PongNoFrameskip-v4-MuZero

Play PongNoFrameskip-v4 with MuZero Policy

Model Description

<!-- Provide a longer summary of what this model is. -->

This implementation applies MuZero to the OpenAI/Gym/Atari PongNoFrameskip-v4 environment using LightZero and DI-engine.

LightZero is an efficient, easy-to-understand open-source toolkit that merges Monte Carlo Tree Search (MCTS) with Deep Reinforcement Learning (RL), simplifying their integration for developers and researchers. More details are in paper LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios.

Model Usage

Install the Dependencies

<details close> <summary>(Click for Details)</summary>

# install huggingface_ding
git clone https://github.com/opendilab/huggingface_ding.git
pip3 install -e ./huggingface_ding/
# install environment dependencies if needed

pip3 install DI-engine[common_env,video]
pip3 install LightZero

</details>

Git Clone from Huggingface and Run the Model

<details close> <summary>(Click for Details)</summary>

# running with trained model
python3 -u run.py

run.py

from lzero.agent import MuZeroAgent
from ding.config import Config
from easydict import EasyDict
import torch

# Pull model from files which are git cloned from huggingface
policy_state_dict = torch.load("pytorch_model.bin", map_location=torch.device("cpu"))
cfg = EasyDict(Config.file_to_dict("policy_config.py").cfg_dict)
# Instantiate the agent
agent = MuZeroAgent(
    env_id="PongNoFrameskip-v4", exp_name="PongNoFrameskip-v4-MuZero", cfg=cfg.exp_config, policy_state_dict=policy_state_dict
)
# Continue training
agent.train(step=5000)
# Render the new agent performance
agent.deploy(enable_save_replay=True)

</details>

Run Model by Using Huggingface_ding

<details close> <summary>(Click for Details)</summary>

# running with trained model
python3 -u run.py

run.py

from lzero.agent import MuZeroAgent
from huggingface_ding import pull_model_from_hub

# Pull model from Hugggingface hub
policy_state_dict, cfg = pull_model_from_hub(repo_id="OpenDILabCommunity/PongNoFrameskip-v4-MuZero")
# Instantiate the agent
agent = MuZeroAgent(
    env_id="PongNoFrameskip-v4", exp_name="PongNoFrameskip-v4-MuZero", cfg=cfg.exp_config, policy_state_dict=policy_state_dict
)
# Continue training
agent.train(step=5000)
# Render the new agent performance
agent.deploy(enable_save_replay=True)

</details>

Model Training

Train the Model and Push to Huggingface_hub

<details close> <summary>(Click for Details)</summary>

#Training Your Own Agent
python3 -u train.py

train.py

from lzero.agent import MuZeroAgent
from huggingface_ding import push_model_to_hub

# Instantiate the agent
agent = MuZeroAgent(env_id="PongNoFrameskip-v4", exp_name="PongNoFrameskip-v4-MuZero")
# Train the agent
return_ = agent.train(step=int(500000))
# Push model to huggingface hub
push_model_to_hub(
    agent=agent.best,
    env_name="OpenAI/Gym/Atari",
    task_name="PongNoFrameskip-v4",
    algo_name="MuZero",
    github_repo_url="https://github.com/opendilab/LightZero",
    github_doc_model_url=None,
    github_doc_env_url=None,
    installation_guide='''
pip3 install DI-engine[common_env,video]
pip3 install LightZero
''',
    usage_file_by_git_clone="./muzero/pong_muzero_deploy.py",
    usage_file_by_huggingface_ding="./muzero/pong_muzero_download.py",
    train_file="./muzero/pong_muzero.py",
    repo_id="OpenDILabCommunity/PongNoFrameskip-v4-MuZero",
    platform_info="[LightZero](https://github.com/opendilab/LightZero) and [DI-engine](https://github.com/opendilab/di-engine)",
    model_description="**LightZero** is an efficient, easy-to-understand open-source toolkit that merges Monte Carlo Tree Search (MCTS) with Deep Reinforcement Learning (RL), simplifying their integration for developers and researchers. More details are in paper [LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios](https://huggingface.co/papers/2310.08348).",
    create_repo=True
)

</details>

Configuration

<details close> <summary>(Click for Details)</summary>

exp_config = {
    'main_config': {
        'exp_name': 'PongNoFrameskip-v4-MuZero',
        'seed': 0,
        'env': {
            'stop_value': 1000000,
            'env_id': 'PongNoFrameskip-v4',
            'env_name': 'PongNoFrameskip-v4',
            'obs_shape': [4, 96, 96],
            'collector_env_num': 8,
            'evaluator_env_num': 3,
            'n_evaluator_episode': 3,
            'manager': {
                'shared_memory': False
            }
        },
        'policy': {
            'on_policy': False,
            'cuda': True,
            'multi_gpu': False,
            'bp_update_sync': True,
            'traj_len_inf': False,
            'model': {
                'observation_shape': [4, 96, 96],
                'frame_stack_num': 4,
                'action_space_size': 6,
                'downsample': True,
                'self_supervised_learning_loss': True,
                'discrete_action_encoding_type': 'one_hot',
                'norm_type': 'BN'
            },
            'use_rnd_model': False,
            'sampled_algo': False,
            'gumbel_algo': False,
            'mcts_ctree': True,
            'collector_env_num': 8,
            'evaluator_env_num': 3,
            'env_type': 'not_board_games',
            'action_type': 'fixed_action_space',
            'battle_mode': 'play_with_bot_mode',
            'monitor_extra_statistics': True,
            'game_segment_length': 400,
            'transform2string': False,
            'gray_scale': False,
            'use_augmentation': True,
            'augmentation': ['shift', 'intensity'],
            'ignore_done': False,
            'update_per_collect': 1000,
            'model_update_ratio': 0.1,
            'batch_size': 256,
            'optim_type': 'SGD',
            'learning_rate': 0.2,
            'target_update_freq': 100,
            'target_update_freq_for_intrinsic_reward': 1000,
            'weight_decay': 0.0001,
            'momentum': 0.9,
            'grad_clip_value': 10,
            'n_episode': 8,
            'num_simulations': 50,
            'discount_factor': 0.997,
            'td_steps': 5,
            'num_unroll_steps': 5,
            'reward_loss_weight': 1,
            'value_loss_weight': 0.25,
            'policy_loss_weight': 1,
            'policy_entropy_loss_weight': 0,
            'ssl_loss_weight': 2,
            'lr_piecewise_constant_decay': True,
            'threshold_training_steps_for_final_lr': 50000,
            'manual_temperature_decay': False,
            'threshold_training_steps_for_final_temperature': 100000,
            'fixed_temperature_value': 0.25,
            'use_ture_chance_label_in_chance_encoder': False,
            'use_priority': True,
            'priority_prob_alpha': 0.6,
            'priority_prob_beta': 0.4,
            'root_dirichlet_alpha': 0.3,
            'root_noise_weight': 0.25,
            'random_collect_episode_num': 0,
            'eps': {
                'eps_greedy_exploration_in_collect': False,
                'type': 'linear',
                'start': 1.0,
                'end': 0.05,
                'decay': 100000
            },
            'cfg_type': 'MuZeroPolicyDict',
            'reanalyze_ratio': 0.0,
            'eval_freq': 2000,
            'replay_buffer_size': 1000000
        },
        'wandb_logger': {
            'gradient_logger': False,
            'video_logger': False,
            'plot_logger': False,
            'action_logger': False,
            'return_logger': False
        }
    },
    'create_config': {
        'env': {
            'type': 'atari_lightzero',
            'import_names': ['zoo.atari.envs.atari_lightzero_env']
        },
        'env_manager': {
            'type': 'subprocess'
        },
        'policy': {
            'type': 'muzero',
            'import_names': ['lzero.policy.muzero']
        }
    }
}

</details>

Training Procedure

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->

• Weights & Biases (wandb): monitor link

Model Information

<!-- Provide the basic links for the model. -->

• Github Repository: repo link
• Doc: Algorithm link
• Configuration: config link
• Demo: video

<!-- Provide the size information for the model. -->

• Parameters total size: 24013.13 KB
• Last Update Date: 2023-12-15

Environments

<!-- Address questions around what environment the model is intended to be trained and deployed at, including the necessary information needed to be provided for future users. -->

• Benchmark: OpenAI/Gym/Atari
• Task: PongNoFrameskip-v4
• Gym version: 0.25.1
• DI-engine version: v0.5.0
• PyTorch version: 2.0.1+cu117
• Doc: Environments link