run_bc.py [103:114]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    evaluations = []
    training_iters = 0
    max_timesteps = args.num_epochs * args.num_steps_per_epoch
    best_score = -100.
    while training_iters < max_timesteps:
        iterations = int(args.eval_freq * args.num_steps_per_epoch)
        utils.print_banner(f"Train step: {training_iters}", separator="*", num_star=90)
        agent.train(data_sampler,
                    iterations=iterations,
                    batch_size=args.batch_size)
        training_iters += iterations
        curr_epoch = int(training_iters // int(args.num_steps_per_epoch))
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



run_offline.py [105:116]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    evaluations = []
    training_iters = 0
    max_timesteps = args.num_epochs * args.num_steps_per_epoch
    best_score = -100.
    while training_iters < max_timesteps:
        iterations = int(args.eval_freq * args.num_steps_per_epoch)
        utils.print_banner(f"Train step: {training_iters}", separator="*", num_star=90)
        agent.train(data_sampler,
                    iterations=iterations,
                    batch_size=args.batch_size)
        training_iters += iterations
        curr_epoch = int(training_iters // int(args.num_steps_per_epoch))
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -