6955fab0f4f0a38ead24cec84152d82769a0cfa2,scripts/bert/run_pretraining_hvd.py,,,#,248

Before Change

            if args.masked_lm_prob:
                raise UserWarning("argument masked_lm_prob is valid only when --raw is set")
            if args.max_predictions_per_seq:
                raise UserWarning("argument max_predictions_per_seq is valid only when "
                                  "--raw is set")
        num_parts = 1 if args.dummy_data_len else num_workers
        part_idx = 0 if args.dummy_data_len else rank
        data_train = get_dataset_fn(args.data, args.batch_size, 1, True,
                                    args.use_avg_len, args.num_buckets,

After Change

        // generate dev dataset from the raw text if needed
        if not args.eval_use_npz:
            data_eval = cache_file
            if not os.path.isfile(cache_file) and rank == 0:
                generate_dev_set(tokenizer, vocab, cache_file, args)

    logging.debug("Random seed set to %d", random_seed)
    mx.random.seed(random_seed)

    if args.data:

Instances