Python cmd_util.mujoco_arg_parser方法代碼示例

本文整理匯總了Python中baselines.common.cmd_util.mujoco_arg_parser方法的典型用法代碼示例。如果您正苦於以下問題：Python cmd_util.mujoco_arg_parser方法的具體用法？Python cmd_util.mujoco_arg_parser怎麽用？Python cmd_util.mujoco_arg_parser使用的例子？那麽, 這裏精選的方法代碼示例或許可以為您提供幫助。您也可以進一步了解該方法所在類baselines.common.cmd_util的用法示例。

在下文中一共展示了cmd_util.mujoco_arg_parser方法的7個代碼示例，這些例子默認根據受歡迎程度排序。您可以為喜歡或者感覺有用的代碼點讚，您的評價將有助於係統推薦出更棒的Python代碼示例。

示例1: main

# 需要導入模塊: from baselines.common import cmd_util [as 別名]
# 或者: from baselines.common.cmd_util import mujoco_arg_parser [as 別名]
def main():
    logger.configure()
    parser = mujoco_arg_parser()
    parser.add_argument('--model-path', default=os.path.join(logger.get_dir(), 'humanoid_policy'))
    parser.set_defaults(num_timesteps=int(2e7))
   
    args = parser.parse_args()
    
    if not args.play:
        # train the model
        train(num_timesteps=args.num_timesteps, seed=args.seed, model_path=args.model_path)
    else:       
        # construct the model object, load pre-trained model and render
        pi = train(num_timesteps=1, seed=args.seed)
        U.load_state(args.model_path)
        env = make_mujoco_env('Humanoid-v2', seed=0)

        ob = env.reset()        
        while True:
            action = pi.act(stochastic=False, ob=ob)[0]
            ob, _, done, _ =  env.step(action)
            env.render()
            if done:
                ob = env.reset()

開發者ID:MaxSobolMark，項目名稱:HardRLWithYoutube，代碼行數:26，代碼來源:run_humanoid.py

示例2: main

# 需要導入模塊: from baselines.common import cmd_util [as 別名]
# 或者: from baselines.common.cmd_util import mujoco_arg_parser [as 別名]
def main():
    logger.configure()
    parser = mujoco_arg_parser()
    parser.add_argument('--model-path', default=os.path.join(logger.get_dir(), 'humanoid_policy'))
    parser.set_defaults(num_timesteps=int(2e7))

    args = parser.parse_args()

    if not args.play:
        # train the model
        train(num_timesteps=args.num_timesteps, seed=args.seed, model_path=args.model_path)
    else:
        # construct the model object, load pre-trained model and render
        pi = train(num_timesteps=1, seed=args.seed)
        U.load_state(args.model_path)
        env = make_mujoco_env('Humanoid-v2', seed=0)

        ob = env.reset()
        while True:
            action = pi.act(stochastic=False, ob=ob)[0]
            ob, _, done, _ =  env.step(action)
            env.render()
            if done:
                ob = env.reset()

開發者ID:hiwonjoon，項目名稱:ICML2019-TREX，代碼行數:26，代碼來源:run_humanoid.py

示例3: main

# 需要導入模塊: from baselines.common import cmd_util [as 別名]
# 或者: from baselines.common.cmd_util import mujoco_arg_parser [as 別名]
def main():
    parser = mujoco_arg_parser()
    parser.add_argument('--lr', type=float, default=3e-4, help="Learning rate")
    parser.add_argument('--sil-update', type=float, default=10, help="Number of updates per iteration")
    parser.add_argument('--sil-value', type=float, default=0.01, help="Weight for value update")
    parser.add_argument('--sil-alpha', type=float, default=0.6, help="Alpha for prioritized replay")
    parser.add_argument('--sil-beta', type=float, default=0.1, help="Beta for prioritized replay")

    args = parser.parse_args()
    logger.configure()
    model, env = train(args.env, num_timesteps=args.num_timesteps, seed=args.seed,
            lr=args.lr,
            sil_update=args.sil_update, sil_value=args.sil_value,
            sil_alpha=args.sil_alpha, sil_beta=args.sil_beta)

    if args.play:
        logger.log("Running trained model")
        obs = np.zeros((env.num_envs,) + env.observation_space.shape)
        obs[:] = env.reset()
        while True:
            actions = model.step(obs)[0]
            obs[:]  = env.step(actions)[0]
            env.render()

開發者ID:junhyukoh，項目名稱:self-imitation-learning，代碼行數:25，代碼來源:run_mujoco_sil.py

示例4: main

# 需要導入模塊: from baselines.common import cmd_util [as 別名]
# 或者: from baselines.common.cmd_util import mujoco_arg_parser [as 別名]
def main():
    logger.configure()
    parser = mujoco_arg_parser()
    parser.add_argument('--model-path', default=os.path.join(logger.get_dir(), 'humanoid_policy'))
    parser.set_defaults(num_timesteps=int(5e7))

    args = parser.parse_args()

    if not args.play:
        # train the model
        train(num_timesteps=args.num_timesteps, seed=args.seed, model_path=args.model_path)
    else:
        # construct the model object, load pre-trained model and render
        pi = train(num_timesteps=1, seed=args.seed)
        U.load_state(args.model_path)
        env = make_mujoco_env('Humanoid-v2', seed=0)

        ob = env.reset()
        while True:
            action = pi.act(stochastic=False, ob=ob)[0]
            ob, _, done, _ =  env.step(action)
            env.render()
            if done:
                ob = env.reset()

開發者ID:openai，項目名稱:baselines，代碼行數:26，代碼來源:run_humanoid.py

示例5: main

# 需要導入模塊: from baselines.common import cmd_util [as 別名]
# 或者: from baselines.common.cmd_util import mujoco_arg_parser [as 別名]
def main():
    args = mujoco_arg_parser().parse_args()
    logger.configure()
    train(args.env, num_timesteps=args.num_timesteps, seed=args.seed)

開發者ID:Hwhitetooth，項目名稱:lirpg，代碼行數:6，代碼來源:run_mujoco.py

示例6: main

# 需要導入模塊: from baselines.common import cmd_util [as 別名]
# 或者: from baselines.common.cmd_util import mujoco_arg_parser [as 別名]
def main():
    args = mujoco_arg_parser().parse_args()
    train(args.env, num_timesteps=args.num_timesteps, seed=args.seed)

開發者ID:Hwhitetooth，項目名稱:lirpg，代碼行數:5，代碼來源:run_mujoco.py

示例7: main

# 需要導入模塊: from baselines.common import cmd_util [as 別名]
# 或者: from baselines.common.cmd_util import mujoco_arg_parser [as 別名]
def main():
    args = mujoco_arg_parser().parse_args()
    logger.configure()
    model, env = train(args.env, num_timesteps=args.num_timesteps, seed=args.seed)

    if args.play:
        logger.log("Running trained model")
        obs = np.zeros((env.num_envs,) + env.observation_space.shape)
        obs[:] = env.reset()
        while True:
            actions = model.step(obs)[0]
            obs[:]  = env.step(actions)[0]
            env.render()

開發者ID:junhyukoh，項目名稱:self-imitation-learning，代碼行數:15，代碼來源:run_mujoco.py

注：本文中的baselines.common.cmd_util.mujoco_arg_parser方法示例由純淨天空整理自Github/MSDocs等開源代碼及文檔管理平台，相關代碼片段篩選自各路編程大神貢獻的開源項目，源碼版權歸原作者所有，傳播和使用請參考對應項目的License；未經允許，請勿轉載。