from easydict import EasyDict
import ding.envs.gym_env

cfg = dict(
    exp_name='LunarLander-v2-SQL',
    env=dict(
        collector_env_num=8,
        evaluator_env_num=8,
        env_id='LunarLander-v2',
        n_evaluator_episode=8,
        stop_value=260,
    ),
    policy=dict(
        cuda=True,
        model=dict(
            obs_shape=8,
            action_shape=4,
            encoder_hidden_size_list=[128, 128, 64],
            dueling=True,
        ),
        nstep=1,
        discount_factor=0.97,
        learn=dict(batch_size=64, learning_rate=0.001, alpha=0.08),
        collect=dict(n_sample=64),
        eval=dict(evaluator=dict(eval_freq=50, )),  # note: this is the times after which you learns to evaluate
        other=dict(
            eps=dict(
                type='exp',
                start=0.95,
                end=0.1,
                decay=10000,
            ),
            replay_buffer=dict(replay_buffer_size=20000, ),
        ),
    ),
    wandb_logger=dict(
        gradient_logger=True, video_logger=True, plot_logger=True, action_logger=True, return_logger=False
    ),
)

cfg = EasyDict(cfg)

env = ding.envs.gym_env.env