# Copyright 2019 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Keras layer that creates a self-attention mask."""

from __future__ import absolute_import
from __future__ import division
# from __future__ import google_type_annotations
from __future__ import print_function

import tensorflow as tf
from official.modeling import tf_utils


@tf.keras.utils.register_keras_serializable(package='Text')
class SelfAttentionMask(tf.keras.layers.Layer):
  """Create 3D attention mask from a 2D tensor mask.

    inputs[0]: from_tensor: 2D or 3D Tensor of shape
      [batch_size, from_seq_length, ...].
    inputs[1]: to_mask: int32 Tensor of shape [batch_size, to_seq_length].

    Returns:
      float Tensor of shape [batch_size, from_seq_length, to_seq_length].
  """

  def call(self, inputs):
    from_tensor = inputs[0]
    to_mask = inputs[1]
    from_shape = tf_utils.get_shape_list(from_tensor, expected_rank=[2, 3])
    batch_size = from_shape[0]
    from_seq_length = from_shape[1]

    to_shape = tf_utils.get_shape_list(to_mask, expected_rank=2)
    to_seq_length = to_shape[1]

    to_mask = tf.cast(
        tf.reshape(to_mask, [batch_size, 1, to_seq_length]),
        dtype=from_tensor.dtype)

    # We don't assume that `from_tensor` is a mask (although it could be). We
    # don't actually care if we attend *from* padding tokens (only *to* padding)
    # tokens so we create a tensor of all ones.
    #
    # `broadcast_ones` = [batch_size, from_seq_length, 1]
    broadcast_ones = tf.ones(
        shape=[batch_size, from_seq_length, 1], dtype=from_tensor.dtype)

    # Here we broadcast along two dimensions to create the mask.
    mask = broadcast_ones * to_mask

    return mask