How to use the fairseq.progress_bar.build_progress_bar function in fairseq

def train(args, trainer, task, epoch_itr):
    """Train the model for one epoch."""
    # Update parameters every N batches
    update_freq = args.update_freq[epoch_itr.epoch - 1] \
        if epoch_itr.epoch <= len(args.update_freq) else args.update_freq[-1]

    # Initialize data iterator
    itr = epoch_itr.next_epoch_itr(
        fix_batches_to_gpus=args.fix_batches_to_gpus,
        shuffle=(epoch_itr.epoch >= args.curriculum),
    )
    itr = iterators.GroupedIterator(itr, update_freq)
    progress = progress_bar.build_progress_bar(
        args, itr, epoch_itr.epoch, no_progress_bar='simple',
    )

    extra_meters = collections.defaultdict(lambda: AverageMeter())
    valid_subsets = args.valid_subset.split(',')
    max_update = args.max_update or math.inf
    for i, samples in enumerate(progress, start=epoch_itr.iterations_in_epoch):
        log_output = trainer.train_step(samples)
        if log_output is None:
            continue

        # log mid-epoch stats
        stats = get_training_stats(trainer)
        for k, v in log_output.items():
            if k in ['loss', 'nll_loss', 'ntokens', 'nsentences', 'sample_size']:
                continue  # these are already logged above

def train(args, trainer, task, epoch_itr):
    """Train the model for one epoch."""
    # Update parameters every N batches
    update_freq = args.update_freq[epoch_itr.epoch - 1] \
        if epoch_itr.epoch <= len(args.update_freq) else args.update_freq[-1]

    # Initialize data iterator
    itr = epoch_itr.next_epoch_itr(
        fix_batches_to_gpus=args.fix_batches_to_gpus,
        shuffle=(epoch_itr.epoch >= args.curriculum),
    )
    itr = iterators.GroupedIterator(itr, update_freq)
    progress = progress_bar.build_progress_bar(
        args, itr, epoch_itr.epoch, no_progress_bar='simple',
    )

    extra_meters = collections.defaultdict(lambda: AverageMeter())
    valid_subsets = args.valid_subset.split(',')
    max_update = args.max_update or math.inf
    for i, samples in enumerate(progress, start=epoch_itr.iterations_in_epoch):
        log_output = trainer.train_step(samples)
        if log_output is None:
            continue

        # log mid-epoch stats
        stats = get_training_stats(trainer)
        for k, v in log_output.items():
            if k in ['loss', 'nll_loss', 'ntokens', 'nsentences', 'sample_size']:
                continue  # these are already logged above

def train(args, trainer, task, epoch_itr):
    """Train the model for one epoch."""
    # Update parameters every N batches
    update_freq = args.update_freq[epoch_itr.epoch - 1] \
        if epoch_itr.epoch <= len(args.update_freq) else args.update_freq[-1]

    # Initialize data iterator
    itr = epoch_itr.next_epoch_itr(
        fix_batches_to_gpus=args.fix_batches_to_gpus,
        shuffle=(epoch_itr.epoch >= args.curriculum),
    )
    itr = iterators.GroupedIterator(itr, update_freq)
    progress = progress_bar.build_progress_bar(
        args, itr, epoch_itr.epoch, no_progress_bar='simple',
    )

    extra_meters = collections.defaultdict(lambda: AverageMeter())
    valid_subsets = args.valid_subset.split(',')
    max_update = args.max_update or math.inf
    for i, samples in enumerate(progress, start=epoch_itr.iterations_in_epoch):
        log_output = trainer.train_step(samples)
        if log_output is None:
            continue

        # log mid-epoch stats
        stats = get_training_stats(trainer)
        for k, v in log_output.items():
            if k in ['loss', 'nll_loss', 'ntokens', 'nsentences', 'sample_size']:
                continue  # these are already logged above

"""Evaluate the model on the validation set(s) and return the losses."""
    valid_losses = []
    for subset in subsets:
        # Initialize data iterator
        itr = data.EpochBatchIterator(
            dataset=task.dataset(subset),
            max_tokens=args.max_tokens,
            max_sentences=args.max_sentences_valid,
            max_positions=trainer.get_model().max_positions(),
            ignore_invalid_inputs=args.skip_invalid_size_inputs_valid_test,
            required_batch_size_multiple=8,
            seed=args.seed,
            num_shards=args.distributed_world_size,
            shard_id=args.distributed_rank,
        ).next_epoch_itr(shuffle=False)
        progress = progress_bar.build_progress_bar(
            args, itr, epoch_itr.epoch,
            prefix='valid on \'{}\' subset'.format(subset),
            no_progress_bar='simple'
        )

        # reset validation loss meters
        for k in ['valid_loss', 'valid_nll_loss', "acc"]:
            meter = trainer.get_meter(k)
            if meter is not None:
                meter.reset()
        extra_meters = collections.defaultdict(lambda: AverageMeter())

        for sample in progress:
            log_output = trainer.valid_step(sample)

            for k, v in log_output.items():

itr = task.get_batch_iterator(
            dataset=task.dataset(subset),
            max_tokens=args.max_tokens_valid,
            max_sentences=args.max_sentences_valid,
            max_positions=utils.resolve_max_positions(
                task.max_positions(),
                trainer.get_model().max_positions(),
            ),
            ignore_invalid_inputs=args.skip_invalid_size_inputs_valid_test,
            required_batch_size_multiple=args.required_batch_size_multiple,
            seed=args.seed,
            num_shards=args.distributed_world_size,
            shard_id=args.distributed_rank,
            num_workers=args.num_workers,
        ).next_epoch_itr(shuffle=False)
        progress = progress_bar.build_progress_bar(
            args, itr, epoch_itr.epoch,
            prefix='valid on \'{}\' subset'.format(subset),
            no_progress_bar='simple'
        )

        # reset validation loss meters
        for k in ['valid_loss', 'valid_nll_loss']:
            meter = trainer.get_meter(k)
            if meter is not None:
                meter.reset()
        extra_meters = collections.defaultdict(lambda: AverageMeter())

        for sample in progress:
            log_output = trainer.valid_step(sample)

            for k, v in log_output.items():

scorer = SequenceScorer(models, task.target_dictionary)
    if use_cuda:
        scorer.cuda()

    score_sum = 0.
    count = 0

    if args.remove_bpe is not None:
        bpe_cont = args.remove_bpe.rstrip()
        bpe_toks = set(i for i in range(len(task.dictionary)) if task.dictionary[i].endswith(bpe_cont))
        bpe_len = len(bpe_cont)
    else:
        bpe_toks = None
        bpe_len = 0

    with progress_bar.build_progress_bar(args, itr) as t:
        results = scorer.score_batched_itr(t, cuda=use_cuda, timer=gen_timer)
        wps_meter = TimeMeter()
        for _, src_tokens, __, hypos in results:
            for hypo in hypos:
                pos_scores = hypo['positional_scores']

                skipped_toks = 0
                if bpe_toks is not None:
                    for i in range(len(hypo['tokens']) - 1):
                        if hypo['tokens'][i].item() in bpe_toks:
                            skipped_toks += 1
                            pos_scores[i + 1] += pos_scores[i]
                            pos_scores[i] = 0

                inf_scores = pos_scores.eq(float('inf')) | pos_scores.eq(float('-inf'))
                if inf_scores.any():

if use_cuda:
        translator.cuda()

    # Initialize fluency scorer (and language model)
    fluency_scorer = FluencyScorer(args.lang_model_path, args.lang_model_data)

    # Generate and compute BLEU score
    scorer = bleu.Scorer(tgt_dict.pad(), tgt_dict.eos(), tgt_dict.unk())
    # Save all sources, targets and hypothesis to compute GLEU score
    sources = []
    targets = []
    hypoths = []

    num_sentences = 0
    has_target = True
    with progress_bar.build_progress_bar(args, itr) as t:
        if args.score_reference:
            translations = translator.score_batched_itr(t, cuda=use_cuda, timer=gen_timer)
        else:
            translations = translator.generate_batched_itr(
                t, maxlen_a=args.max_len_a, maxlen_b=args.max_len_b,
                cuda=use_cuda, timer=gen_timer, prefix_size=args.prefix_size,
            )

        wps_meter = TimeMeter()
        for sample_id, src_tokens, target_tokens, hypos in translations:
            # Process input and ground truth
            has_target = target_tokens is not None
            target_tokens = target_tokens.int().cpu() if has_target else None

            # Either retrieve the original sentences or regenerate them from tokens.
            if align_dict is not None:

itr = get_dataset_itr(args, task)

    # Initialize generator
    gen_timer = StopwatchMeter()
    generator = task.build_generator(args)

    num_sentences = 0

    if not os.path.exists(args.results_path):
        os.makedirs(args.results_path)

    sp = spm.SentencePieceProcessor()
    sp.Load(os.path.join(args.data, "spm.model"))

    res_files = prepare_result_files(args)
    with progress_bar.build_progress_bar(args, itr) as t:
        wps_meter = TimeMeter()
        for sample in t:
            sample = utils.move_to_cuda(sample) if use_cuda else sample
            if "net_input" not in sample:
                continue

            prefix_tokens = None
            if args.prefix_size > 0:
                prefix_tokens = sample["target"][:, : args.prefix_size]

            gen_timer.start()
            hypos = task.inference_step(generator, models, sample, prefix_tokens)
            num_generated_tokens = sum(len(h[0]["tokens"]) for h in hypos)
            gen_timer.stop(num_generated_tokens)

            for i, sample_id in enumerate(sample["id"].tolist()):

"""Evaluate the model on the validation set(s) and return the losses."""
    valid_losses = []
    for subset in subsets:
        # Initialize data iterator
        itr = data.EpochBatchIterator(
            dataset=task.dataset(subset),
            max_tokens=args.max_tokens,
            max_sentences=args.max_sentences_valid,
            max_positions=trainer.get_model().max_positions(),
            ignore_invalid_inputs=args.skip_invalid_size_inputs_valid_test,
            required_batch_size_multiple=8,
            seed=args.seed,
            num_shards=args.distributed_world_size,
            shard_id=args.distributed_rank,
        ).next_epoch_itr(shuffle=False)
        progress = progress_bar.build_progress_bar(
            args, itr, epoch_itr.epoch,
            prefix='valid on \'{}\' subset'.format(subset),
            no_progress_bar='simple'
        )

        # reset validation loss meters
        for k in ['valid_loss', 'valid_nll_loss']:
            meter = trainer.get_meter(k)
            if meter is not None:
                meter.reset()
        extra_meters = collections.defaultdict(lambda: AverageMeter())

        for sample in progress:
            log_output = trainer.valid_step(sample)

            for k, v in log_output.items():