What Does "fg_sum" Mean?

[doora1202]

Looking at the following code, there is a variable named fg_sum. What does this mean? Also, why is there a condition fg_sum > 0?

if loss_cfgs.CORNER_LOSS_REGULARIZATION and fg_sum > 0:
                # TODO: NEED to BE CHECK
                fg_rcnn_reg = rcnn_reg.view(rcnn_batch_size, -1)[fg_mask]
                fg_roi_boxes3d = roi_boxes3d.view(-1, code_size)[fg_mask]

                fg_roi_boxes3d = fg_roi_boxes3d.view(1, -1, code_size)
                batch_anchors = fg_roi_boxes3d.clone().detach()
                roi_ry = fg_roi_boxes3d[:, :, 6].view(-1)
                roi_xyz = fg_roi_boxes3d[:, :, 0:3].view(-1, 3)
                batch_anchors[:, :, 0:3] = 0
                rcnn_boxes3d = self.box_coder.decode_torch(
                    fg_rcnn_reg.view(batch_anchors.shape[0], -1, code_size), batch_anchors
                ).view(-1, code_size)

                rcnn_boxes3d = common_utils.rotate_points_along_z(
                    rcnn_boxes3d.unsqueeze(dim=1), roi_ry
                ).squeeze(dim=1)
                rcnn_boxes3d[:, 0:3] += roi_xyz

                loss_corner = loss_utils.get_corner_loss_lidar(
                    rcnn_boxes3d[:, 0:7],
                    gt_of_rois_src[fg_mask][:, 0:7]
                )
                loss_corner = loss_corner.mean()
                loss_corner = loss_corner * loss_cfgs.LOSS_WEIGHTS['rcnn_corner_weight']

                rcnn_loss_reg += loss_corner
                tb_dict['rcnn_loss_corner'] = loss_corner.item()