from easydict import EasyDict

collector_env_num = 8
evaluator_env_num = 5
minigrid_ppo_stdim_config = dict(
    exp_name="minigrid_empty8_onppo_stdim_seed0",
    env=dict(
        collector_env_num=collector_env_num,
        evaluator_env_num=evaluator_env_num,
        n_evaluator_episode=evaluator_env_num,
        # typical MiniGrid env id:
        # {'MiniGrid-Empty-8x8-v0', 'MiniGrid-FourRooms-v0', 'MiniGrid-DoorKey-8x8-v0','MiniGrid-DoorKey-16x16-v0'},
        # please refer to https://github.com/Farama-Foundation/MiniGrid for details.
        env_id='MiniGrid-Empty-8x8-v0',
        max_step=300,
        stop_value=0.96,
    ),
    policy=dict(
        cuda=True,
        recompute_adv=True,
        action_space='discrete',
        model=dict(
            obs_shape=2835,
            action_shape=7,
            action_space='discrete',
            encoder_hidden_size_list=[256, 128, 64, 64],
        ),
        aux_model=dict(
            encode_shape=64,
            heads=[1, 1],
            loss_type='infonce',
            temperature=1.0,
        ),
        # the weight of the auxiliary loss to the TD loss
        aux_loss_weight=0.003,
        learn=dict(
            epoch_per_collect=10,
            update_per_collect=1,
            batch_size=320,
            learning_rate=3e-4,
            value_weight=0.5,
            entropy_weight=0.001,
            clip_ratio=0.2,
            adv_norm=True,
            value_norm=True,
        ),
        collect=dict(
            collector_env_num=collector_env_num,
            n_sample=int(3200),
            unroll_len=1,
            discount_factor=0.99,
            gae_lambda=0.95,
        ),
    ),
)
minigrid_ppo_stdim_config = EasyDict(minigrid_ppo_stdim_config)
main_config = minigrid_ppo_stdim_config
minigrid_ppo_stdim_create_config = dict(
    env=dict(
        type='minigrid',
        import_names=['dizoo.minigrid.envs.minigrid_env'],
    ),
    env_manager=dict(type='subprocess'),
    policy=dict(type='ppo_stdim'),
)
minigrid_ppo_stdim_create_config = EasyDict(minigrid_ppo_stdim_create_config)
create_config = minigrid_ppo_stdim_create_config

if __name__ == "__main__":
    # or you can enter `ding -m serial -c minigrid_onppo_stdim_config.py -s 0`
    from ding.entry import serial_pipeline_onpolicy
    serial_pipeline_onpolicy([main_config, create_config], seed=0)