How to use the nnunet.utilities.tensor_utilities.sum_tensor function in nnunet

tp = net_output * y_onehot
    fp = net_output * (1 - y_onehot)
    fn = (1 - net_output) * y_onehot

    if mask is not None:
        tp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(tp, dim=1)), dim=1)
        fp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fp, dim=1)), dim=1)
        fn = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fn, dim=1)), dim=1)

    if square:
        tp = tp ** 2
        fp = fp ** 2
        fn = fn ** 2

    tp = sum_tensor(tp, axes, keepdim=False)
    fp = sum_tensor(fp, axes, keepdim=False)
    fn = sum_tensor(fn, axes, keepdim=False)

    return tp, fp, fn

y_onehot = y_onehot.cuda(net_output.device.index)
                y_onehot.scatter_(1, gt, 1)

        if self.batch_dice:
            axes = [0] + list(range(2, len(shp_x)))
        else:
            axes = list(range(2, len(shp_x)))

        if self.apply_nonlin is not None:
            softmax_output = self.apply_nonlin(net_output)
        
        # no object value
        bg_onehot = 1 - y_onehot
        squared_error = (y_onehot - softmax_output)**2
        specificity_part = sum_tensor(squared_error*y_onehot, axes)/(sum_tensor(y_onehot, axes)+self.smooth)
        sensitivity_part = sum_tensor(squared_error*bg_onehot, axes)/(sum_tensor(bg_onehot, axes)+self.smooth)

        ss = self.r * specificity_part + (1-self.r) * sensitivity_part

        if not self.do_bg:
            if self.batch_dice:
                ss = ss[1:]
            else:
                ss = ss[:, 1:]
        ss = ss.mean()

        return ss

tp = net_output * y_onehot
    fp = net_output * (1 - y_onehot)
    fn = (1 - net_output) * y_onehot

    if mask is not None:
        tp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(tp, dim=1)), dim=1)
        fp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fp, dim=1)), dim=1)
        fn = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fn, dim=1)), dim=1)

    if square:
        tp = tp ** 2
        fp = fp ** 2
        fn = fn ** 2

    tp = sum_tensor(tp, axes, keepdim=False)
    fp = sum_tensor(fp, axes, keepdim=False)
    fn = sum_tensor(fn, axes, keepdim=False)

    return tp, fp, fn

tp = net_output * y_onehot
    fp = net_output * (1 - y_onehot)
    fn = (1 - net_output) * y_onehot

    if mask is not None:
        tp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(tp, dim=1)), dim=1)
        fp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fp, dim=1)), dim=1)
        fn = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fn, dim=1)), dim=1)

    if square:
        tp = tp ** 2
        fp = fp ** 2
        fn = fn ** 2

    tp = sum_tensor(tp, axes, keepdim=False)
    fp = sum_tensor(fp, axes, keepdim=False)
    fn = sum_tensor(fn, axes, keepdim=False)

    return tp, fp, fn

tp = net_output * y_onehot
    fp = net_output * (1 - y_onehot)
    fn = (1 - net_output) * y_onehot

    if mask is not None:
        tp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(tp, dim=1)), dim=1)
        fp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fp, dim=1)), dim=1)
        fn = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fn, dim=1)), dim=1)

    if square:
        tp = tp ** 2
        fp = fp ** 2
        fn = fn ** 2

    tp = sum_tensor(tp, axes, keepdim=False)
    fp = sum_tensor(fp, axes, keepdim=False)
    fn = sum_tensor(fn, axes, keepdim=False)

    return tp, fp, fn

fp = net_output * (1 - y_onehot)
    fn = (1 - net_output) * y_onehot

    if mask is not None:
        tp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(tp, dim=1)), dim=1)
        fp = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fp, dim=1)), dim=1)
        fn = torch.stack(tuple(x_i * mask[:, 0] for x_i in torch.unbind(fn, dim=1)), dim=1)

    if square:
        tp = tp ** 2
        fp = fp ** 2
        fn = fn ** 2

    tp = sum_tensor(tp, axes, keepdim=False)
    fp = sum_tensor(fp, axes, keepdim=False)
    fn = sum_tensor(fn, axes, keepdim=False)

    return tp, fp, fn