1, branch7x7, branch7x7dbl, branch_pool])
end_points['mixed_17x17x768d'] = net
# mixed_7: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768e'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 192, [1, 1])
with tf.variable_scope('branch7x7'):
branch7x7 = ops.conv2d(net, 192, [1, 1])
branch7x7 = ops.conv2d(branch7x7, 192, [1, 7])
branch7x7 = ops.conv2d(branch7x7, 192, [7, 1])
with tf.variable_scope('branch7x7dbl'):
branch7x7dbl = ops.conv2d(net, 192, [1, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(
3, [branch1x1, branch7x7, branch7x7dbl, branch_pool])
end_points['mixed_17x17x768e'] = net
# Auxiliary Head logits
aux_logits = tf.identity(end_points['mixed_17x17x768e'])
with tf.variable_scope('aux_logits'):
aux_logits = ops.avg_pool(aux_logits, [5, 5],
stride=3,
padding='VALID')
aux_logits = ops.conv2d(aux_logits,
128, [1, 1],
scope='proj')
# Shape of feature map before the final layer.
shape = aux_logits.get_shape()
aux_logits = ops.conv2d(aux_logits,
768,
shape[1:3],
stddev=0.01,
padding='VALID')
aux_logits = ops.flatten(aux_logits)
aux_logits = ops.fc(aux_logits,
num_classes,
activation=None,
stddev=0.001,
restore=restore_logits)
end_points['aux_logits'] = aux_logits
# mixed_8: 8 x 8 x 1280.
# Note that the scope below is not changed to not void previous
# checkpoints.
# (TODO) Fix the scope when appropriate.
with tf.variable_scope('mixed_17x17x1280a'):
with tf.variable_scope('branch3x3'):
branch3x3 = ops.conv2d(net, 192, [1, 1])
branch3x3 = ops.conv2d(branch3x3,
320, [3, 3],
stride=2,
padding='VALID')
with tf.variable_scope('branch7x7x3'):
branch7x7x3 = ops.conv2d(net, 192, [1, 1])
branch7x7x3 = ops.conv2d(branch7x7x3, 192, [1, 7])
branch7x7x3 = ops.conv2d(branch7x7x3, 192, [7, 1])
branch7x7x3 = ops.conv2d(branch7x7x3,
192, [3, 3],
stride=2,
padding='VALID')
with tf.variable_scope('branch_pool'):
branch_pool = ops.max_pool(net, [3, 3],
stride=2,
padding='VALID')
net = tf.concat(3, [branch3x3, branch7x7x3, branch_pool])
end_points['mixed_17x17x1280a'] = net
# mixed_9: 8 x 8 x 2048.
with tf.variable_scope('mixed_8x8x2048a'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 320, [1, 1])
with tf.variable_scope('branch3x3'):
branch3x3 = ops.conv2d(net, 384, [1, 1])
branch3x3 = tf.concat(3, [
ops.conv2d(branch3x3, 384, [1, 3]),
ops.conv2d(branch3x3, 384, [3, 1])
])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 448, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 384, [3, 3])
branch3x3dbl = tf.concat(3, [
ops.conv2d(branch3x3dbl, 384, [1, 3]),
ops.conv2d(branch3x3dbl, 384, [3, 1])
])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(
3, [branch1x1, branch3x3, branch3x3dbl, branch_pool])
end_points['mixed_8x8x2048a'] = net
# mixed_10: 8 x 8 x 2048.
with tf.variable_scope('mixed_8x8x2048b'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 320, [1, 1])
with tf.variable_scope('branch3x3'):
branch3x3 = ops.conv2d(net, 384, [1, 1])
branch3x3 = tf.concat(3, [
ops.conv2d(branch3x3, 384, [1, 3]),
ops.conv2d(branch3x3, 384, [3, 1])
])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 448, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 384, [3, 3])
branch3x3dbl = tf.concat(3, [
ops.conv2d(branch3x3dbl, 384, [1, 3]),
ops.conv2d(branch3x3dbl, 384, [3, 1])
])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(
3, [branch1x1, branch3x3, branch3x3dbl, branch_pool])
end_points['mixed_8x8x2048b'] = net
# Final pooling and prediction
with tf.variable_scope('logits'):
shape = net.get_shape()
net = ops.avg_pool(net,
shape[1:3],
padding='VALID',
scope='pool')
# 1 x 1 x 2048
net = ops.dropout(net, dropout_keep_prob, scope='dropout')
net = ops.flatten(net, scope='flatten')
# 2048
logits = ops.fc(net,
num_classes,
activation=None,
scope='logits',
restore=restore_logits)
# 1000
end_points['logits'] = logits
end_points['predictions'] = tf.nn.softmax(
logits, name='predictions')
return logits, end_points
def inference(inputs, batch_size, is_training=True):
'''
Neural network for efficient piecewise crf training
Parameters
----------
inputs: tensor
Input tensor
batch_size: int
Batch size
is_training: bool
Flag that determines whether the net is used for validation or training
Returns
-------
unary: tensor
Unary scores
pairwise_surr_map: tensor
Pairwise (surrounding neighbourhood) scores
'''
vgg_layers, vgg_layer_names = read_vgg_init(FLAGS.vgg_init_file)
conv1_sz = 64
conv2_sz = 128
conv3_sz = 256
conv4_sz = 512
conv5_sz = 512
conv6_1_sz = 1024
conv6_1_kernel = 1
conv6_sz = 512
unary_pairwise_size = 512
k = 3
bn_params = {
# Decay for the moving averages.
'decay': 0.999,
# epsilon to prevent 0s in variance.
'epsilon': 0.001,
'center': False,
'scale': False,
}
with scopes.arg_scope([slim.ops.conv2d],
stddev=0.01,
weight_decay=FLAGS.reg_factor,
is_training=is_training):
with scopes.arg_scope([ops.conv2d], is_training=is_training):
# FeatMap-Net
# VGG-16 part
net = convolve(inputs, conv1_sz, k, 'conv1_1', vgg_layers)
net = convolve(net, conv1_sz, k, 'conv1_2', vgg_layers)
net = ops.max_pool(net, [2, 2], scope='pool1')
net = convolve(net, conv2_sz, k, 'conv2_1', vgg_layers)
net = convolve(net, conv2_sz, k, 'conv2_2', vgg_layers)
net = ops.max_pool(net, [2, 2], scope='pool2')
net = convolve(net, conv3_sz, k, 'conv3_1', vgg_layers)
net = convolve(net, conv3_sz, k, 'conv3_2', vgg_layers)
net = convolve(net, conv3_sz, k, 'conv3_3', vgg_layers)
net = ops.max_pool(net, [2, 2], scope='pool3')
net = convolve(net, conv4_sz, k, 'conv4_1', vgg_layers)
net = convolve(net, conv4_sz, k, 'conv4_2', vgg_layers)
net = convolve(net, conv4_sz, k, 'conv4_3', vgg_layers)
net = ops.max_pool(net, [2, 2], scope='pool4')
net = convolve(net, conv5_sz, k, 'conv5_1', vgg_layers)
net = convolve(net, conv5_sz, k, 'conv5_2', vgg_layers)
net = convolve(net, conv5_sz, k, 'conv5_3', vgg_layers)
net = slim.ops.max_pool(net, [2, 2],
stride=1,
padding='SAME',
scope='pool5')
# Additional block
with scopes.arg_scope([ops.conv2d], batch_norm_params=bn_params):
net = convolve(net, conv6_1_sz, conv6_1_kernel, 'conv6_1')
net = convolve(net, conv6_sz, k, 'conv6_2')
featmap = convolve(net, conv6_sz, k, 'conv6_3')
unary = convolve(featmap, unary_pairwise_size, 1, 'unary_1')
# Pairwise-Net - surrounding neighbourhood
# reshape the featmap tensor
pairwise_surr = tf.transpose(featmap, perm=[1, 2, 3, 0])
pairwise_surr = tf.reshape(pairwise_surr,
shape=[-1, conv6_sz, batch_size])
first_indices = tf.gather(pairwise_surr,
indices.FIRST_INDICES_SURR)
second_indices = tf.gather(pairwise_surr,
indices.SECOND_INDICES_SURR)
pairwise_surr_map = tf.concat(1,
[first_indices, second_indices])
pairwise_surr_map = tf.reshape(
pairwise_surr_map,
shape=[
indices.NUMBER_OF_NEIGHBOURS_SURR, -1, 2 * conv6_sz,
batch_size
])
pairwise_surr_map = tf.transpose(pairwise_surr_map,
perm=[3, 0, 1, 2])
# apply convolution layers
pairwise_surr_map = convolve(pairwise_surr_map,
unary_pairwise_size, 1,
'pairwise_surr_1')
# Pairwise-Net - surrounding neighbourhood
# reshape the featmap tensor
pairwise_above_below = tf.transpose(featmap, perm=[1, 2, 3, 0])
pairwise_above_below = tf.reshape(
pairwise_above_below, shape=[-1, conv6_sz, batch_size])
first_indices_ab = tf.gather(pairwise_above_below,
indices.FIRST_INDICES_AB)
second_indices_ab = tf.gather(pairwise_above_below,
indices.SECOND_INDICES_AB)
pairwise_above_below_map = tf.concat(
1, [first_indices_ab, second_indices_ab])
pairwise_above_below_map = tf.reshape(
pairwise_above_below_map,
shape=[
indices.NUMBER_OF_NEIGHBOURS_AB, -1, 2 * conv6_sz,
batch_size
])
pairwise_above_below_map = tf.transpose(
pairwise_above_below_map, perm=[3, 0, 1, 2])
# apply convolution layers
pairwise_above_below_map = convolve(pairwise_above_below_map,
unary_pairwise_size, 1,
'pairwise_above_below_1')
# Unary-Net
unary = convolve(unary,
FLAGS.num_classes,
1,
'unary_2',
activation=None)
pairwise_surr_map = convolve(pairwise_surr_map,
FLAGS.num_classes * FLAGS.num_classes,
1,
'pairwise_surr_2',
activation=None)
# Pairwise-Net - above/below neighbourhood
# reshape the featmap tensor
pairwise_above_below_map = convolve(pairwise_above_below_map,
FLAGS.num_classes *
FLAGS.num_classes,
1,
'pairwise_above_below_2',
activation=None)
return unary, pairwise_surr_map, pairwise_above_below_map
def inference(inputs, batch_size, is_training=True):
'''
Neural network for efficient piecewise crf training
Parameters
----------
inputs: tensor
Input tensor
batch_size: int
Batch size
is_training: bool
Flag that determines whether the net is used for validation or training
Returns
-------
unary: tensor
Unary scores
pairwise_surr_map: tensor
Pairwise (surrounding neighbourhood) scores
'''
vgg_layers, vgg_layer_names = read_vgg_init(FLAGS.vgg_init_file)
conv1_sz = 64
conv2_sz = 128
conv3_sz = 256
conv4_sz = 512
conv5_sz = 512
conv6_1_sz = 1024
conv6_1_kernel = 1
conv6_sz = 512
unary_pairwise_size = 512
k = 3
bn_params = {
# Decay for the moving averages.
'decay': 0.999,
# epsilon to prevent 0s in variance.
'epsilon': 0.001,
'center': False,
'scale': False,
}
with scopes.arg_scope([slim.ops.conv2d], stddev=0.01, weight_decay=FLAGS.reg_factor, is_training=is_training):
with scopes.arg_scope([ops.conv2d], is_training=is_training):
# FeatMap-Net
# VGG-16 part
net = convolve(inputs, conv1_sz, k, 'conv1_1', vgg_layers)
net = convolve(net, conv1_sz, k, 'conv1_2', vgg_layers)
net = ops.max_pool(net, [2, 2], scope='pool1')
net = convolve(net, conv2_sz, k, 'conv2_1', vgg_layers)
net = convolve(net, conv2_sz, k, 'conv2_2', vgg_layers)
net = ops.max_pool(net, [2, 2], scope='pool2')
net = convolve(net, conv3_sz, k, 'conv3_1', vgg_layers)
net = convolve(net, conv3_sz, k, 'conv3_2', vgg_layers)
net = convolve(net, conv3_sz, k, 'conv3_3', vgg_layers)
net = ops.max_pool(net, [2, 2], scope='pool3')
net = convolve(net, conv4_sz, k, 'conv4_1', vgg_layers)
net = convolve(net, conv4_sz, k, 'conv4_2', vgg_layers)
net = convolve(net, conv4_sz, k, 'conv4_3', vgg_layers)
net = ops.max_pool(net, [2, 2], scope='pool4')
net = convolve(net, conv5_sz, k, 'conv5_1', vgg_layers)
net = convolve(net, conv5_sz, k, 'conv5_2', vgg_layers)
net = convolve(net, conv5_sz, k, 'conv5_3', vgg_layers)
net = slim.ops.max_pool(net, [2, 2], stride=1, padding='SAME', scope='pool5')
# Additional block
with scopes.arg_scope([ops.conv2d], batch_norm_params=bn_params):
net = convolve(net, conv6_1_sz, conv6_1_kernel, 'conv6_1')
net = convolve(net, conv6_sz, k, 'conv6_2')
featmap = convolve(net, conv6_sz, k, 'conv6_3')
unary = convolve(featmap, unary_pairwise_size, 1, 'unary_1')
# Pairwise-Net - surrounding neighbourhood
# reshape the featmap tensor
pairwise_surr = tf.transpose(featmap, perm=[1, 2, 3, 0])
pairwise_surr = tf.reshape(pairwise_surr, shape=[-1, conv6_sz, batch_size])
first_indices = tf.gather(pairwise_surr, indices.FIRST_INDICES_SURR)
second_indices = tf.gather(pairwise_surr, indices.SECOND_INDICES_SURR)
pairwise_surr_map = tf.concat(1, [first_indices, second_indices])
pairwise_surr_map = tf.reshape(pairwise_surr_map,
shape=[indices.NUMBER_OF_NEIGHBOURS_SURR, -1, 2 * conv6_sz, batch_size])
pairwise_surr_map = tf.transpose(pairwise_surr_map, perm=[3, 0, 1, 2])
# apply convolution layers
pairwise_surr_map = convolve(pairwise_surr_map, unary_pairwise_size, 1, 'pairwise_surr_1')
# Pairwise-Net - surrounding neighbourhood
# reshape the featmap tensor
pairwise_above_below = tf.transpose(featmap, perm=[1, 2, 3, 0])
pairwise_above_below = tf.reshape(pairwise_above_below, shape=[-1, conv6_sz, batch_size])
first_indices_ab = tf.gather(pairwise_above_below, indices.FIRST_INDICES_AB)
second_indices_ab = tf.gather(pairwise_above_below, indices.SECOND_INDICES_AB)
pairwise_above_below_map = tf.concat(1, [first_indices_ab, second_indices_ab])
pairwise_above_below_map = tf.reshape(pairwise_above_below_map,
shape=[indices.NUMBER_OF_NEIGHBOURS_AB, -1, 2 * conv6_sz,
batch_size])
pairwise_above_below_map = tf.transpose(pairwise_above_below_map, perm=[3, 0, 1, 2])
# apply convolution layers
pairwise_above_below_map = convolve(pairwise_above_below_map, unary_pairwise_size, 1,
'pairwise_above_below_1')
# Unary-Net
unary = convolve(unary, FLAGS.num_classes, 1, 'unary_2', activation=None)
pairwise_surr_map = convolve(pairwise_surr_map, FLAGS.num_classes * FLAGS.num_classes, 1,
'pairwise_surr_2', activation=None)
# Pairwise-Net - above/below neighbourhood
# reshape the featmap tensor
pairwise_above_below_map = convolve(pairwise_above_below_map, FLAGS.num_classes * FLAGS.num_classes, 1,
'pairwise_above_below_2', activation=None)
return unary, pairwise_surr_map, pairwise_above_below_map
def inception_v3(inputs,
dropout_keep_prob=0.8,
num_classes=1000,
is_training=True,
restore_logits=True,
scope=''):
"""Latest Inception from http://arxiv.org/abs/1512.00567.
"Rethinking the Inception Architecture for Computer Vision"
Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens,
Zbigniew Wojna
Args:
inputs: a tensor of size [batch_size, height, width, channels].
dropout_keep_prob: dropout keep_prob.
num_classes: number of predicted classes.
is_training: whether is training or not.
restore_logits: whether or not the logits layers should be restored.
Useful for fine-tuning a model with different num_classes.
scope: Optional scope for op_scope.
Returns:
a list containing 'logits', 'aux_logits' Tensors.
"""
# end_points will collect relevant activations for external use, for example
# summaries or losses.
end_points = {}
with tf.op_scope([inputs], scope, 'inception_v3'):
with scopes.arg_scope([ops.conv2d, ops.fc, ops.batch_norm, ops.dropout],
is_training=is_training):
with scopes.arg_scope([ops.conv2d, ops.max_pool, ops.avg_pool],
stride=1, padding='VALID'):
# 299 x 299 x 3
end_points['conv0'] = ops.conv2d(inputs, 32, [3, 3], stride=2,
scope='conv0')
# 149 x 149 x 32
end_points['conv1'] = ops.conv2d(end_points['conv0'], 32, [3, 3],
scope='conv1')
# 147 x 147 x 32
end_points['conv2'] = ops.conv2d(end_points['conv1'], 64, [3, 3],
padding='SAME', scope='conv2')
# 147 x 147 x 64
end_points['pool1'] = ops.max_pool(end_points['conv2'], [3, 3],
stride=2, scope='pool1')
# 73 x 73 x 64
end_points['conv3'] = ops.conv2d(end_points['pool1'], 80, [1, 1],
scope='conv3')
# 73 x 73 x 80.
end_points['conv4'] = ops.conv2d(end_points['conv3'], 192, [3, 3],
scope='conv4')
# 71 x 71 x 192.
end_points['pool2'] = ops.max_pool(end_points['conv4'], [3, 3],
stride=2, scope='pool2')
# 35 x 35 x 192.
net = end_points['pool2']
# Inception blocks
with scopes.arg_scope([ops.conv2d, ops.max_pool, ops.avg_pool],
stride=1, padding='SAME'):
# mixed: 35 x 35 x 256.
with tf.variable_scope('mixed_35x35x256a'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 64, [1, 1])
with tf.variable_scope('branch5x5'):
branch5x5 = ops.conv2d(net, 48, [1, 1])
branch5x5 = ops.conv2d(branch5x5, 64, [5, 5])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 64, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 32, [1, 1])
net = tf.concat(3, [branch1x1, branch5x5, branch3x3dbl, branch_pool])
end_points['mixed_35x35x256a'] = net
# mixed_1: 35 x 35 x 288.
with tf.variable_scope('mixed_35x35x288a'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 64, [1, 1])
with tf.variable_scope('branch5x5'):
branch5x5 = ops.conv2d(net, 48, [1, 1])
branch5x5 = ops.conv2d(branch5x5, 64, [5, 5])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 64, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 64, [1, 1])
net = tf.concat(3, [branch1x1, branch5x5, branch3x3dbl, branch_pool])
end_points['mixed_35x35x288a'] = net
# mixed_2: 35 x 35 x 288.
with tf.variable_scope('mixed_35x35x288b'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 64, [1, 1])
with tf.variable_scope('branch5x5'):
branch5x5 = ops.conv2d(net, 48, [1, 1])
branch5x5 = ops.conv2d(branch5x5, 64, [5, 5])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 64, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 64, [1, 1])
net = tf.concat(3, [branch1x1, branch5x5, branch3x3dbl, branch_pool])
end_points['mixed_35x35x288b'] = net
# mixed_3: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768a'):
with tf.variable_scope('branch3x3'):
branch3x3 = ops.conv2d(net, 384, [3, 3], stride=2, padding='VALID')
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 64, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3])
branch3x3dbl = ops.conv2d(branch3x3dbl, 96, [3, 3],
stride=2, padding='VALID')
with tf.variable_scope('branch_pool'):
branch_pool = ops.max_pool(net, [3, 3], stride=2, padding='VALID')
net = tf.concat(3, [branch3x3, branch3x3dbl, branch_pool])
end_points['mixed_17x17x768a'] = net
# mixed4: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768b'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 192, [1, 1])
with tf.variable_scope('branch7x7'):
branch7x7 = ops.conv2d(net, 128, [1, 1])
branch7x7 = ops.conv2d(branch7x7, 128, [1, 7])
branch7x7 = ops.conv2d(branch7x7, 192, [7, 1])
with tf.variable_scope('branch7x7dbl'):
branch7x7dbl = ops.conv2d(net, 128, [1, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 128, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 128, [1, 7])
branch7x7dbl = ops.conv2d(branch7x7dbl, 128, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(3, [branch1x1, branch7x7, branch7x7dbl, branch_pool])
end_points['mixed_17x17x768b'] = net
# mixed_5: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768c'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 192, [1, 1])
with tf.variable_scope('branch7x7'):
branch7x7 = ops.conv2d(net, 160, [1, 1])
branch7x7 = ops.conv2d(branch7x7, 160, [1, 7])
branch7x7 = ops.conv2d(branch7x7, 192, [7, 1])
with tf.variable_scope('branch7x7dbl'):
branch7x7dbl = ops.conv2d(net, 160, [1, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [1, 7])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(3, [branch1x1, branch7x7, branch7x7dbl, branch_pool])
end_points['mixed_17x17x768c'] = net
# mixed_6: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768d'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 192, [1, 1])
with tf.variable_scope('branch7x7'):
branch7x7 = ops.conv2d(net, 160, [1, 1])
branch7x7 = ops.conv2d(branch7x7, 160, [1, 7])
branch7x7 = ops.conv2d(branch7x7, 192, [7, 1])
with tf.variable_scope('branch7x7dbl'):
branch7x7dbl = ops.conv2d(net, 160, [1, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [1, 7])
branch7x7dbl = ops.conv2d(branch7x7dbl, 160, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(3, [branch1x1, branch7x7, branch7x7dbl, branch_pool])
end_points['mixed_17x17x768d'] = net
# mixed_7: 17 x 17 x 768.
with tf.variable_scope('mixed_17x17x768e'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 192, [1, 1])
with tf.variable_scope('branch7x7'):
branch7x7 = ops.conv2d(net, 192, [1, 1])
branch7x7 = ops.conv2d(branch7x7, 192, [1, 7])
branch7x7 = ops.conv2d(branch7x7, 192, [7, 1])
with tf.variable_scope('branch7x7dbl'):
branch7x7dbl = ops.conv2d(net, 192, [1, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [7, 1])
branch7x7dbl = ops.conv2d(branch7x7dbl, 192, [1, 7])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(3, [branch1x1, branch7x7, branch7x7dbl, branch_pool])
end_points['mixed_17x17x768e'] = net
# Auxiliary Head logits
aux_logits = tf.identity(end_points['mixed_17x17x768e'])
with tf.variable_scope('aux_logits'):
aux_logits = ops.avg_pool(aux_logits, [5, 5], stride=3,
padding='VALID')
aux_logits = ops.conv2d(aux_logits, 128, [1, 1], scope='proj')
# Shape of feature map before the final layer.
shape = aux_logits.get_shape()
aux_logits = ops.conv2d(aux_logits, 768, shape[1:3], stddev=0.01,
padding='VALID')
aux_logits = ops.flatten(aux_logits)
aux_logits = ops.fc(aux_logits, num_classes, activation=None,
stddev=0.001, restore=restore_logits)
end_points['aux_logits'] = aux_logits
# mixed_8: 8 x 8 x 1280.
# Note that the scope below is not changed to not void previous
# checkpoints.
# (TODO) Fix the scope when appropriate.
with tf.variable_scope('mixed_17x17x1280a'):
with tf.variable_scope('branch3x3'):
branch3x3 = ops.conv2d(net, 192, [1, 1])
branch3x3 = ops.conv2d(branch3x3, 320, [3, 3], stride=2,
padding='VALID')
with tf.variable_scope('branch7x7x3'):
branch7x7x3 = ops.conv2d(net, 192, [1, 1])
branch7x7x3 = ops.conv2d(branch7x7x3, 192, [1, 7])
branch7x7x3 = ops.conv2d(branch7x7x3, 192, [7, 1])
branch7x7x3 = ops.conv2d(branch7x7x3, 192, [3, 3],
stride=2, padding='VALID')
with tf.variable_scope('branch_pool'):
branch_pool = ops.max_pool(net, [3, 3], stride=2, padding='VALID')
net = tf.concat(3, [branch3x3, branch7x7x3, branch_pool])
end_points['mixed_17x17x1280a'] = net
# mixed_9: 8 x 8 x 2048.
with tf.variable_scope('mixed_8x8x2048a'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 320, [1, 1])
with tf.variable_scope('branch3x3'):
branch3x3 = ops.conv2d(net, 384, [1, 1])
branch3x3 = tf.concat(3, [ops.conv2d(branch3x3, 384, [1, 3]),
ops.conv2d(branch3x3, 384, [3, 1])])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 448, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 384, [3, 3])
branch3x3dbl = tf.concat(3, [ops.conv2d(branch3x3dbl, 384, [1, 3]),
ops.conv2d(branch3x3dbl, 384, [3, 1])])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(3, [branch1x1, branch3x3, branch3x3dbl, branch_pool])
end_points['mixed_8x8x2048a'] = net
# mixed_10: 8 x 8 x 2048.
with tf.variable_scope('mixed_8x8x2048b'):
with tf.variable_scope('branch1x1'):
branch1x1 = ops.conv2d(net, 320, [1, 1])
with tf.variable_scope('branch3x3'):
branch3x3 = ops.conv2d(net, 384, [1, 1])
branch3x3 = tf.concat(3, [ops.conv2d(branch3x3, 384, [1, 3]),
ops.conv2d(branch3x3, 384, [3, 1])])
with tf.variable_scope('branch3x3dbl'):
branch3x3dbl = ops.conv2d(net, 448, [1, 1])
branch3x3dbl = ops.conv2d(branch3x3dbl, 384, [3, 3])
branch3x3dbl = tf.concat(3, [ops.conv2d(branch3x3dbl, 384, [1, 3]),
ops.conv2d(branch3x3dbl, 384, [3, 1])])
with tf.variable_scope('branch_pool'):
branch_pool = ops.avg_pool(net, [3, 3])
branch_pool = ops.conv2d(branch_pool, 192, [1, 1])
net = tf.concat(3, [branch1x1, branch3x3, branch3x3dbl, branch_pool])
end_points['mixed_8x8x2048b'] = net
# Final pooling and prediction
with tf.variable_scope('logits'):
shape = net.get_shape()
net = ops.avg_pool(net, shape[1:3], padding='VALID', scope='pool')
# 1 x 1 x 2048
net = ops.dropout(net, dropout_keep_prob, scope='dropout')
net = ops.flatten(net, scope='flatten')
# 2048
logits = ops.fc(net, num_classes, activation=None, scope='logits',
restore=restore_logits)
# 1000
end_points['logits'] = logits
end_points['predictions'] = tf.nn.softmax(logits, name='predictions')
return logits, end_points
def testGlobalMaxPool(self):
height, width = 3, 3
with self.test_session():
images = tf.random_uniform((5, height, width, 3), seed=1)
output = ops.max_pool(images, images.get_shape()[1:3], stride=1)
self.assertListEqual(output.get_shape().as_list(), [5, 1, 1, 3])
def testCreateMaxPoolStrideSAME(self):
height, width = 3, 3
with self.test_session():
images = tf.random_uniform((5, height, width, 3), seed=1)
output = ops.max_pool(images, [3, 3], stride=1, padding='SAME')
self.assertListEqual(output.get_shape().as_list(), [5, height, width, 3])
def testCreateMaxPoolWithScope(self):
height, width = 3, 3
with self.test_session():
images = tf.random_uniform((5, height, width, 3), seed=1)
output = ops.max_pool(images, [3, 3], scope='pool1')
self.assertEquals(output.op.name, 'pool1/MaxPool')