Function classification_loss_1

loss.py:188–196

(prob_volume, depth_values, interval, depth_gt, mask, weight)

    else:
        return ce_loss.sum()
def classification_loss_1(prob_volume, depth_values, interval, depth_gt, mask, weight):
    depth_gt_volume = depth_gt.unsqueeze(1).expand_as(depth_values)  # (b, d, h, w)

    gt_index_volume = (
            ((depth_values - interval / 2) <= depth_gt_volume).float() * ((depth_values + interval / 2) > depth_gt_volume).float())

    pos_w = (depth_gt_volume.shape[1]-1)/1.0 # pos_w = neg_num / pos_num
    loss = binary_cross_entropy_with_logits(prob_volume, gt_index_volume, pos_weight=pos_w,mask=mask,weight=weight)
    return loss

def classification_loss(prob_volume, depth_values, interval, depth_gt, mask, weight):
    depth_gt_volume = depth_gt.unsqueeze(1).expand_as(depth_values)  # (b, d, h, w)