pendulum_td3_config.py 1.6 KB
Newer Older
N
v0.1.0  
niuyazhe 已提交
1 2 3
from easydict import EasyDict

pendulum_td3_config = dict(
4
    exp_name='pendulum_td3',
N
v0.1.0  
niuyazhe 已提交
5 6 7 8 9 10 11 12 13 14 15
    env=dict(
        collector_env_num=8,
        evaluator_env_num=5,
        # (bool) Scale output action into legal range.
        act_scale=True,
        n_evaluator_episode=5,
        stop_value=-250,
    ),
    policy=dict(
        cuda=False,
        priority=False,
16
        random_collect_size=800,
N
v0.1.0  
niuyazhe 已提交
17 18 19 20
        model=dict(
            obs_shape=3,
            action_shape=1,
            twin_critic=True,
21
            action_space='regression',
N
v0.1.0  
niuyazhe 已提交
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58
        ),
        learn=dict(
            update_per_collect=2,
            batch_size=128,
            learning_rate_actor=0.001,
            learning_rate_critic=0.001,
            ignore_done=True,
            actor_update_freq=2,
            noise=True,
            noise_sigma=0.1,
            noise_range=dict(
                min=-0.5,
                max=0.5,
            ),
        ),
        collect=dict(
            n_sample=48,
            noise_sigma=0.1,
            collector=dict(collect_print_freq=1000, ),
        ),
        eval=dict(evaluator=dict(eval_freq=100, ), ),
        other=dict(replay_buffer=dict(replay_buffer_size=20000, ), ),
    ),
)
pendulum_td3_config = EasyDict(pendulum_td3_config)
main_config = pendulum_td3_config

pendulum_td3_create_config = dict(
    env=dict(
        type='pendulum',
        import_names=['dizoo.classic_control.pendulum.envs.pendulum_env'],
    ),
    env_manager=dict(type='base'),
    policy=dict(type='td3'),
)
pendulum_td3_create_config = EasyDict(pendulum_td3_create_config)
create_config = pendulum_td3_create_config