from easydict import EasyDict

agent_num = 10
collector_env_num = 16
evaluator_env_num = 8

main_config = dict(
    exp_name='smac_MMM_coma_seed0',
    env=dict(
        map_name='MMM',
        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,
        stop_value=0.999,
        n_evaluator_episode=32,
        manager=dict(
            shared_memory=False,
            reset_timeout=6000,
        ),
    ),
    policy=dict(
        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.
            obs_shape=dict(
                agent_state=186,
                global_state=290,
            ),
            # (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=16,
            # (List[int]) The size of hidden layer
            actor_hidden_size_list=[64],
        ),
        # used in state_num of hidden_state
        collect=dict(
            n_episode=32,
            unroll_len=10,
            env_num=collector_env_num,
        ),
        eval=dict(env_num=evaluator_env_num, evaluator=dict(eval_freq=100, )),
        other=dict(
            eps=dict(
                type='exp',
                start=0.5,
                end=0.01,
                decay=200000,
            ),
            replay_buffer=dict(
                # (int) max size of replay buffer
                replay_buffer_size=5000,
                # (int) max use count of data, if count is bigger than this value, the data will be removed from buffer
                max_use=10,
            ),
        ),
    ),
)
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='coma'),
    collector=dict(type='episode', get_train_sample=True),
)
create_config = EasyDict(create_config)

if __name__ == '__main__':

    from ding.entry import serial_pipeline
    serial_pipeline((main_config, create_config), seed=0)