tensorflow/models · error · ValueError

hidden_size must be a multiple of 2.

Error message

hidden_size must be a multiple of 2.

What it means

Error "hidden_size must be a multiple of 2." thrown in tensorflow/models.

Source

Thrown at official/projects/detr/modeling/detr.py:152

  def __init__(self,
               backbone,
               backbone_endpoint_name,
               num_queries,
               hidden_size,
               num_classes,
               num_encoder_layers=6,
               num_decoder_layers=6,
               dropout_rate=0.1,
               **kwargs):
    super().__init__(**kwargs)
    self._num_queries = num_queries
    self._hidden_size = hidden_size
    self._num_classes = num_classes
    self._num_encoder_layers = num_encoder_layers
    self._num_decoder_layers = num_decoder_layers
    self._dropout_rate = dropout_rate
    if hidden_size % 2 != 0:
      raise ValueError("hidden_size must be a multiple of 2.")
    self._backbone = backbone
    self._backbone_endpoint_name = backbone_endpoint_name

  def build(self, input_shape=None):
    self._input_proj = tf_keras.layers.Conv2D(
        self._hidden_size, 1, name="detr/conv2d")
    self._build_detection_decoder()
    super().build(input_shape)

  def _build_detection_decoder(self):
    """Builds detection decoder."""
    self._transformer = DETRTransformer(
        num_encoder_layers=self._num_encoder_layers,
        num_decoder_layers=self._num_decoder_layers,
        dropout_rate=self._dropout_rate)
    self._query_embeddings = self.add_weight(
        "detr/query_embeddings",
        shape=[self._num_queries, self._hidden_size],

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/detr/modeling/detr.py:152 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/0a06d677c5bf41bc. Report an issue: GitHub.