VoxelRCNN, assert xyz.shape[0] == xyz_batch_cnt.sum(),

[beyondpzk]

When I use " POINT_CLOUD_RANGE: [-40, -134.4, -2, 40, 70.4, 2]" and set batch_size=8 in KITTI, training is ok, but when I change the batch_size=12, it outputs the following errors:

pcdet/ops/pointnet2/pointnet2_stack/voxel_query_utils.py", line 77, in forward
    assert xyz.shape[0] == xyz_batch_cnt.sum(), 'xyz: %s, xyz_batch_cnt: %s' % (str(xyz.shape), str(new_xyz_batch_cnt))
AssertionError: xyz: torch.Size([1563008, 3]), xyz_batch_cnt: tensor([27648, 27648, 27648, 27648, 27648, 27648, 27648, 27648, 27648, 27648,
        27648, 27648], device='cuda:0', dtype=torch.int32)