File size: 3,401 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 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 |
from easydict import EasyDict
agent_num = 27
collector_env_num = 8
evaluator_env_num = 8
special_global_state = True
main_config = dict(
exp_name='smac_27m30m_mappo_seed0',
env=dict(
map_name='27m_vs_30m',
difficulty=7,
reward_only_positive=True,
mirror_opponent=False,
agent_num=agent_num,
collector_env_num=collector_env_num,
evaluator_env_num=evaluator_env_num,
n_evaluator_episode=32,
stop_value=0.99,
death_mask=False,
special_global_state=special_global_state,
manager=dict(
shared_memory=False,
reset_timeout=6000,
),
),
policy=dict(
cuda=True,
on_policy=True,
multi_agent=True,
continuous=False,
model=dict(
# (int) agent_num: The number of the agent.
# For SMAC 3s5z, agent_num=8; for 2c_vs_64zg, agent_num=2.
agent_num=agent_num,
# (int) obs_shape: The shapeension of observation of each agent.
# For 3s5z, obs_shape=150; for 2c_vs_64zg, agent_num=404.
# (int) global_obs_shape: The shapeension of global observation.
# For 3s5z, obs_shape=216; for 2c_vs_64zg, agent_num=342.
agent_obs_shape=348,
global_obs_shape=1454,
# (int) action_shape: The number of action which each agent can take.
# action_shape= the number of common action (6) + the number of enemies.
# For 3s5z, obs_shape=14 (6+8); for 2c_vs_64zg, agent_num=70 (6+64).
action_shape=36,
# (List[int]) The size of hidden layer
# hidden_size_list=[64],
# delete encode in code
actor_head_hidden_size=512,
critic_head_hidden_size=1024,
),
# used in state_num of hidden_state
learn=dict(
epoch_per_collect=5,
batch_size=3200,
learning_rate=5e-4,
# ==============================================================
# The following configs is algorithm-specific
# ==============================================================
# (float) The loss weight of value network, policy network weight is set to 1
value_weight=0.5,
# (float) The loss weight of entropy regularization, policy network weight is set to 1
entropy_weight=0.01,
# (float) PPO clip ratio, defaults to 0.2
clip_ratio=0.2,
# (bool) Whether to use advantage norm in a whole training batch
adv_norm=False,
value_norm=True,
ppo_param_init=True,
grad_clip_type='clip_norm',
grad_clip_value=10,
ignore_done=False,
),
collect=dict(env_num=collector_env_num, n_sample=3200),
eval=dict(
evaluator=dict(eval_freq=100, ),
env_num=evaluator_env_num,
),
),
)
main_config = EasyDict(main_config)
create_config = dict(
env=dict(
type='smac',
import_names=['dizoo.smac.envs.smac_env'],
),
env_manager=dict(type='subprocess'),
policy=dict(type='ppo'),
)
create_config = EasyDict(create_config)
if __name__ == '__main__':
from ding.entry import serial_pipeline_onpolicy
serial_pipeline_onpolicy((main_config, create_config), seed=0)
|