Function train

tools/program.py:200–706

(
    config,
    train_dataloader,
    valid_dataloader,
    device,
    model,
    loss_class,
    optimizer,
    lr_scheduler,
    post_process_class,
    eval_class,
    pre_best_model_dict,
    logger,
    step_pre_epoch,
    log_writer=None,
    scaler=None,
    amp_level="O2",
    amp_custom_black_list=[],
    amp_custom_white_list=[],
    amp_dtype="float16",
    wd_scheduler=None,
    ema=None,
)

def train(
    config,
    train_dataloader,
    valid_dataloader,
    device,
    model,
    loss_class,
    optimizer,
    lr_scheduler,
    post_process_class,
    eval_class,
    pre_best_model_dict,
    logger,
    step_pre_epoch,
    log_writer=None,
    scaler=None,
    amp_level="O2",
    amp_custom_black_list=[],
    amp_custom_white_list=[],
    amp_dtype="float16",
    wd_scheduler=None,
    ema=None,
):
    cal_metric_during_train = config["Global"].get("cal_metric_during_train", False)
    calc_epoch_interval = config["Global"].get("calc_epoch_interval", 1)
    log_smooth_window = config["Global"]["log_smooth_window"]
    epoch_num = config["Global"]["epoch_num"]
    print_batch_step = config["Global"]["print_batch_step"]
    eval_batch_step = config["Global"]["eval_batch_step"]
    eval_batch_epoch = config["Global"].get("eval_batch_epoch", None)
    profiler_options = config["profiler_options"]
    print_mem_info = config["Global"].get("print_mem_info", True)
    uniform_output_enabled = config["Global"].get("uniform_output_enabled", False)

    global_step = 0
    if "global_step" in pre_best_model_dict:
        global_step = pre_best_model_dict["global_step"]
    start_eval_step = 0
    if isinstance(eval_batch_step, list) and len(eval_batch_step) >= 2:
        start_eval_step = eval_batch_step[0] if not eval_batch_epoch else 0
        eval_batch_step = (
            eval_batch_step[1]
            if not eval_batch_epoch
            else step_pre_epoch * eval_batch_epoch
        )
        if len(valid_dataloader) == 0:
            logger.info(
                "No Images in eval dataset, evaluation during training "
                "will be disabled"
            )
            start_eval_step = 1e111
        logger.info(
            "During the training process, after the {}th iteration, "
            "an evaluation is run every {} iterations".format(
                start_eval_step, eval_batch_step
            )
        )
    save_epoch_step = config["Global"]["save_epoch_step"]