已经弄明白了，打扰了

图中清晰代码如下：

    timer = Timer()
    avg_loss = 0.0
    iters_per_epoch = len(batch_sampler)
    best_mean_iou = -1.0
    best_model_iter = -1
    train_reader_cost = 0.0
    train_batch_cost = 0.0
    timer.start()

    iter = start_iter
    while iter < iters:
        for data in loader:
            iter += 1
            if iter > iters:
                break
            train_reader_cost += timer.elapsed_time()
            images = data[0]
            labels = data[1].astype('int64')
            edges = None
            if len(data) == 3:
                edges = data[2].astype('int64')

            if nranks > 1:
                logits_list = ddp_model(images)
            else:
                logits_list = model(images)
            loss = loss_computation(
                logits_list=logits_list,
                labels=labels,
                losses=losses,
                edges=edges)
            loss.backward()