tensorflow/models · error · ValueError

Duplicate feats found for output level {}.

Error message

Duplicate feats found for output level {}.

What it means

Error "Duplicate feats found for output level {}." thrown in tensorflow/models.

Source

Thrown at official/vision/modeling/backbones/spinenet.py:404

      x = self._block_group(
          inputs=x,
          filters=target_num_filters,
          strides=1,
          block_fn_cand=target_block_fn,
          block_repeats=self._block_repeats,
          stochastic_depth_drop_rate=nn_layers.get_stochastic_depth_rate(
              self._init_stochastic_depth_rate, i + 1, len(self._block_specs)),
          name='scale_permuted_block_{}'.format(i + 1))

      net.append(x)
      net_sizes.append(target_width)
      net_block_fns.append(target_block_fn)
      num_outgoing_connections.append(0)

      # Save output feats.
      if block_spec.is_output:
        if block_spec.level in endpoints:
          raise ValueError('Duplicate feats found for output level {}.'.format(
              block_spec.level))
        if (block_spec.level < self._min_level or
            block_spec.level > self._max_level):
          logging.warning(
              'SpineNet output level %s out of range [min_level, max_level] = '
              '[%s, %s] will not be used for further processing.',
              block_spec.level, self._min_level, self._max_level)
        endpoints[str(block_spec.level)] = x

    return endpoints

  def _build_endpoints(self, net):
    """Matches filter size for endpoints before sharing conv layers."""
    endpoints = {}
    for level in range(self._min_level, self._max_level + 1):
      x = layers.Conv2D(
          filters=self._endpoints_num_filters,
          kernel_size=1,

View on GitHub (pinned to e006f5f0d5)

Solutions

1. Remove the duplicate feature entry for the reported output level from the SpineNet arch spec.
2. Ensure each output level maps to a single feature in the config's arch specification.

When it happens

Trigger: Thrown at official/vision/modeling/backbones/spinenet.py:404 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/befd97547d22adb6. Report an issue: GitHub.