tensorflow/models · error · RuntimeError

When input hidden_cls to EncoderScaffold %s is a list, it mu

Error message

When input hidden_cls to EncoderScaffold %s is a list, it must contain classes or instances with size specified by num_hidden_instances, got %d vs %d.

What it means

Error "When input hidden_cls to EncoderScaffold %s is a list, it must contain classes or instances with size specified by num_hidden_instances, got %d vs %d." thrown in tensorflow/models.

Source

Thrown at official/nlp/modeling/networks/encoder_scaffold.py:205

      embeddings = (
          tf_keras.layers.Dropout(
              rate=embedding_cfg['dropout_rate'])(embeddings))  # pyrefly: ignore[unsupported-operation]

      mask_cfg = {} if mask_cfg is None else mask_cfg
      if inspect.isclass(mask_cls):
        mask_layer = mask_cls(**mask_cfg)
      else:
        mask_layer = mask_cls
      attention_mask = mask_layer(embeddings, mask)

    data = embeddings

    layer_output_data = []
    hidden_layers = []
    hidden_cfg = hidden_cfg if hidden_cfg else {}

    if isinstance(hidden_cls, list) and len(hidden_cls) != num_hidden_instances:
      raise RuntimeError(
          ('When input hidden_cls to EncoderScaffold %s is a list, it must '
           'contain classes or instances with size specified by '
           'num_hidden_instances, got %d vs %d.') % self.name, len(hidden_cls),
          num_hidden_instances)
    # Consider supporting customized init states.
    recursive_states = None
    for i in range(num_hidden_instances):
      if isinstance(hidden_cls, list):
        cur_hidden_cls = hidden_cls[i]
      else:
        cur_hidden_cls = hidden_cls
      if inspect.isclass(cur_hidden_cls):
        if hidden_cfg and 'attention_cfg' in hidden_cfg and (
            layer_idx_as_attention_seed):
          hidden_cfg = copy.deepcopy(hidden_cfg)
          hidden_cfg['attention_cfg']['seed'] = i
        if feed_layer_idx:
          hidden_cfg['layer_idx'] = i

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/nlp/modeling/networks/encoder_scaffold.py:205 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/8262f79a7d22e2cb. Report an issue: GitHub.