Anurag459/smart_elevator_system

1import random2from tasks import easy_env, medium_env, hard_env3from grader import compute_score4 5def run_env(env_name, env_fn):6    env = env_fn()7    state = env.reset()8 9    print(f"[START] task={env_name} env=elevator model=baseline")10 11    total_reward = 0.012    rewards = []13    done = False14    step_num = 015 16    try:17        while not done:18            action = random.choice([0, 1, 2])19            state, reward, done, _ = env.step(action)20 21            reward = float(reward)22            total_reward += reward23            rewards.append(reward)24            step_num += 125 26            print(27                f"[STEP] step={step_num} action={action} reward={reward:.2f} done={str(done).lower()} error=null"28            )29 30        score = compute_score(total_reward)31        score = max(0.0, min(1.0, score))  # clamp32 33        success = "true" if score > 0 else "false"34 35    except Exception as e:36        success = "false"37        print(f"[STEP] step={step_num} action=error reward=0.00 done=true error={str(e)}")38 39    rewards_str = ",".join(f"{r:.2f}" for r in rewards)40 41    print(42        f"[END] success={success} steps={step_num} score={score:.2f} rewards={rewards_str}"43    )44 45 46if __name__ == "__main__":47    run_env("easy", easy_env)48    run_env("medium", medium_env)49    run_env("hard", hard_env)