Python common_layers.comma_separated_string_to_integer_list方法代码示例

# 需要导入模块: from tensor2tensor.layers import common_layers [as 别名]
# 或者: from tensor2tensor.layers.common_layers import comma_separated_string_to_integer_list [as 别名]
def body(self, features):
    assert self._hparams.block_size > 0
    assert not common_layers.is_xla_compiled()
    assert "targets_segmentation" not in features

    decoder_output = super(TransformerBlockParallel, self).body(features)
    assert not isinstance(decoder_output, tuple)
    assert len(decoder_output.shape) == 4

    relu_dropout_broadcast_dims = (
        common_layers.comma_separated_string_to_integer_list(
            getattr(self._hparams, "relu_dropout_broadcast_dims", "")))

    with tf.variable_scope("block_size_%d" % self._hparams.block_size):
      block_output = common_layers.dense_relu_dense(
          decoder_output,
          self._hparams.block_size * self._hparams.filter_size,
          self._hparams.block_size * self._hparams.hidden_size,
          dropout=self._hparams.relu_dropout,
          dropout_broadcast_dims=relu_dropout_broadcast_dims)

    batch_size, length = common_layers.shape_list(decoder_output)[:2]
    block_output = tf.reshape(block_output, [
        batch_size,
        length,
        self._hparams.block_size,
        self._hparams.hidden_size
    ])

    block_output = common_layers.layer_postprocess(
        decoder_output, block_output, self._hparams)

    return block_output 

# 需要导入模块: from tensor2tensor.layers import common_layers [as 别名]
# 或者: from tensor2tensor.layers.common_layers import comma_separated_string_to_integer_list [as 别名]
def recurrent_transformer_decoder(
    decoder_input,
    encoder_output,
    decoder_self_attention_bias,
    encoder_decoder_attention_bias,
    hparams,
    name="decoder",
    nonpadding=None,
    save_weights_to=None,
    make_image_summary=True):
  """Recurrent decoder function."""
  x = decoder_input
  attention_dropout_broadcast_dims = (
      common_layers.comma_separated_string_to_integer_list(
          getattr(hparams, "attention_dropout_broadcast_dims", "")))
  with tf.variable_scope(name):
    ffn_unit = functools.partial(
        # use encoder ffn, since decoder ffn use left padding
        universal_transformer_util.transformer_encoder_ffn_unit,
        hparams=hparams,
        nonpadding_mask=nonpadding)

    attention_unit = functools.partial(
        universal_transformer_util.transformer_decoder_attention_unit,
        hparams=hparams,
        encoder_output=encoder_output,
        decoder_self_attention_bias=decoder_self_attention_bias,
        encoder_decoder_attention_bias=encoder_decoder_attention_bias,
        attention_dropout_broadcast_dims=attention_dropout_broadcast_dims,
        save_weights_to=save_weights_to,
        make_image_summary=make_image_summary)

    x, extra_output = universal_transformer_util.universal_transformer_layer(
        x, hparams, ffn_unit, attention_unit)

    return common_layers.layer_preprocess(x, hparams), extra_output 

# 需要导入模块: from tensor2tensor.layers import common_layers [as 别名]
# 或者: from tensor2tensor.layers.common_layers import comma_separated_string_to_integer_list [as 别名]
def universal_transformer_decoder(decoder_input,
                                  encoder_output,
                                  decoder_self_attention_bias,
                                  encoder_decoder_attention_bias,
                                  hparams,
                                  name="decoder",
                                  nonpadding=None,
                                  save_weights_to=None,
                                  make_image_summary=True):
  """Universal Transformer decoder function.

  Prepares all the arguments and the inputs and passes it to a
  core_universal_transformer_layer to decoder.

  Args:
    decoder_input: a Tensor
    encoder_output: a Tensor
    decoder_self_attention_bias: bias Tensor for self-attention
      (see common_attention.attention_bias())
    encoder_decoder_attention_bias: bias Tensor for encoder-decoder attention
      (see common_attention.attention_bias())
    hparams: hyperparameters for model
    name: a string
    nonpadding: optional Tensor with shape [batch_size, encoder_length]
      indicating what positions are not padding.  This is used
      to mask out padding in convoltutional layers.  We generally only
      need this mask for "packed" datasets, because for ordinary datasets,
      no padding is ever followed by nonpadding.
    save_weights_to: an optional dictionary to capture attention weights
      for vizualization; the weights tensor will be appended there under
      a string key created from the variable scope (including name).
    make_image_summary: Whether to make an attention image summary.

  Returns:
    y: the output Tensors
    extra_output: which can be used to pass extra information to the body
  """
  x = decoder_input
  attention_dropout_broadcast_dims = (
      common_layers.comma_separated_string_to_integer_list(
          getattr(hparams, "attention_dropout_broadcast_dims", "")))
  with tf.variable_scope(name):
    ffn_unit = functools.partial(
        transformer_decoder_ffn_unit,
        hparams=hparams,
        nonpadding_mask=nonpadding)

    attention_unit = functools.partial(
        transformer_decoder_attention_unit,
        hparams=hparams,
        encoder_output=encoder_output,
        decoder_self_attention_bias=decoder_self_attention_bias,
        encoder_decoder_attention_bias=encoder_decoder_attention_bias,
        attention_dropout_broadcast_dims=attention_dropout_broadcast_dims,
        save_weights_to=save_weights_to,
        make_image_summary=make_image_summary)

    x, extra_output = universal_transformer_layer(
        x, hparams, ffn_unit, attention_unit)

    return common_layers.layer_preprocess(x, hparams), extra_output