File size: 2,009 Bytes
079c32c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 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 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 |
from easydict import EasyDict
ant_td3_config = dict(
exp_name='ant_td3_seed0',
env=dict(
env_id='Ant-v3',
norm_obs=dict(use_norm=False, ),
norm_reward=dict(use_norm=False, ),
collector_env_num=1,
evaluator_env_num=8,
n_evaluator_episode=8,
stop_value=6000,
manager=dict(shared_memory=False, reset_inplace=True),
),
policy=dict(
cuda=True,
random_collect_size=25000,
model=dict(
obs_shape=111,
action_shape=8,
twin_critic=True,
actor_head_hidden_size=256,
critic_head_hidden_size=256,
action_space='regression',
),
learn=dict(
update_per_collect=1,
batch_size=256,
learning_rate_actor=1e-3,
learning_rate_critic=1e-3,
ignore_done=False,
target_theta=0.005,
discount_factor=0.99,
actor_update_freq=2,
noise=True,
noise_sigma=0.2,
noise_range=dict(
min=-0.5,
max=0.5,
),
),
collect=dict(
n_sample=1,
unroll_len=1,
noise_sigma=0.1,
),
other=dict(replay_buffer=dict(replay_buffer_size=1000000, ), ),
)
)
ant_td3_config = EasyDict(ant_td3_config)
main_config = ant_td3_config
ant_td3_create_config = dict(
env=dict(
type='mujoco',
import_names=['dizoo.mujoco.envs.mujoco_env'],
),
env_manager=dict(type='base'),
policy=dict(
type='td3',
import_names=['ding.policy.td3'],
),
replay_buffer=dict(type='naive', ),
)
ant_td3_create_config = EasyDict(ant_td3_create_config)
create_config = ant_td3_create_config
if __name__ == "__main__":
# or you can enter `ding -m serial -c ant_td3_config.py -s 0 --env-step 1e7`
from ding.entry import serial_pipeline
serial_pipeline((main_config, create_config), seed=0)
|