Add an encoder scaffold.

PiperOrigin-RevId: 286477560

Author: tensorflower-gardener

official/nlp/modeling/layers/__init__.py

@@ -18,4 +18,5 @@
 from official.nlp.modeling.layers.masked_softmax import MaskedSoftmax
 from official.nlp.modeling.layers.on_device_embedding import OnDeviceEmbedding
 from official.nlp.modeling.layers.position_embedding import PositionEmbedding
+from official.nlp.modeling.layers.self_attention_mask import SelfAttentionMask
 from official.nlp.modeling.layers.transformer import Transformer

official/nlp/modeling/layers/self_attention_mask.py

@@ -0,0 +1,63 @@
+# 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

official/nlp/modeling/networks/albert_transformer_encoder.py

@@ -24,7 +24,6 @@
 from tensorflow.python.keras.engine import network  # pylint: disable=g-direct-tensorflow-import
 from official.modeling import activations
 from official.nlp.modeling import layers
-from official.nlp.modeling.networks import transformer_encoder
 
 
 @tf.keras.utils.register_keras_serializable(package='Text')
@@ -159,7 +158,7 @@ def __init__(self,
       embeddings = tf.cast(embeddings, tf.float16)
 
     data = embeddings
-    attention_mask = transformer_encoder.MakeAttentionMaskLayer()([data, mask])
+    attention_mask = layers.SelfAttentionMask()([data, mask])
     shared_layer = layers.Transformer(
         num_attention_heads=num_attention_heads,
         intermediate_size=intermediate_size,

official/nlp/modeling/networks/encoder_scaffold.py

@@ -0,0 +1,249 @@
+# 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.
+# ==============================================================================
+"""Transformer-based text encoder network."""
+
+from __future__ import absolute_import
+from __future__ import division
+# from __future__ import google_type_annotations
+from __future__ import print_function
+
+import inspect
+import tensorflow as tf
+
+from tensorflow.python.keras.engine import network  # pylint: disable=g-direct-tensorflow-import
+from official.nlp.modeling import layers
+
+
+@tf.keras.utils.register_keras_serializable(package='Text')
+class EncoderScaffold(network.Network):
+  """Bi-directional Transformer-based encoder network scaffold.
+
+  This network allows users to flexibly implement an encoder similar to the one
+  described in "BERT: Pre-training of Deep Bidirectional Transformers for
+  Language Understanding" (https://arxiv.org/abs/1810.04805).
+
+  In this network, users can choose to provide a custom embedding subnetwork
+  (which will replace the standard embedding logic) and/or a custom hidden layer
+  class (which will replace the Transformer instantiation in the encoder). For
+  each of these custom injection points, users can pass either a class or a
+  class instance. If a class is passed, that class will be instantiated using
+  the 'embedding_cfg' or 'hidden_cfg' argument, respectively; if an instance
+  is passed, that instance will be invoked. (In the case of hidden_cls, the
+  instance will be invoked 'num_hidden_instances' times.
+
+  If the hidden_cls is not overridden, a default transformer layer will be
+  instantiated.
+
+  Attributes:
+    num_output_classes: The output size of the classification layer.
+    classification_layer_initializer: The initializer for the classification
+      layer.
+    classification_layer_dtype: The dtype for the classification layer.
+    embedding_cls: The class or instance to use to embed the input data. This
+      class or instance defines the inputs to this encoder. If embedding_cls is
+      not set, a default embedding network (from the original BERT paper) will
+      be created.
+    embedding_cfg: A dict of kwargs to pass to the embedding_cls, if it needs to
+      be instantiated. If embedding_cls is not set, a config dict must be
+      passed to 'embedding_cfg' with the following values:
+      "vocab_size": The size of the token vocabulary.
+      "type_vocab_size": The size of the type vocabulary.
+      "hidden_size": The hidden size for this encoder.
+      "max_seq_length": The maximum sequence length for this encoder.
+      "seq_length": The sequence length for this encoder.
+      "initializer": The initializer for the embedding portion of this encoder.
+      "dropout_rate": The dropout rate to apply before the encoding layers.
+      "dtype": (Optional): The dtype of the embedding layers.
+    embedding_data: A reference to the embedding weights that will be used to
+      train the masked language model, if necessary. This is optional, and only
+      needed if (1) you are overriding embedding_cls and (2) are doing standard
+      pretraining.
+    num_hidden_instances: The number of times to instantiate and/or invoke the
+      hidden_cls.
+    hidden_cls: The class or instance to encode the input data. If hidden_cls is
+      not set, a KerasBERT transformer layer will be used as the encoder class.
+    hidden_cfg: A dict of kwargs to pass to the hidden_cls, if it needs to be
+      instantiated. If hidden_cls is not set, a config dict must be passed to
+      'hidden_cfg' with the following values:
+        "num_attention_heads": The number of attention heads. The hidden size
+          must be divisible by num_attention_heads.
+        "intermediate_size": The intermediate size of the transformer.
+        "intermediate_activation": The activation to apply in the transfomer.
+        "dropout_rate": The overall dropout rate for the transformer layers.
+        "attention_dropout_rate": The dropout rate for the attention layers.
+        "kernel_initializer": The initializer for the transformer layers.
+        "dtype": The dtype of the transformer.
+  """
+
+  def __init__(
+      self,
+      num_output_classes,
+      classification_layer_initializer=tf.keras.initializers.TruncatedNormal(
+          stddev=0.02),
+      classification_layer_dtype=tf.float32,
+      embedding_cls=None,
+      embedding_cfg=None,
+      embedding_data=None,
+      num_hidden_instances=1,
+      hidden_cls=layers.Transformer,
+      hidden_cfg=None,
+      **kwargs):
+    print(embedding_cfg)
+    self._self_setattr_tracking = False
+    self._hidden_cls = hidden_cls
+    self._hidden_cfg = hidden_cfg
+    self._num_hidden_instances = num_hidden_instances
+    self._num_output_classes = num_output_classes
+    self._classification_layer_initializer = classification_layer_initializer
+    self._embedding_cls = embedding_cls
+    self._embedding_cfg = embedding_cfg
+    self._embedding_data = embedding_data
+    self._kwargs = kwargs
+
+    if embedding_cls:
+      if inspect.isclass(embedding_cls):
+        self._embedding_network = embedding_cls(embedding_cfg)
+      else:
+        self._embedding_network = embedding_cls
+      inputs = self._embedding_network.inputs
+      embeddings, mask = self._embedding_network(inputs)
+    else:
+      self._embedding_network = None
+      word_ids = tf.keras.layers.Input(
+          shape=(embedding_cfg['seq_length'],),
+          dtype=tf.int32,
+          name='input_word_ids')
+      mask = tf.keras.layers.Input(
+          shape=(embedding_cfg['seq_length'],),
+          dtype=tf.int32,
+          name='input_mask')
+      type_ids = tf.keras.layers.Input(
+          shape=(embedding_cfg['seq_length'],),
+          dtype=tf.int32,
+          name='input_type_ids')
+      inputs = [word_ids, mask, type_ids]
+
+      self._embedding_layer = layers.OnDeviceEmbedding(
+          vocab_size=embedding_cfg['vocab_size'],
+          embedding_width=embedding_cfg['hidden_size'],
+          initializer=embedding_cfg['initializer'],
+          name='word_embeddings')
+
+      word_embeddings = self._embedding_layer(word_ids)
+
+      # Always uses dynamic slicing for simplicity.
+      self._position_embedding_layer = layers.PositionEmbedding(
+          initializer=embedding_cfg['initializer'],
+          use_dynamic_slicing=True,
+          max_sequence_length=embedding_cfg['max_seq_length'])
+      position_embeddings = self._position_embedding_layer(word_embeddings)
+
+      type_embeddings = (
+          layers.OnDeviceEmbedding(
+              vocab_size=embedding_cfg['type_vocab_size'],
+              embedding_width=embedding_cfg['hidden_size'],
+              initializer=embedding_cfg['initializer'],
+              use_one_hot=True,
+              name='type_embeddings')(type_ids))
+
+      embeddings = tf.keras.layers.Add()(
+          [word_embeddings, position_embeddings, type_embeddings])
+      embeddings = (
+          tf.keras.layers.LayerNormalization(
+              name='embeddings/layer_norm',
+              axis=-1,
+              epsilon=1e-12,
+              dtype=tf.float32)(embeddings))
+      embeddings = (
+          tf.keras.layers.Dropout(
+              rate=embedding_cfg['dropout_rate'], dtype=tf.float32)(embeddings))
+
+      if embedding_cfg.get('dtype') == 'float16':
+        embeddings = tf.cast(embeddings, tf.float16)
+
+    attention_mask = layers.SelfAttentionMask()([embeddings, mask])
+    data = embeddings
+
+    for _ in range(num_hidden_instances):
+      if inspect.isclass(hidden_cls):
+        layer = self._hidden_cls(**hidden_cfg)
+      else:
+        layer = self._hidden_cls
+      data = layer([data, attention_mask])
+
+    first_token_tensor = (
+        tf.keras.layers.Lambda(lambda x: tf.squeeze(x[:, 0:1, :], axis=1))(data)
+    )
+    cls_output = tf.keras.layers.Dense(
+        units=num_output_classes,
+        activation='tanh',
+        kernel_initializer=classification_layer_initializer,
+        dtype=classification_layer_dtype,
+        name='cls_transform')(
+            first_token_tensor)
+
+    super(EncoderScaffold, self).__init__(
+        inputs=inputs, outputs=[data, cls_output], **kwargs)
+
+  def get_config(self):
+    config_dict = {
+        'num_hidden_instances':
+            self._num_hidden_instances,
+        'num_output_classes':
+            self._num_output_classes,
+        'classification_layer_initializer':
+            self._classification_layer_initializer,
+        'embedding_cls':
+            self._embedding_network,
+        'embedding_cfg':
+            self._embedding_cfg,
+        'hidden_cfg':
+            self._hidden_cfg,
+    }
+    if inspect.isclass(self._hidden_cls):
+      config_dict['hidden_cls_string'] = tf.keras.utils.get_registered_name(
+          self._hidden_cls)
+    else:
+      config_dict['hidden_cls'] = self._hidden_cls
+
+    config_dict.update(self._kwargs)
+    return config_dict
+
+  @classmethod
+  def from_config(cls, config, custom_objects=None):
+    if 'hidden_cls_string' in config:
+      config['hidden_cls'] = tf.keras.utils.get_registered_object(
+          config['hidden_cls_string'], custom_objects=custom_objects)
+      del config['hidden_cls_string']
+    return cls(**config)
+
+  def get_embedding_table(self):
+    if self._embedding_network is None:
+      # In this case, we don't have a custom embedding network and can return
+      # the standard embedding data.
+      return self._embedding_layer.embeddings
+
+    if self._embedding_data is None:
+      raise RuntimeError(('The EncoderScaffold %s does not have a reference '
+                          'to the embedding data. This is required when you '
+                          'pass a custom embedding network to the scaffold. '
+                          'It is also possible that you are trying to get '
+                          'embedding data from an embedding scaffold with a '
+                          'custom embedding network where the scaffold has '
+                          'been serialized and deserialized. Unfortunately, '
+                          'accessing custom embedding references after '
+                          'serialization is not yet supported.') % self.name)
+    else:
+      return self._embedding_data