Python tensorflow.contrib.slim.variance_scaling_initializer() Examples

The following are 30 code examples of tensorflow.contrib.slim.variance_scaling_initializer(). You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may also want to check out all available functions/classes of the module tensorflow.contrib.slim , or try the search function .
Example #1
Source File: iCAN_ResNet50_HICO.py    From iCAN with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     weight_decay=cfg.TRAIN.WEIGHT_DECAY,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': ops.GraphKeys.UPDATE_OPS
    }
    with arg_scope(
        [slim.conv2d, slim.fully_connected],
        weights_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
        weights_initializer = slim.variance_scaling_initializer(),
        biases_regularizer  = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY), 
        biases_initializer  = tf.constant_initializer(0.0),
        trainable           = is_training,
        activation_fn       = tf.nn.relu,
        normalizer_fn       = slim.batch_norm,
        normalizer_params   = batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #2
Source File: resnet_gluoncv.py    From RetinaNet_Tensorflow_Rotation with MIT License 6 votes vote down vote up 
def resnet_arg_scope(freeze_norm, is_training=True, weight_decay=0.0001,
                     batch_norm_decay=0.9, batch_norm_epsilon=1e-5, batch_norm_scale=True):

    batch_norm_params = {
        'is_training': False, 'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon, 'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': tf.GraphKeys.UPDATE_OPS,
        'data_format': DATA_FORMAT
    }
    with slim.arg_scope(
            [slim.conv2d],
            weights_regularizer=slim.l2_regularizer(weight_decay),
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params):
        with slim.arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #3
Source File: rebar.py    From Gun-Detector with Apache License 2.0 6 votes vote down vote up 
def _create_baseline(self, n_output=1, n_hidden=100,
                       is_zero_init=False,
                       collection='BASELINE'):
    # center input
    h = self._x
    if self.mean_xs is not None:
      h -= self.mean_xs

    if is_zero_init:
      initializer = init_ops.zeros_initializer()
    else:
      initializer = slim.variance_scaling_initializer()

    with slim.arg_scope([slim.fully_connected],
                        variables_collections=[collection, Q_COLLECTION],
                        trainable=False,
                        weights_initializer=initializer):
      h = slim.fully_connected(h, n_hidden, activation_fn=tf.nn.tanh)
      baseline = slim.fully_connected(h, n_output, activation_fn=None)

      if n_output == 1:
        baseline = tf.reshape(baseline, [-1])  # very important to reshape
    return baseline
Example #4
Source File: network.py    From sense_classification with Apache License 2.0 6 votes vote down vote up 
def network_arg_scope(is_training=True,
                      weight_decay=cfg.train.weight_decay,
                      batch_norm_decay=0.997,
                      batch_norm_epsilon=1e-5,
                      batch_norm_scale=True):
    batch_norm_params = {
        'is_training': is_training, 'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon, 'scale': batch_norm_scale,
        'updates_collections': ops.GraphKeys.UPDATE_OPS,
        #'variables_collections': [ tf.GraphKeys.TRAINABLE_VARIABLES ],
        'trainable': cfg.train.bn_training,
    }

    with slim.arg_scope(
            [slim.conv2d, slim.separable_convolution2d],
            weights_regularizer=slim.l2_regularizer(weight_decay),
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu6,
            #activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params,
            padding='SAME'):
        with slim.arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #5
Source File: pyramid_network.py    From Master-R-CNN with Apache License 2.0 6 votes vote down vote up 
def _extra_conv_arg_scope_with_bn(weight_decay=0.00001,
                     activation_fn=None,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):

  batch_norm_params = {
      'decay': batch_norm_decay,
      'epsilon': batch_norm_epsilon,
      'scale': batch_norm_scale,
      'updates_collections': tf.GraphKeys.UPDATE_OPS_EXTRA,
  }

  with slim.arg_scope(
      [slim.conv2d],
      weights_regularizer=slim.l2_regularizer(weight_decay),
      weights_initializer=slim.variance_scaling_initializer(),
      activation_fn=tf.nn.relu,
      normalizer_fn=slim.batch_norm,
      normalizer_params=batch_norm_params):
    with slim.arg_scope([slim.batch_norm], **batch_norm_params):
      with slim.arg_scope([slim.max_pool2d], padding='SAME') as arg_sc:
        return arg_sc
Example #6
Source File: resnet_v1.py    From tf-faster-rcnn with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
  batch_norm_params = {
    'is_training': False,
    'decay': batch_norm_decay,
    'epsilon': batch_norm_epsilon,
    'scale': batch_norm_scale,
    'trainable': False,
    'updates_collections': tf.GraphKeys.UPDATE_OPS
  }

  with arg_scope(
      [slim.conv2d],
      weights_regularizer=slim.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
      weights_initializer=slim.variance_scaling_initializer(),
      trainable=is_training,
      activation_fn=tf.nn.relu,
      normalizer_fn=slim.batch_norm,
      normalizer_params=batch_norm_params):
    with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
      return arg_sc
Example #7
Source File: cifar_model.py    From sact with Apache License 2.0 6 votes vote down vote up 
def resnet_arg_scope(is_training=True):
  """Sets up the default arguments for the CIFAR-10 resnet model."""
  batch_norm_params = {
      'is_training': is_training,
      'decay': 0.9,
      'epsilon': 0.001,
      'scale': True,
      # This forces batch_norm to compute the moving averages in-place
      # instead of using a global collection which does not work with tf.cond.
      # 'updates_collections': None,
  }

  with slim.arg_scope([slim.conv2d, slim.batch_norm], activation_fn=lrelu):
    with slim.arg_scope(
        [slim.conv2d],
        weights_regularizer=slim.l2_regularizer(0.0002),
        weights_initializer=slim.variance_scaling_initializer(),
        normalizer_fn=slim.batch_norm,
        normalizer_params=batch_norm_params):
      with slim.arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
        return arg_sc
Example #8
Source File: resnet_v1.py    From iter-reason with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
  batch_norm_params = {
    'is_training': False,
    'decay': batch_norm_decay,
    'epsilon': batch_norm_epsilon,
    'scale': batch_norm_scale,
    'trainable': False,
    'updates_collections': tf.GraphKeys.UPDATE_OPS
  }

  with slim.arg_scope(
      [slim.conv2d],
      weights_regularizer=slim.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
      weights_initializer=slim.variance_scaling_initializer(),
      trainable=is_training,
      activation_fn=tf.nn.relu,
      normalizer_fn=slim.batch_norm,
      normalizer_params=batch_norm_params):
    with slim.arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
      return arg_sc
Example #9
Source File: resnet_v1.py    From iCAN with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
  batch_norm_params = {
    'is_training': False,
    'decay': batch_norm_decay,
    'epsilon': batch_norm_epsilon,
    'scale': batch_norm_scale,
    'trainable': False,
    'updates_collections': tf.GraphKeys.UPDATE_OPS
  }

  with arg_scope(
      [slim.conv2d],
      weights_regularizer=slim.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
      weights_initializer=slim.variance_scaling_initializer(),
      trainable=is_training,
      activation_fn=tf.nn.relu,
      normalizer_fn=slim.batch_norm,
      normalizer_params=batch_norm_params):
    with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
      return arg_sc
Example #10
Source File: iCAN_ResNet50_VCOCO.py    From iCAN with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     weight_decay=cfg.TRAIN.WEIGHT_DECAY,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': ops.GraphKeys.UPDATE_OPS
    }
    with arg_scope(
        [slim.conv2d, slim.fully_connected],
        weights_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
        weights_initializer = slim.variance_scaling_initializer(),
        biases_regularizer  = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY), 
        biases_initializer  = tf.constant_initializer(0.0),
        trainable           = is_training,
        activation_fn       = tf.nn.relu,
        normalizer_fn       = slim.batch_norm,
        normalizer_params   = batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #11
Source File: iCAN_ResNet50_VCOCO_Early.py    From iCAN with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     weight_decay=cfg.TRAIN.WEIGHT_DECAY,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': ops.GraphKeys.UPDATE_OPS
    }
    with arg_scope(
        [slim.conv2d, slim.fully_connected],
        weights_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
        weights_initializer = slim.variance_scaling_initializer(),
        biases_regularizer  = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY), 
        biases_initializer  = tf.constant_initializer(0.0),
        trainable           = is_training,
        activation_fn       = tf.nn.relu,
        normalizer_fn       = slim.batch_norm,
        normalizer_params   = batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #12
Source File: faster_rcnn_wrapper.py    From anime-face-detector with MIT License 6 votes vote down vote up 
def _resnet_arg_scope():
        batch_norm_params = {
            'is_training': False,
            'decay': 0.997,
            'epsilon': 1e-5,
            'scale': True,
            'trainable': False,
            'updates_collections': tf.GraphKeys.UPDATE_OPS
        }
        with arg_scope([slim.conv2d],
                       weights_regularizer=slim.l2_regularizer(0.0001),
                       weights_initializer=slim.variance_scaling_initializer(),
                       trainable=False,
                       activation_fn=tf.nn.relu,
                       normalizer_fn=slim.batch_norm,
                       normalizer_params=batch_norm_params):
            with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
                return arg_sc
Example #13
Source File: TIN_HICO.py    From Transferable-Interactiveness-Network with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     weight_decay=cfg.TRAIN.WEIGHT_DECAY,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': ops.GraphKeys.UPDATE_OPS
    }
    with arg_scope(
        [slim.conv2d, slim.fully_connected],
        weights_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
        weights_initializer = slim.variance_scaling_initializer(),
        biases_regularizer  = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY), 
        biases_initializer  = tf.constant_initializer(0.0),
        trainable           = is_training,
        activation_fn       = tf.nn.relu,
        normalizer_fn       = slim.batch_norm,
        normalizer_params   = batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #14
Source File: TIN_VCOCO.py    From Transferable-Interactiveness-Network with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     weight_decay=cfg.TRAIN.WEIGHT_DECAY,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': ops.GraphKeys.UPDATE_OPS
    }
    with arg_scope(
        [slim.conv2d, slim.fully_connected],
        weights_regularizer = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
        weights_initializer = slim.variance_scaling_initializer(),
        biases_regularizer  = tf.contrib.layers.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY), 
        biases_initializer  = tf.constant_initializer(0.0),
        trainable           = is_training,
        activation_fn       = tf.nn.relu,
        normalizer_fn       = slim.batch_norm,
        normalizer_params   = batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #15
Source File: rebar.py    From hands-detection with MIT License 6 votes vote down vote up 
def _create_baseline(self, n_output=1, n_hidden=100,
                       is_zero_init=False,
                       collection='BASELINE'):
    # center input
    h = self._x
    if self.mean_xs is not None:
      h -= self.mean_xs

    if is_zero_init:
      initializer = init_ops.zeros_initializer()
    else:
      initializer = slim.variance_scaling_initializer()

    with slim.arg_scope([slim.fully_connected],
                        variables_collections=[collection, Q_COLLECTION],
                        trainable=False,
                        weights_initializer=initializer):
      h = slim.fully_connected(h, n_hidden, activation_fn=tf.nn.tanh)
      baseline = slim.fully_connected(h, n_output, activation_fn=None)

      if n_output == 1:
        baseline = tf.reshape(baseline, [-1])  # very important to reshape
    return baseline
Example #16
Source File: network.py    From C3AE_Age_Estimation with Apache License 2.0 6 votes vote down vote up 
def network_arg_scope(
        is_training=True, weight_decay=cfg.train.weight_decay, batch_norm_decay=0.997,
        batch_norm_epsilon=1e-5, batch_norm_scale=False):
    batch_norm_params = {
        'is_training': is_training, 'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon, 'scale': batch_norm_scale,
        'updates_collections': ops.GraphKeys.UPDATE_OPS,
        #'variables_collections': [ tf.GraphKeys.TRAINABLE_VARIABLES ],
        'trainable': cfg.train.bn_training,
    }

    with slim.arg_scope(
            [slim.conv2d, slim.separable_convolution2d],
            weights_regularizer=slim.l2_regularizer(weight_decay),
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=h_swish,
            #activation_fn=tf.nn.relu6,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params,
            padding='valid'):
        with slim.arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #17
Source File: rebar.py    From object_detection_kitti with Apache License 2.0 6 votes vote down vote up 
def _create_baseline(self, n_output=1, n_hidden=100,
                       is_zero_init=False,
                       collection='BASELINE'):
    # center input
    h = self._x
    if self.mean_xs is not None:
      h -= self.mean_xs

    if is_zero_init:
      initializer = init_ops.zeros_initializer()
    else:
      initializer = slim.variance_scaling_initializer()

    with slim.arg_scope([slim.fully_connected],
                        variables_collections=[collection, Q_COLLECTION],
                        trainable=False,
                        weights_initializer=initializer):
      h = slim.fully_connected(h, n_hidden, activation_fn=tf.nn.tanh)
      baseline = slim.fully_connected(h, n_output, activation_fn=None)

      if n_output == 1:
        baseline = tf.reshape(baseline, [-1])  # very important to reshape
    return baseline
Example #18
Source File: rebar.py    From object_detection_with_tensorflow with MIT License 6 votes vote down vote up 
def _create_baseline(self, n_output=1, n_hidden=100,
                       is_zero_init=False,
                       collection='BASELINE'):
    # center input
    h = self._x
    if self.mean_xs is not None:
      h -= self.mean_xs

    if is_zero_init:
      initializer = init_ops.zeros_initializer()
    else:
      initializer = slim.variance_scaling_initializer()

    with slim.arg_scope([slim.fully_connected],
                        variables_collections=[collection, Q_COLLECTION],
                        trainable=False,
                        weights_initializer=initializer):
      h = slim.fully_connected(h, n_hidden, activation_fn=tf.nn.tanh)
      baseline = slim.fully_connected(h, n_output, activation_fn=None)

      if n_output == 1:
        baseline = tf.reshape(baseline, [-1])  # very important to reshape
    return baseline
Example #19
Source File: rebar.py    From g-tensorflow-models with Apache License 2.0 6 votes vote down vote up 
def _create_baseline(self, n_output=1, n_hidden=100,
                       is_zero_init=False,
                       collection='BASELINE'):
    # center input
    h = self._x
    if self.mean_xs is not None:
      h -= self.mean_xs

    if is_zero_init:
      initializer = init_ops.zeros_initializer()
    else:
      initializer = slim.variance_scaling_initializer()

    with slim.arg_scope([slim.fully_connected],
                        variables_collections=[collection, Q_COLLECTION],
                        trainable=False,
                        weights_initializer=initializer):
      h = slim.fully_connected(h, n_hidden, activation_fn=tf.nn.tanh)
      baseline = slim.fully_connected(h, n_output, activation_fn=None)

      if n_output == 1:
        baseline = tf.reshape(baseline, [-1])  # very important to reshape
    return baseline
Example #20
Source File: rebar.py    From models with Apache License 2.0 6 votes vote down vote up 
def _create_baseline(self, n_output=1, n_hidden=100,
                       is_zero_init=False,
                       collection='BASELINE'):
    # center input
    h = self._x
    if self.mean_xs is not None:
      h -= self.mean_xs

    if is_zero_init:
      initializer = init_ops.zeros_initializer()
    else:
      initializer = slim.variance_scaling_initializer()

    with slim.arg_scope([slim.fully_connected],
                        variables_collections=[collection, Q_COLLECTION],
                        trainable=False,
                        weights_initializer=initializer):
      h = slim.fully_connected(h, n_hidden, activation_fn=tf.nn.tanh)
      baseline = slim.fully_connected(h, n_output, activation_fn=None)

      if n_output == 1:
        baseline = tf.reshape(baseline, [-1])  # very important to reshape
    return baseline
Example #21
Source File: rebar.py    From multilabel-image-classification-tensorflow with MIT License 6 votes vote down vote up 
def _create_baseline(self, n_output=1, n_hidden=100,
                       is_zero_init=False,
                       collection='BASELINE'):
    # center input
    h = self._x
    if self.mean_xs is not None:
      h -= self.mean_xs

    if is_zero_init:
      initializer = init_ops.zeros_initializer()
    else:
      initializer = slim.variance_scaling_initializer()

    with slim.arg_scope([slim.fully_connected],
                        variables_collections=[collection, Q_COLLECTION],
                        trainable=False,
                        weights_initializer=initializer):
      h = slim.fully_connected(h, n_hidden, activation_fn=tf.nn.tanh)
      baseline = slim.fully_connected(h, n_output, activation_fn=None)

      if n_output == 1:
        baseline = tf.reshape(baseline, [-1])  # very important to reshape
    return baseline
Example #22
Source File: pyramid_network.py    From FastMaskRCNN with Apache License 2.0 6 votes vote down vote up 
def _extra_conv_arg_scope_with_bn(weight_decay=0.00001,
                     activation_fn=None,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):

  batch_norm_params = {
      'decay': batch_norm_decay,
      'epsilon': batch_norm_epsilon,
      'scale': batch_norm_scale,
      'updates_collections': tf.GraphKeys.UPDATE_OPS,
  }

  with slim.arg_scope(
      [slim.conv2d],
      weights_regularizer=slim.l2_regularizer(weight_decay),
      weights_initializer=slim.variance_scaling_initializer(),
      activation_fn=tf.nn.relu,
      normalizer_fn=slim.batch_norm,
      normalizer_params=batch_norm_params):
    with slim.arg_scope([slim.batch_norm], **batch_norm_params):
      with slim.arg_scope([slim.max_pool2d], padding='SAME') as arg_sc:
        return arg_sc
Example #23
Source File: resnet_v1.py    From SSH-TensorFlow with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': tf.GraphKeys.UPDATE_OPS
    }

    with arg_scope(
            [slim.conv2d],
            weights_regularizer=slim.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #24
Source File: rebar.py    From yolo_v2 with Apache License 2.0 6 votes vote down vote up 
def _create_baseline(self, n_output=1, n_hidden=100,
                       is_zero_init=False,
                       collection='BASELINE'):
    # center input
    h = self._x
    if self.mean_xs is not None:
      h -= self.mean_xs

    if is_zero_init:
      initializer = init_ops.zeros_initializer()
    else:
      initializer = slim.variance_scaling_initializer()

    with slim.arg_scope([slim.fully_connected],
                        variables_collections=[collection, Q_COLLECTION],
                        trainable=False,
                        weights_initializer=initializer):
      h = slim.fully_connected(h, n_hidden, activation_fn=tf.nn.tanh)
      baseline = slim.fully_connected(h, n_output, activation_fn=None)

      if n_output == 1:
        baseline = tf.reshape(baseline, [-1])  # very important to reshape
    return baseline
Example #25
Source File: extract_pool5.py    From zero-shot-gcn with MIT License 6 votes vote down vote up 
def inception_arg_scope(is_training=True,
                        batch_norm_decay=0.997,
                        batch_norm_epsilon=1e-5,
                        batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'trainable': False,
        'updates_collections': tf.GraphKeys.UPDATE_OPS
    }
    with arg_scope(
            [slim.conv2d],
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #26
Source File: resnet_gluoncv.py    From R3Det_Tensorflow with MIT License 6 votes vote down vote up 
def resnet_arg_scope(freeze_norm, is_training=True, weight_decay=0.0001,
                     batch_norm_decay=0.9, batch_norm_epsilon=1e-5, batch_norm_scale=True):

    batch_norm_params = {
        'is_training': False, 'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon, 'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': tf.GraphKeys.UPDATE_OPS,
        'data_format': DATA_FORMAT
    }
    with slim.arg_scope(
            [slim.conv2d],
            weights_regularizer=slim.l2_regularizer(weight_decay),
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params):
        with slim.arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #27
Source File: extract_pool5.py    From zero-shot-gcn with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': tf.GraphKeys.UPDATE_OPS
    }
    with arg_scope(
            [slim.conv2d],
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #28
Source File: resnet_v1.py    From densecap-tensorflow with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': tf.GraphKeys.UPDATE_OPS
    }

    with arg_scope(
            [slim.conv2d],
            weights_regularizer=slim.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #29
Source File: ckpt_restore_test.py    From densecap-tensorflow with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': tf.GraphKeys.UPDATE_OPS
    }

    with arg_scope(
            [slim.conv2d],
            # weights_regularizer=slim.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
            weights_regularizer=None,
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc
Example #30
Source File: resnet_v1.py    From tf_ctpn with MIT License 6 votes vote down vote up 
def resnet_arg_scope(is_training=True,
                     batch_norm_decay=0.997,
                     batch_norm_epsilon=1e-5,
                     batch_norm_scale=True):
    batch_norm_params = {
        'is_training': False,
        'decay': batch_norm_decay,
        'epsilon': batch_norm_epsilon,
        'scale': batch_norm_scale,
        'trainable': False,
        'updates_collections': tf.GraphKeys.UPDATE_OPS
    }

    with arg_scope(
            [slim.conv2d],
            weights_regularizer=slim.l2_regularizer(cfg.TRAIN.WEIGHT_DECAY),
            weights_initializer=slim.variance_scaling_initializer(),
            trainable=is_training,
            activation_fn=tf.nn.relu,
            normalizer_fn=slim.batch_norm,
            normalizer_params=batch_norm_params):
        with arg_scope([slim.batch_norm], **batch_norm_params) as arg_sc:
            return arg_sc