tensorflow/models · error · ValueError

Expected features is a rank-5 tensor. Got shape %s

Error message

Expected features is a rank-5 tensor. Got shape %s

What it means

Error "Expected features is a rank-5 tensor. Got shape %s" thrown in tensorflow/models.

Source

Thrown at official/projects/const_cl/modeling/heads/instance_reconstructor.py:173

    return tf.concat([inputs, temporal_position, spatial_position], axis=-1)

  def _keyframe_roi_pooling(self,
                            features: tf.Tensor,
                            boxes: tf.Tensor,
                            training: bool = True) -> tf.Tensor:
    """Pools ROI features on the keyframe.

    Args:
      features: a 5D tensor in shape [B, T, H, W, C].
      boxes: normalized box coordinates, a 4D tensor in shape [B, T', N, 4].
      training: whether in training mode.

    Returns:
      roi_feature: pooled ROI-features in shape [B, N, C].
    """
    if features.shape.ndims != 5:
      raise ValueError('Expected features is a rank-5 tensor. Got shape %s' %
                       features.shape)

    keyframe_index = tf.shape(boxes)[1] // 2
    t, h, w = _get_shape(features)[1:4]
    roi_features = {'0': features[:, t // 2, ...]}
    keyframe_boxes = boxes[:, keyframe_index, ...]
    unnormalized_boxes = keyframe_boxes * tf.convert_to_tensor(
        [h, w, h, w], keyframe_boxes.dtype)
    # roi_features in shape [B, N, h, w, C]
    roi_features = self._roi_aligner(
        roi_features, unnormalized_boxes, training=training)

    roi_shape = _get_shape(roi_features)
    # Perform average_pooling on ROI-pooled features.
    roi_features = tf.reshape(roi_features, [-1] + roi_shape[2:])
    roi_features = tf.reduce_mean(roi_features, axis=[1, 2])
    roi_features = tf.reshape(roi_features, roi_shape[:2] + roi_shape[-1:])
    return roi_features

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/const_cl/modeling/heads/instance_reconstructor.py:173 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/dc8984c49dfdf37a. Report an issue: GitHub.