Generalized Focal Loss Head is inherited from AnchorHead?

[iumyx2612]

the bbox_coder in GFLHead is
bbox_coder=dict(type='DistancePointBBoxCoder')

However, anchor_generator for GFLHead is

anchor_generator=dict(
    type='AnchorGenerator',
    ratios=[1.0],
    octave_base_scale=8,
    scales_per_octave=1,
    strides=[8, 16, 32, 64, 128]),

Therefore, GFLHead needs a function to find center of Anchor Box generated by anchor_generator

def anchor_center(self, anchors):
    """Get anchor centers from anchors.

    Args:
        anchors (Tensor): Anchor list with shape (N, 4), "xyxy" format.

    Returns:
        Tensor: Anchor centers with shape (N, 2), "xy" format.
    """
    anchors_cx = (anchors[..., 2] + anchors[..., 0]) / 2
    anchors_cy = (anchors[..., 3] + anchors[..., 1]) / 2
    return torch.stack([anchors_cx, anchors_cy], dim=-1)

Why don't MMDet implementation inherit from AnchorFreeHead and use
prior_generator = MlvlPointGenerator(strides)