Python v1.range方法代码示例

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def layer_norm(x, reduction_indices, epsilon=1e-9, gain=None, bias=None,
               per_element=True, scope=None):
  """DOC."""
  reduction_indices = ensure_list(reduction_indices)
  mean = tf.reduce_mean(x, reduction_indices, keep_dims=True)
  variance = tf.reduce_mean(tf.squared_difference(x, mean),
                            reduction_indices, keep_dims=True)
  normalized = (x - mean) / tf.sqrt(variance + epsilon)
  dtype = x.dtype
  shape = x.get_shape().as_list()
  for i in six.moves.range(len(shape)):
    if i not in reduction_indices or not per_element:
      shape[i] = 1
  with tf.variable_scope(scope or 'layer_norm'):
    if gain is None:
      gain = tf.get_variable('gain', shape=shape, dtype=dtype,
                             initializer=tf.ones_initializer())
    if bias is None:
      bias = tf.get_variable('bias', shape=shape, dtype=dtype,
                             initializer=tf.zeros_initializer())
  return gain*normalized+bias 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def _distributional_to_value(value_d, size, subscale, threshold):
  """Get a scalar value out of a value distribution in distributional RL."""
  half = size // 2
  value_range = (tf.to_float(tf.range(-half, half)) + 0.5) * subscale
  probs = tf.nn.softmax(value_d)

  if threshold == 0.0:
    return tf.reduce_sum(probs * value_range, axis=-1)

  # accumulated_probs[..., i] is the sum of probabilities in buckets upto i
  # so it is the probability that value <= i'th bucket value
  accumulated_probs = tf.cumsum(probs, axis=-1)
  # New probs are 0 on all lower buckets, until the threshold
  probs = tf.where(accumulated_probs < threshold, tf.zeros_like(probs), probs)
  probs /= tf.reduce_sum(probs, axis=-1, keepdims=True)  # Re-normalize.
  return tf.reduce_sum(probs * value_range, axis=-1) 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def __init__(self, *args, **kwargs):
    with tf.Graph().as_default():
      self._batch_env = SimulatedBatchEnv(*args, **kwargs)

      self._actions_t = tf.placeholder(shape=(self.batch_size,), dtype=tf.int32)
      self._rewards_t, self._dones_t = self._batch_env.simulate(self._actions_t)
      with tf.control_dependencies([self._rewards_t]):
        self._obs_t = self._batch_env.observ
      self._indices_t = tf.placeholder(shape=(self.batch_size,), dtype=tf.int32)
      self._reset_op = self._batch_env.reset(
          tf.range(self.batch_size, dtype=tf.int32)
      )

      self._sess = tf.Session()
      self._sess.run(tf.global_variables_initializer())
      self._batch_env.initialize(self._sess) 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def compute_batch_indices(batch_size, beam_size):
  """Computes the i'th coordinate that contains the batch index for gathers.

  Batch pos is a tensor like [[0,0,0,0,],[1,1,1,1],..]. It says which
  batch the beam item is in. This will create the i of the i,j coordinate
  needed for the gather.

  Args:
    batch_size: Batch size
    beam_size: Size of the beam.
  Returns:
    batch_pos: [batch_size, beam_size] tensor of ids
  """
  batch_pos = tf.range(batch_size * beam_size) // beam_size
  batch_pos = tf.reshape(batch_pos, [batch_size, beam_size])
  return batch_pos 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def pad_batch(features, batch_multiple):
  """Pad batch dim of features to nearest multiple of batch_multiple."""
  feature = list(features.items())[0][1]
  batch_size = tf.shape(feature)[0]
  mod = batch_size % batch_multiple
  has_mod = tf.cast(tf.cast(mod, tf.bool), tf.int32)
  batch_padding = batch_multiple * has_mod - mod

  padded_features = {}
  for k, feature in features.items():
    rank = len(feature.shape)
    paddings = [[0, 0] for _ in range(rank)]
    paddings[0][1] = batch_padding
    padded_feature = tf.pad(feature, paddings)
    padded_features[k] = padded_feature
  return padded_features


# TODO(lukaszkaiser): refactor the API to not be just a list of self params
#   but make sense for other uses too. 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def __init__(self, *args, **kwargs):
    super(TransformerMemory, self).__init__(*args, **kwargs)

    hparams = self._hparams
    self.recurrent_memory_by_layer = {}
    for layer in range(hparams.num_decoder_layers or hparams.num_hidden_layers):
      layer_name = "layer_%d" % layer
      if hparams.memory_type == "neural_memory":
        memory = transformer_memory.TransformerMemory(
            batch_size=int(hparams.batch_size / hparams.max_length),
            key_depth=hparams.hidden_size,
            val_depth=hparams.hidden_size,
            memory_size=hparams.split_targets_chunk_length,
            sharpen_factor=1.,
            name=layer_name + "/recurrent_memory")
      elif hparams.memory_type == "transformer_xl":
        memory = transformer_memory.RecentTokensMemory(
            layer_name + "/recurrent_memory", hparams)
      else:
        raise ValueError("Unsupported memory type: %s" % hparams.memory_type)
      self.recurrent_memory_by_layer[layer_name] = memory 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def compress(x, c, is_2d, hparams, name):
  """Compress."""
  with tf.variable_scope(name):
    # Run compression by strided convs.
    cur = x
    k1 = (3, 3) if is_2d else (3, 1)
    k2 = (2, 2) if is_2d else (2, 1)
    cur = residual_conv(cur, hparams.num_compress_steps, k1, hparams, "rc")
    if c is not None and hparams.do_attend_compress:
      cur = attend(cur, c, hparams, "compress_attend")
    for i in range(hparams.num_compress_steps):
      if hparams.do_residual_compress:
        cur = residual_conv(cur, hparams.num_compress_steps, k1, hparams,
                            "rc_%d" % i)
      cur = common_layers.conv_block(
          cur, hparams.hidden_size, [((1, 1), k2)],
          strides=k2, name="compress_%d" % i)
    return cur 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def shuffle_layer(inputs, shuffle_fn=rol):
  """Shuffles the elements according to bitwise left or right rotation.

  Args:
    inputs: Tensor input from previous layer
    shuffle_fn: Shift function rol or ror

  Returns:
    tf.Tensor: Inputs shifted according to shuffle_fn
  """

  length = tf.shape(inputs)[1]
  n_bits = tf.log(tf.cast(length - 1, tf.float32)) / tf.log(2.0)
  n_bits = tf.cast(n_bits, tf.int32) + 1

  indices = tf.range(0, length)
  rev_indices = shuffle_fn(indices, n_bits)
  return tf.gather(inputs, rev_indices, axis=1) 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def residual_shuffle_network(inputs, hparams):
  """Residual Shuffle-Exchange network with weight sharing.

  Args:
    inputs: inputs to the Shuffle-Exchange network. Should be in length of power
      of 2.
    hparams: Model configuration

  Returns:
    tf.Tensor: Outputs of the Shuffle-Exchange last layer
  """
  input_shape = tf.shape(inputs)
  n_bits = tf.log(tf.cast(input_shape[1] - 1, tf.float32)) / tf.log(2.0)
  n_bits = tf.cast(n_bits, tf.int32) + 1

  block_out = inputs

  for k in range(hparams.num_hidden_layers):
    with tf.variable_scope("benes_block_" + str(k), reuse=tf.AUTO_REUSE):
      forward_output = forward_part(block_out, hparams, n_bits)
      block_out = reverse_part(forward_output, hparams, n_bits)

  return RSU("last_layer", hparams.dropout, hparams.mode)(block_out) 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def get_timing_signal(length,
                      min_timescale=1,
                      max_timescale=1e4,
                      num_timescales=16):
  """Create Tensor of sinusoids of different frequencies.

  Args:
    length: Length of the Tensor to create, i.e. Number of steps.
    min_timescale: a float
    max_timescale: a float
    num_timescales: an int

  Returns:
    Tensor of shape (length, 2*num_timescales)
  """
  positions = to_float(tf.range(length))
  log_timescale_increment = (
      math.log(max_timescale / min_timescale) / (num_timescales - 1))
  inv_timescales = min_timescale * tf.exp(
      to_float(tf.range(num_timescales)) * -log_timescale_increment)
  scaled_time = tf.expand_dims(positions, 1) * tf.expand_dims(inv_timescales, 0)
  return tf.concat([tf.sin(scaled_time), tf.cos(scaled_time)], axis=1) 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def smoothing_cross_entropy_factored(a, b, labels, confidence):
  """Memory-efficient computation of smoothing cross-entropy.

  Avoids realizing the entire logits matrix at once.

  Args:
    a: a Tensor with shape [batch, inner_dim]
    b: a Tensor with shape [vocab_size, inner_dim]
    labels: an integer Tensor with shape [batch]
    confidence: a float

  Returns:
    A Tensor with shape [batch]
  """
  num_splits = 16
  vocab_size = shape_list(b)[0]
  labels = approximate_split(labels, num_splits)
  a = approximate_split(a, num_splits)
  parts = []
  for part in range(num_splits):
    with tf.control_dependencies(parts[-1:]):
      logits = tf.matmul(a[part], b, transpose_b=True)
      parts.append(
          smoothing_cross_entropy(logits, labels[part], vocab_size, confidence))
  return tf.concat(parts, 0) 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def argmax_with_score(logits, axis=None):
  """Argmax along with the value."""
  axis = axis or len(logits.get_shape()) - 1
  predictions = tf.argmax(logits, axis=axis)

  logits_shape = shape_list(logits)
  prefix_shape, vocab_size = logits_shape[:-1], logits_shape[-1]
  prefix_size = 1
  for d in prefix_shape:
    prefix_size *= d

  # Flatten to extract scores
  flat_logits = tf.reshape(logits, [prefix_size, vocab_size])
  flat_predictions = tf.reshape(predictions, [prefix_size])
  flat_indices = tf.stack(
      [tf.range(tf.to_int64(prefix_size)),
       tf.to_int64(flat_predictions)],
      axis=1)
  flat_scores = tf.gather_nd(flat_logits, flat_indices)

  # Unflatten
  scores = tf.reshape(flat_scores, prefix_shape)

  return predictions, scores 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def summarize_video(video, prefix, max_outputs=1):
  """Summarize the video using image summaries starting with prefix."""
  video_shape = shape_list(video)
  if len(video_shape) != 5:
    raise ValueError("Assuming videos given as tensors in the format "
                     "[batch, time, height, width, channels] but got one "
                     "of shape: %s" % str(video_shape))
  if tf.executing_eagerly():
    return
  if video.get_shape().as_list()[1] is None:
    tf.summary.image(
        "%s_last_frame" % prefix,
        tf.cast(video[:, -1, :, :, :], tf.uint8),
        max_outputs=max_outputs)
  else:
    for k in range(video_shape[1]):
      tf.summary.image(
          "%s_frame_%d" % (prefix, k),
          tf.cast(video[:, k, :, :, :], tf.uint8),
          max_outputs=max_outputs) 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def patch_discriminator(x, filters=64, filter_size=5, n=4,
                        name="patch_discrim"):
  """Patch descriminator."""
  with tf.variable_scope(name):
    x_shape = shape_list(x)
    spatial_dims = [x_shape[1] // 4, x_shape[2] // 4]
    x = tf.random_crop(x, [x_shape[0]] + spatial_dims + [x_shape[3]])
    for i in range(n):
      x = general_conv(
          x=x,
          num_filters=filters * 2**i,
          filter_size=filter_size,
          stride=2 if i != n - 1 else 1,
          stddev=0.02,
          padding="SAME",
          name="c%d" % i,
          do_norm="instance" if i != 0 else False,
          do_relu=i != n - 1,
          relufactor=0.2)
    x = tf.reduce_mean(x, [1, 2])
    return x 

# 需要导入模块: from tensorflow.compat import v1 [as 别名]
# 或者: from tensorflow.compat.v1 import range [as 别名]
def testExtractblocks(self):

    batch_size = 1
    num_heads = 3
    height = 6
    width = 10
    depth = 15
    block_h = 3
    block_w = 2
    t = np.random.rand(batch_size * num_heads, height, width, depth)
    a = common_attention._extract_blocks(t, block_h, block_w)
    self.evaluate(tf.global_variables_initializer())
    res = self.evaluate(a)
    self.assertEqual(res.shape, (batch_size * num_heads, height//block_h,
                                 width//block_w, block_h, block_w, depth))
    # also check if the content is right
    out = np.zeros((batch_size*num_heads, height//block_h,
                    width//block_w, block_h, block_w, depth))
    for b in range(batch_size*num_heads):
      for x in range(height//block_h):
        for y in range(width//block_w):
          for v in range(block_h):
            for w in range(block_w):
              out[b, x, y, v, w] = t[b, block_h*x+v, block_w*y+w]
    self.assertAllClose(res, out)