scripts/launcher_distributed.py [379:451]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
                function_map[value]()
            except Exception as e:
                print(f"An error occurred in function {value}: {e}")
                raise e
        else:
            print(f"No function defined for value {value}")


if __name__ == "__main__":

    report_error = 0
    args, _ = parse_arge()
    print(args)

    # get the current working directory
    current_working_directory = os.getcwd()

    # print output to the console
    print(current_working_directory)

    jinja_env = jinja2.Environment()

    # Dynamically modify fine-tuning yaml file.
    template = jinja_env.from_string(Path(args.tune_finetune_yaml).open().read())
    train_path = os.path.join(args.train_dir, "train.jsonl")
    metric_logger = "DiskLogger"
    if len(args.wandb_api_key) > 0:
        metric_logger = "WandBLogger"

    Path(args.tune_finetune_yaml).open("w").write(
        template.render(
            train_path=train_path,
            log_dir=args.log_dir,
            model_dir=args.model_dir,
            model_output_dir=args.model_output_dir,
            metric_logger=metric_logger,
        )
    )

    epoch = get_epoch(args.tune_finetune_yaml)

    # Dynamically modify Evaluation yaml file.
    template = jinja_env.from_string(Path(args.tune_eval_yaml).open().read())
    Path(args.tune_eval_yaml).open("w").write(
        template.render(
            model_dir=args.model_dir,
            model_output_dir=os.path.join(args.model_output_dir, f"epoch_{epoch}"),
        )
    )

    # Dynamically modify Quantization yaml file.
    template = jinja_env.from_string(Path(args.tune_quant_yaml).open().read())
    Path(args.tune_quant_yaml).open("w").write(
        template.render(
            model_output_dir=os.path.join(args.model_output_dir, f"epoch_{epoch}")
        )
    )

    try:
        print("Starting training...")
        training_function()

        if report_error == 1:
            sys.exit(1)

        print(f"Training completed with code: {report_error}")

    except Exception as e:
        # Log the error
        print(f"Error occurred during training: {str(e)}")

        # Exit with a non-zero status code
        sys.exit(1)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -