tensorflow/models · error · ValueError

Classification heads should have unique names.

Error message

Classification heads should have unique names.

What it means

Error "Classification heads should have unique names." thrown in tensorflow/models.

Source

Thrown at official/nlp/modeling/models/bert_pretrainer.py:214

      customized_masked_lm: Optional[tf_keras.layers.Layer] = None,
      name: str = 'bert',
      **kwargs):
    super().__init__(self, name=name, **kwargs)
    self._config = {
        'encoder_network': encoder_network,
        'mlm_initializer': mlm_initializer,
        'mlm_activation': mlm_activation,
        'classification_heads': classification_heads,
        'name': name,
    }
    self.encoder_network = encoder_network
    # Makes sure the weights are built.
    _ = self.encoder_network(self.encoder_network.inputs)
    inputs = copy.copy(self.encoder_network.inputs)
    self.classification_heads = classification_heads or []
    if len(set([cls.name for cls in self.classification_heads])) != len(
        self.classification_heads):
      raise ValueError('Classification heads should have unique names.')

    self.masked_lm = customized_masked_lm or layers.MaskedLM(
        embedding_table=self.encoder_network.get_embedding_table(),
        activation=mlm_activation,
        initializer=mlm_initializer,
        name='cls/predictions')
    masked_lm_positions = tf_keras.layers.Input(
        shape=(None,), name='masked_lm_positions', dtype=tf.int32)
    if isinstance(inputs, dict):
      inputs['masked_lm_positions'] = masked_lm_positions
    else:
      inputs.append(masked_lm_positions)
    self.inputs = inputs

  def call(self, inputs):  # pytype: disable=signature-mismatch  # overriding-parameter-count-checks
    if isinstance(inputs, list):
      logging.warning('List inputs to BertPretrainer are discouraged.')
      inputs = dict([

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/nlp/modeling/models/bert_pretrainer.py:214 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tensorflow/models@e006f5f0d5 (2026-08-24). Data as JSON: /api/errors/fc94b3fcd744a860. Report an issue: GitHub.