当前位置: 首页>>代码示例>>Python>>正文


Python atari_wrappers.make_atari方法代码示例

本文整理汇总了Python中baselines.common.atari_wrappers.make_atari方法的典型用法代码示例。如果您正苦于以下问题:Python atari_wrappers.make_atari方法的具体用法?Python atari_wrappers.make_atari怎么用?Python atari_wrappers.make_atari使用的例子?那么, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在baselines.common.atari_wrappers的用法示例。


在下文中一共展示了atari_wrappers.make_atari方法的15个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Python代码示例。

示例1: main

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def main():
    logger.configure()
    env = make_atari('PongNoFrameskip-v4')
    env = bench.Monitor(env, logger.get_dir())
    env = deepq.wrap_atari_dqn(env)

    model = deepq.learn(
        env,
        "conv_only",
        convs=[(32, 8, 4), (64, 4, 2), (64, 3, 1)],
        hiddens=[256],
        dueling=True,
        lr=1e-4,
        total_timesteps=int(1e7),
        buffer_size=10000,
        exploration_fraction=0.1,
        exploration_final_eps=0.01,
        train_freq=4,
        learning_starts=10000,
        target_network_update_freq=1000,
        gamma=0.99,
    )

    model.save('pong_model.pkl')
    env.close() 
开发者ID:hiwonjoon,项目名称:ICML2019-TREX,代码行数:27,代码来源:train_pong.py

示例2: make_env

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def make_env(env_id, env_type, subrank=0, seed=None, reward_scale=1.0, gamestate=None, wrapper_kwargs={}):
    mpi_rank = MPI.COMM_WORLD.Get_rank() if MPI else 0
    if env_type == 'atari':
        env = make_atari(env_id)
    elif env_type == 'retro':
        import retro
        gamestate = gamestate or retro.State.DEFAULT
        env = retro_wrappers.make_retro(game=env_id, max_episode_steps=10000, use_restricted_actions=retro.Actions.DISCRETE, state=gamestate)
    else:
        env = gym.make(env_id)

    env.seed(seed + subrank if seed is not None else None)
    env = Monitor(env,
                  logger.get_dir() and os.path.join(logger.get_dir(), str(mpi_rank) + '.' + str(subrank)),
                  allow_early_resets=True)

    if env_type == 'atari':
        env = wrap_deepmind(env, **wrapper_kwargs)
    elif env_type == 'retro':
        env = retro_wrappers.wrap_deepmind_retro(env, **wrapper_kwargs)

    if reward_scale != 1:
        env = retro_wrappers.RewardScaler(env, reward_scale)

    return env 
开发者ID:hiwonjoon,项目名称:ICML2019-TREX,代码行数:27,代码来源:cmd_util.py

示例3: make_atari_env

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def make_atari_env(env_id, num_env, seed, hparams=None, wrapper_kwargs=None, start_index=0, nsteps=5, **kwargs):
    """
    Create a wrapped, monitored SubprocVecEnv for Atari.
    """
    if wrapper_kwargs is None: wrapper_kwargs = {}
    def make_env(rank): # pylint: disable=C0111
        def _thunk():
            env = make_atari(env_id)
            env.seed(seed + rank)
            env = Monitor(env, logger.get_dir() and os.path.join(logger.get_dir(), str(rank)))

            if rank == start_index and 'video_log_dir' in kwargs:
                env = VideoLogMonitor(env, kwargs['video_log_dir'] + '_rgb', write_attention_video=kwargs['write_attention_video'], hparams=hparams, nsteps=nsteps)

            return wrap_deepmind(env, **wrapper_kwargs)
        return _thunk
    set_global_seeds(seed)

    env_fns = [make_env(i + start_index) for i in range(num_env)]

    global my_subproc_vec_env
    assert my_subproc_vec_env == None
    my_subproc_vec_env = SubprocVecEnv(env_fns)

    return my_subproc_vec_env 
开发者ID:vik-goel,项目名称:MOREL,代码行数:27,代码来源:cmd_util.py

示例4: train

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def train(env_id, num_timesteps, seed, policy, lrschedule, num_cpu):
    def make_env(rank):
        def _thunk():
            env = make_atari(env_id)
            env.seed(seed + rank)
            env = bench.Monitor(env, logger.get_dir() and os.path.join(logger.get_dir(), str(rank)))
            gym.logger.setLevel(logging.WARN)
            return wrap_deepmind(env)
        return _thunk
    set_global_seeds(seed)
    env = SubprocVecEnv([make_env(i) for i in range(num_cpu)])
    if policy == 'cnn':
        policy_fn = CnnPolicy
    elif policy == 'lstm':
        policy_fn = LstmPolicy
    elif policy == 'lnlstm':
        policy_fn = LnLstmPolicy
    learn(policy_fn, env, seed, total_timesteps=int(num_timesteps * 1.1), lrschedule=lrschedule)
    env.close() 
开发者ID:cxxgtxy,项目名称:deeprl-baselines,代码行数:21,代码来源:run_atari.py

示例5: train

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def train(env_id, num_timesteps, seed, policy, lrschedule, num_cpu):
    def make_env(rank):
        def _thunk():
            env = make_atari(env_id)
            env.seed(seed + rank)
            env = bench.Monitor(env, logger.get_dir() and os.path.join(logger.get_dir(), str(rank)))
            gym.logger.setLevel(logging.WARN)
            return wrap_deepmind(env)
        return _thunk
    set_global_seeds(seed)
    env = SubprocVecEnv([make_env(i) for i in range(num_cpu)])
    if policy == 'cnn':
        policy_fn = AcerCnnPolicy
    elif policy == 'lstm':
        policy_fn = AcerLstmPolicy
    else:
        print("Policy {} not implemented".format(policy))
        return
    learn(policy_fn, env, seed, total_timesteps=int(num_timesteps * 1.1), lrschedule=lrschedule)
    env.close() 
开发者ID:cxxgtxy,项目名称:deeprl-baselines,代码行数:22,代码来源:run_atari.py

示例6: main

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def main():
    parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)
    parser.add_argument('--env', help='environment ID', default='BreakoutNoFrameskip-v4')
    parser.add_argument('--seed', help='RNG seed', type=int, default=0)
    parser.add_argument('--prioritized', type=int, default=1)
    parser.add_argument('--dueling', type=int, default=1)
    parser.add_argument('--num-timesteps', type=int, default=int(10e6))
    args = parser.parse_args()
    logger.configure()
    set_global_seeds(args.seed)
    env = make_atari(args.env)
    env = bench.Monitor(env, logger.get_dir())
    env = deepq.wrap_atari_dqn(env)
    model = deepq.models.cnn_to_mlp(
        convs=[(32, 8, 4), (64, 4, 2), (64, 3, 1)],
        hiddens=[256],
        dueling=bool(args.dueling),
    )
    act = deepq.learn(
        env,
        q_func=model,
        lr=1e-4,
        max_timesteps=args.num_timesteps,
        buffer_size=10000,
        exploration_fraction=0.1,
        exploration_final_eps=0.01,
        train_freq=4,
        learning_starts=10000,
        target_network_update_freq=1000,
        gamma=0.99,
        prioritized_replay=bool(args.prioritized)
    )
    # act.save("pong_model.pkl") XXX
    env.close() 
开发者ID:Hwhitetooth,项目名称:lirpg,代码行数:36,代码来源:run_atari.py

示例7: train

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def train(env_id, num_timesteps, seed):
    from baselines.trpo_mpi.nosharing_cnn_policy import CnnPolicy
    from baselines.trpo_mpi import trpo_mpi
    import baselines.common.tf_util as U
    rank = MPI.COMM_WORLD.Get_rank()
    sess = U.single_threaded_session()
    sess.__enter__()
    if rank == 0:
        logger.configure()
    else:
        logger.configure(format_strs=[])

    workerseed = seed + 10000 * MPI.COMM_WORLD.Get_rank()
    set_global_seeds(workerseed)
    env = make_atari(env_id)
    def policy_fn(name, ob_space, ac_space): #pylint: disable=W0613
        return CnnPolicy(name=name, ob_space=env.observation_space, ac_space=env.action_space)
    env = bench.Monitor(env, logger.get_dir() and osp.join(logger.get_dir(), str(rank)))
    env.seed(workerseed)

    env = wrap_deepmind(env)
    env.seed(workerseed)

    trpo_mpi.learn(env, policy_fn, timesteps_per_batch=512, max_kl=0.001, cg_iters=10, cg_damping=1e-3,
        max_timesteps=int(num_timesteps * 1.1), gamma=0.98, lam=1.0, vf_iters=3, vf_stepsize=1e-4, entcoeff=0.00)
    env.close() 
开发者ID:Hwhitetooth,项目名称:lirpg,代码行数:28,代码来源:run_atari.py

示例8: train

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def train(env_id, num_timesteps, seed):
    from baselines.ppo1 import pposgd_simple, cnn_policy
    import baselines.common.tf_util as U
    rank = MPI.COMM_WORLD.Get_rank()
    sess = U.single_threaded_session()
    sess.__enter__()
    if rank == 0:
        logger.configure()
    else:
        logger.configure(format_strs=[])
    workerseed = seed + 10000 * MPI.COMM_WORLD.Get_rank()
    set_global_seeds(workerseed)
    env = make_atari(env_id)
    def policy_fn(name, ob_space, ac_space): #pylint: disable=W0613
        return cnn_policy.CnnPolicy(name=name, ob_space=ob_space, ac_space=ac_space)
    env = bench.Monitor(env, logger.get_dir() and
        osp.join(logger.get_dir(), str(rank)))
    env.seed(workerseed)

    env = wrap_deepmind(env)
    env.seed(workerseed)

    pposgd_simple.learn(env, policy_fn,
        max_timesteps=int(num_timesteps * 1.1),
        timesteps_per_actorbatch=256,
        clip_param=0.2, entcoeff=0.01,
        optim_epochs=4, optim_stepsize=1e-3, optim_batchsize=64,
        gamma=0.99, lam=0.95,
        schedule='linear'
    )
    env.close() 
开发者ID:Hwhitetooth,项目名称:lirpg,代码行数:33,代码来源:run_atari.py

示例9: make_atari_env

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def make_atari_env(env_id, num_env, seed, wrapper_kwargs=None, start_index=0):
    """
    Create a wrapped, monitored SubprocVecEnv for Atari.
    """
    if wrapper_kwargs is None: wrapper_kwargs = {}
    def make_env(rank): # pylint: disable=C0111
        def _thunk():
            env = make_atari(env_id)
            env.seed(seed + rank)
            env = Monitor(env, logger.get_dir() and os.path.join(logger.get_dir(), str(rank)))
            return wrap_deepmind(env, **wrapper_kwargs)
        return _thunk
    set_global_seeds(seed)
    return SubprocVecEnv([make_env(i + start_index) for i in range(num_env)]) 
开发者ID:Hwhitetooth,项目名称:lirpg,代码行数:16,代码来源:cmd_util.py

示例10: make_env

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def make_env(env_name, rank, seed):
    env = make_atari(env_name)
    env.seed(seed + rank)
    env = wrap_deepmind(env)
    return env 
开发者ID:lnpalmer,项目名称:A2C,代码行数:7,代码来源:envs.py

示例11: main

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def main():
    parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)
    parser.add_argument('--env', help='environment ID', default='BreakoutNoFrameskip-v4')
    parser.add_argument('--seed', help='RNG seed', type=int, default=0)
    parser.add_argument('--prioritized', type=int, default=1)
    parser.add_argument('--prioritized-replay-alpha', type=float, default=0.6)
    parser.add_argument('--dueling', type=int, default=1)
    parser.add_argument('--num-timesteps', type=int, default=int(10e6))
    parser.add_argument('--checkpoint-freq', type=int, default=10000)
    parser.add_argument('--checkpoint-path', type=str, default=None)

    args = parser.parse_args()
    logger.configure()
    set_global_seeds(args.seed)
    env = make_atari(args.env)
    env = bench.Monitor(env, logger.get_dir())
    env = deepq.wrap_atari_dqn(env)

    deepq.learn(
        env,
        "conv_only",
        convs=[(32, 8, 4), (64, 4, 2), (64, 3, 1)],
        hiddens=[256],
        dueling=bool(args.dueling),
        lr=1e-4,
        total_timesteps=args.num_timesteps,
        buffer_size=10000,
        exploration_fraction=0.1,
        exploration_final_eps=0.01,
        train_freq=4,
        learning_starts=10000,
        target_network_update_freq=1000,
        gamma=0.99,
        prioritized_replay=bool(args.prioritized),
        prioritized_replay_alpha=args.prioritized_replay_alpha,
        checkpoint_freq=args.checkpoint_freq,
        checkpoint_path=args.checkpoint_path,
    )

    env.close() 
开发者ID:MaxSobolMark,项目名称:HardRLWithYoutube,代码行数:42,代码来源:run_atari.py

示例12: train

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def train(env_id, num_timesteps, seed):
    from baselines.ppo1 import pposgd_simple, cnn_policy
    import baselines.common.tf_util as U
    rank = MPI.COMM_WORLD.Get_rank()
    sess = U.single_threaded_session()
    sess.__enter__()
    if rank == 0:
        logger.configure()
    else:
        logger.configure(format_strs=[])
    workerseed = seed + 10000 * MPI.COMM_WORLD.Get_rank() if seed is not None else None
    set_global_seeds(workerseed)
    env = make_atari(env_id)
    def policy_fn(name, ob_space, ac_space): #pylint: disable=W0613
        return cnn_policy.CnnPolicy(name=name, ob_space=ob_space, ac_space=ac_space)
    env = bench.Monitor(env, logger.get_dir() and
        osp.join(logger.get_dir(), str(rank)))
    env.seed(workerseed)

    env = wrap_deepmind(env)
    env.seed(workerseed)

    pposgd_simple.learn(env, policy_fn,
        max_timesteps=int(num_timesteps * 1.1),
        timesteps_per_actorbatch=256,
        clip_param=0.2, entcoeff=0.01,
        optim_epochs=4, optim_stepsize=1e-3, optim_batchsize=64,
        gamma=0.99, lam=0.95,
        schedule='linear'
    )
    env.close() 
开发者ID:MaxSobolMark,项目名称:HardRLWithYoutube,代码行数:33,代码来源:run_atari.py

示例13: make_atari_env

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def make_atari_env(env_id, num_env, seed, wrapper_kwargs=None, start_index=0):
    """
    Create a wrapped, monitored SubprocVecEnv for Atari.
    """
    if wrapper_kwargs is None: wrapper_kwargs = {}
    mpi_rank = MPI.COMM_WORLD.Get_rank() if MPI else 0
    def make_env(rank): # pylint: disable=C0111
        def _thunk():
            env = make_atari(env_id)
            env.seed(seed + 10000*mpi_rank + rank if seed is not None else None)
            env = Monitor(env, logger.get_dir() and os.path.join(logger.get_dir(), str(mpi_rank) + '.' + str(rank)))
            return wrap_deepmind(env, **wrapper_kwargs)
        return _thunk
    set_global_seeds(seed)
    return SubprocVecEnv([make_env(i + start_index) for i in range(num_env)]) 
开发者ID:MaxSobolMark,项目名称:HardRLWithYoutube,代码行数:17,代码来源:cmd_util.py

示例14: make_env

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def make_env(env_id, seed, rank, episode_life=True):
    def _thunk():
        random_seed(seed)
        if env_id.startswith("dm"):
            import dm_control2gym
            _, domain, task = env_id.split('-')
            env = dm_control2gym.make(domain_name=domain, task_name=task)
        else:
            env = gym.make(env_id)
        is_atari = hasattr(gym.envs, 'atari') and isinstance(
            env.unwrapped, gym.envs.atari.atari_env.AtariEnv)
        if is_atari:
            env = make_atari(env_id)
        env.seed(seed + rank)
        env = OriginalReturnWrapper(env)
        if is_atari:
            env = wrap_deepmind(env,
                                episode_life=episode_life,
                                clip_rewards=False,
                                frame_stack=False,
                                scale=False)
            obs_shape = env.observation_space.shape
            if len(obs_shape) == 3:
                env = TransposeImage(env)
            env = FrameStack(env, 4)

        return env

    return _thunk 
开发者ID:ShangtongZhang,项目名称:DeepRL,代码行数:31,代码来源:envs.py

示例15: main

# 需要导入模块: from baselines.common import atari_wrappers [as 别名]
# 或者: from baselines.common.atari_wrappers import make_atari [as 别名]
def main():
    parser = argparse.ArgumentParser(formatter_class=argparse.ArgumentDefaultsHelpFormatter)
    parser.add_argument('--env', help='environment ID', default='BreakoutNoFrameskip-v4')
    parser.add_argument('--seed', help='RNG seed', type=int, default=0)
    parser.add_argument('--prioritized', type=int, default=1)
    parser.add_argument('--prioritized-replay-alpha', type=float, default=0.6)
    parser.add_argument('--dueling', type=int, default=1)
    parser.add_argument('--num-timesteps', type=int, default=int(10e6))
    args = parser.parse_args()
    logger.configure()
    set_global_seeds(args.seed)
    env = make_atari(args.env)
    env = bench.Monitor(env, logger.get_dir())
    env = deepq.wrap_atari_dqn(env)
    model = deepq.models.cnn_to_mlp(
        convs=[(32, 8, 4), (64, 4, 2), (64, 3, 1)],
        hiddens=[256],
        dueling=bool(args.dueling),
    )

    deepq.learn(
        env,
        q_func=model,
        lr=1e-4,
        max_timesteps=args.num_timesteps,
        buffer_size=10000,
        exploration_fraction=0.1,
        exploration_final_eps=0.01,
        train_freq=4,
        learning_starts=10000,
        target_network_update_freq=1000,
        gamma=0.99,
        prioritized_replay=bool(args.prioritized),
        prioritized_replay_alpha=args.prioritized_replay_alpha
    )

    env.close() 
开发者ID:flyyufelix,项目名称:sonic_contest,代码行数:39,代码来源:run_atari.py


注:本文中的baselines.common.atari_wrappers.make_atari方法示例由纯净天空整理自Github/MSDocs等开源代码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。