tensorflow/models · error · ValueError

unpool_length is not supported by append_dense_inputs now.

Error message

unpool_length is not supported by append_dense_inputs now.

What it means

Error "unpool_length is not supported by append_dense_inputs now." thrown in tensorflow/models.

Source

Thrown at official/nlp/modeling/networks/funnel_transformer.py:494

      mask = inputs.get('input_mask')
      type_ids = inputs.get('input_type_ids')
      word_embeddings = inputs.get('input_word_embeddings', None)

      dense_inputs = inputs.get('dense_inputs', None)
      dense_mask = inputs.get('dense_mask', None)
      dense_type_ids = inputs.get('dense_type_ids', None)
    else:
      raise ValueError('Unexpected inputs type to %s.' % self.__class__)

    if word_embeddings is None:
      word_embeddings = self._embedding_layer(word_ids)

    if dense_inputs is not None:
      # Allow concatenation of the dense embeddings at sequence end if requested
      # and `unpool_length`` is set as zero
      if self._append_dense_inputs:
        if self._unpool_length != 0:
          raise ValueError(
              'unpool_length is not supported by append_dense_inputs now.'
          )
        word_embeddings = tf.concat([word_embeddings, dense_inputs], axis=1)
        type_ids = tf.concat([type_ids, dense_type_ids], axis=1)
        mask = tf.concat([mask, dense_mask], axis=1)
      else:
        # Concat the dense embeddings at sequence begin so unpool_len can
        # control embedding not being pooled.
        word_embeddings = tf.concat([dense_inputs, word_embeddings], axis=1)
        type_ids = tf.concat([dense_type_ids, type_ids], axis=1)
        mask = tf.concat([dense_mask, mask], axis=1)
    # absolute position embeddings
    position_embeddings = self._position_embedding_layer(word_embeddings)
    type_embeddings = self._type_embedding_layer(type_ids)

    embeddings = tf_keras.layers.add(
        [word_embeddings, position_embeddings, type_embeddings])
    embeddings = self._embedding_norm_layer(embeddings)

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/nlp/modeling/networks/funnel_transformer.py:494 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/28bbb482f094fca1. Report an issue: GitHub.