tensorflow/models · error · ValueError

At least one of encoder_input_tokens and encoder_dense_input

Error message

At least one of encoder_input_tokens and encoder_dense_inputs must be provided.

What it means

Error "At least one of encoder_input_tokens and encoder_dense_inputs must be provided." thrown in tensorflow/models.

Source

Thrown at official/nlp/modeling/models/t5.py:1433

          compute_dtype=self.compute_dtype)

  def encode(self,
             encoder_input_tokens=None,
             encoder_segment_ids=None,
             encoder_dense_inputs=None,
             encoder_dense_segment_ids=None,
             training=False):
    eligible_position_array = []
    if encoder_input_tokens is not None:
      eligible_position_array.append(
          tf.cast(tf.not_equal(encoder_input_tokens, 0), self.compute_dtype))
    if encoder_dense_inputs is not None:
      eligible_dense_positions = tf.cast(
          tf.reduce_any(tf.not_equal(encoder_dense_inputs, 0), axis=-1),
          self.compute_dtype)
      eligible_position_array.append(eligible_dense_positions)
    if not eligible_position_array:
      raise ValueError("At least one of encoder_input_tokens and"
                       " encoder_dense_inputs must be provided.")

    eligible_positions = tf.concat(eligible_position_array, axis=1)
    encoder_mask = make_attention_mask(
        eligible_positions, eligible_positions, dtype=tf.bool)

    encoder_segment_id_array = []
    if encoder_segment_ids is not None:
      encoder_segment_id_array.append(encoder_segment_ids)
    if encoder_dense_segment_ids is not None:
      encoder_segment_id_array.append(encoder_dense_segment_ids)
    if encoder_segment_id_array:
      encoder_segment_ids = tf.concat(encoder_segment_id_array, axis=1)
      segment_mask = make_attention_mask(
          encoder_segment_ids, encoder_segment_ids, tf.equal, dtype=tf.bool)
      encoder_mask = tf.math.logical_and(encoder_mask, segment_mask)
    encoder_mask = (1.0 - tf.cast(encoder_mask, self.compute_dtype)) * -1e9
    return self.encoder(

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/nlp/modeling/models/t5.py:1433 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/b91e247513f8dc61. Report an issue: GitHub.