初窥Tensorflow Object Detection API 源码之（1.1） Resnet

大名鼎鼎的Resnet众所周知，其结构我也不想太多论述，详见论文

Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun
Deep Residual Learning for Image Recognition.
https://arxiv.org/abs/1512.03385

Overview

models/research/slim/nets/resnet_v1.py

resnet_v1

为resnet50，101，152，200提供总框架，blocks提供了网络结构配置

def resnet_v1(inputs,
              blocks,
              num_classes=None,
              is_training=True,
              global_pool=True,
              output_stride=None,
              include_root_block=True,
              spatial_squeeze=True,
              reuse=None,
              scope=None):
  with tf.variable_scope(scope, 'resnet_v1', [inputs], reuse=reuse) as sc:
    end_points_collection = sc.original_name_scope + '_end_points'
    with slim.arg_scope([slim.conv2d, bottleneck,
                         resnet_utils.stack_blocks_dense],
                        outputs_collections=end_points_collection):
      with slim.arg_scope([slim.batch_norm], is_training=is_training):
        net = inputs
        if include_root_block:
          if output_stride is not None:
            if output_stride % 4 != 0:
              raise ValueError('The output_stride needs to be a multiple of 4.')
            output_stride /= 4
          net = resnet_utils.conv2d_same(net, 64, 7, stride=2, scope='conv1')
          net = slim.max_pool2d(net, [3, 3], stride=2, scope='pool1')
        net = resnet_utils.stack_blocks_dense(net, blocks, output_stride)
        # Convert end_points_collection into a dictionary of end_points.
        end_points = slim.utils.convert_collection_to_dict(
            end_points_collection)

        if global_pool:
          # Global average pooling.
          net = tf.reduce_mean(net, [1, 2], name='pool5', keep_dims=True)
          end_points['global_pool'] = net
        if num_classes:
          net = slim.conv2d(net, num_classes, [1, 1], activation_fn=None,
                            normalizer_fn=None, scope='logits')
          end_points[sc.name + '/logits'] = net
          if spatial_squeeze:
            net = tf.squeeze(net, [1, 2], name='SpatialSqueeze')
            end_points[sc.name + '/spatial_squeeze'] = net
          end_points['predictions'] = slim.softmax(net, scope='predictions')
        return net, end_points
resnet_v1.default_image_size = 224

resnet_utils.stack_blocks_dense

resnet_utils.stack_blocks_dense会解开blocks

@slim.add_arg_scope
def stack_blocks_dense(net, blocks, output_stride=None,
                       outputs_collections=None):
  current_stride = 1

  # The atrous convolution rate parameter.
  rate = 1

  for block in blocks:
    with tf.variable_scope(block.scope, 'block', [net]) as sc:
      for i, unit in enumerate(block.args):
        if output_stride is not None and current_stride > output_stride:
          raise ValueError('The target output_stride cannot be reached.')

        with tf.variable_scope('unit_%d' % (i + 1), values=[net]):
          # If we have reached the target output_stride, then we need to employ
          # atrous convolution with stride=1 and multiply the atrous rate by the
          # current unit's stride for use in subsequent layers.
          if output_stride is not None and current_stride == output_stride:
            net = block.unit_fn(net, rate=rate, **dict(unit, stride=1))
            rate *= unit.get('stride', 1)

          else:
            net = block.unit_fn(net, rate=1, **unit)
            current_stride *= unit.get('stride', 1)
      net = slim.utils.collect_named_outputs(outputs_collections, sc.name, net)

  if output_stride is not None and current_stride != output_stride:
    raise ValueError('The target output_stride cannot be reached.')

  return net

bottleneck

Resnet的核心元结构

def bottleneck(inputs,
               depth,
               depth_bottleneck,
               stride,
               rate=1,
               outputs_collections=None,
               scope=None,
               use_bounded_activations=False):
  with tf.variable_scope(scope, 'bottleneck_v1', [inputs]) as sc:
    depth_in = slim.utils.last_dimension(inputs.get_shape(), min_rank=4)
    if depth == depth_in:
      shortcut = resnet_utils.subsample(inputs, stride, 'shortcut')
    else:
      shortcut = slim.conv2d(
          inputs,
          depth, [1, 1],
          stride=stride,
          activation_fn=tf.nn.relu6 if use_bounded_activations else None,
          scope='shortcut')

    residual = slim.conv2d(inputs, depth_bottleneck, [1, 1], stride=1,
                           scope='conv1')
    residual = resnet_utils.conv2d_same(residual, depth_bottleneck, 3, stride,
                                        rate=rate, scope='conv2')
    residual = slim.conv2d(residual, depth, [1, 1], stride=1,
                           activation_fn=None, scope='conv3')

    if use_bounded_activations:
      # Use clip_by_value to simulate bandpass activation.
      residual = tf.clip_by_value(residual, -6.0, 6.0)
      output = tf.nn.relu6(shortcut + residual)
    else:
      output = tf.nn.relu(shortcut + residual)

    return slim.utils.collect_named_outputs(outputs_collections,
                                            sc.name,
                                            output)

resnet_utils.Block

class Block(collections.namedtuple('Block', ['scope', 'unit_fn', 'args'])):

Block继承了collection.namedtuple
如此，可以解开resnet_v1_block：
{
    'scope':scope,
    'unit_fn':bottleneck,
    'args':[{
    'depth':base_depth*4,
    'depth_bottleneck':base_depth,
    'stride':stride
    }]
}

resnet_v1_50

def resnet_v1_50(inputs,
                 num_classes=None,
                 is_training=True,
                 global_pool=True,
                 output_stride=None,
                 spatial_squeeze=True,
                 reuse=None,
                 scope='resnet_v1_50'):
  blocks = [
      resnet_v1_block('block1', base_depth=64, num_units=3, stride=2),
      resnet_v1_block('block2', base_depth=128, num_units=4, stride=2),
      resnet_v1_block('block3', base_depth=256, num_units=6, stride=2),
      resnet_v1_block('block4', base_depth=512, num_units=3, stride=1),
  ]
  return resnet_v1(inputs, blocks, num_classes, is_training,
                   global_pool=global_pool, output_stride=output_stride,
                   include_root_block=True, spatial_squeeze=spatial_squeeze,
                   reuse=reuse, scope=scope)
resnet_v1_50.default_image_size = resnet_v1.default_image_size

resnet_v1_101

def resnet_v1_101(inputs,
                  num_classes=None,
                  is_training=True,
                  global_pool=True,
                  output_stride=None,
                  spatial_squeeze=True,
                  reuse=None,
                  scope='resnet_v1_101'):
  """ResNet-101 model of [1]. See resnet_v1() for arg and return description."""
  blocks = [
      resnet_v1_block('block1', base_depth=64, num_units=3, stride=2),
      resnet_v1_block('block2', base_depth=128, num_units=4, stride=2),
      resnet_v1_block('block3', base_depth=256, num_units=23, stride=2),
      resnet_v1_block('block4', base_depth=512, num_units=3, stride=1),
  ]
  return resnet_v1(inputs, blocks, num_classes, is_training,
                   global_pool=global_pool, output_stride=output_stride,
                   include_root_block=True, spatial_squeeze=spatial_squeeze,
                   reuse=reuse, scope=scope)
resnet_v1_101.default_image_size = resnet_v1.default_image_size

与resnet50的区别在于blocks的配置