You are calling `simple_test_bboxes` in `dense_test_mixins`,

Error message

You are calling `simple_test_bboxes` in `dense_test_mixins`, but the `dense_test_mixins`will be deprecated soon. Please use `simple_test` instead.

What it means

Warns that the DenseTestMixins.simple_test_bboxes path is deprecated in favor of the head's unified simple_test method. The mixin still runs (it forwards feats and calls get_results) but will be removed.

Source

Thrown at mmdet/models/dense_heads/dense_test_mixins.py:44

            feats (tuple[torch.Tensor]): Multi-level features from the
                upstream network, each is a 4D-tensor.
            img_metas (list[dict]): List of image information.
            rescale (bool, optional): Whether to rescale the results.
                Defaults to False.

        Returns:
            list[obj:`InstanceData`]: Detection results of each
                image after the post process. \
                Each item usually contains following keys. \

                - scores (Tensor): Classification scores, has a shape
                  (num_instance,)
                - labels (Tensor): Labels of bboxes, has a shape
                  (num_instances,).
                - bboxes (Tensor): Has a shape (num_instances, 4),
                  the last dimension 4 arrange as (x1, y1, x2, y2).
        """
        warnings.warn('You are calling `simple_test_bboxes` in '
                      '`dense_test_mixins`, but the `dense_test_mixins`'
                      'will be deprecated soon. Please use '
                      '`simple_test` instead.')
        outs = self.forward(feats)
        results_list = self.get_results(
            *outs, img_metas=img_metas, rescale=rescale)
        return results_list

    def aug_test_bboxes(self, feats, img_metas, rescale=False):
        """Test det bboxes with test time augmentation, can be applied in
        DenseHead except for ``RPNHead`` and its variants, e.g., ``GARPNHead``,
        etc.

        Args:
            feats (list[Tensor]): the outer list indicates test-time
                augmentations and inner Tensor should have a shape NxCxHxW,
                which contains features for all images in the batch.
            img_metas (list[list[dict]]): the outer list indicates test-time

View on GitHub (pinned to cfd5d3a985)

Solutions

Switch to head.simple_test(feats, img_metas, rescale=True) which returns results_list
Refactor custom heads to override simple_test rather than mixin methods

Example fix

# before
bboxes, labels = self.bbox_head.simple_test_bboxes(feats, img_metas, ...) 
# after
results_list = self.bbox_head.simple_test(feats, img_metas, rescale=True)

Defensive patterns

Strategy: validation

Validate before calling

assert not hasattr(head, 'simple_test_bboxes') or hasattr(head, 'simple_test'), 'prefer simple_test'
results_list = head.simple_test(feats, img_metas, rescale=True)

Type guard

def has_simple_test(head):
    return callable(getattr(head, 'simple_test', None))

Prevention

Use the detector's standard test/inference entry points rather than head mixins

When it happens

Trigger: Calling head.simple_test_bboxes(...) directly on a head that inherits DenseTestMixins (e.g. older YOLO/Retina-style heads).

Common situations: Custom test scripts or inherited head code calling simple_test_bboxes instead of the detector's standard test path.

Related errors

AI-assisted analysis of open-mmlab/mmdetection@cfd5d3a985 (2026-08-27). Data as JSON: /api/errors/a8dd7278a468850b. Report an issue: GitHub.