トレーニング速度が 17% 向上 4 番目のパラダイムのオープンソース強化学習研究フレームワークは、シングルおよびマルチエージェントのトレーニングをサポートします。

<code>pip install openrl</code>

<code>conda install -c openrl openrl</code>

<code># train_ppo.pyfrom openrl.envs.common import makefrom openrl.modules.common import PPONet as Netfrom openrl.runners.common import PPOAgent as Agentenv = make ("CartPole-v1", env_num=9) # 创建环境，并设置环境并行数为 9net = Net (env) # 创建神经网络agent = Agent (net) # 初始化智能体agent.train (total_time_steps=20000) # 开始训练，并设置环境运行总步数为 20000</code>

<code># train_ppo.pyfrom openrl.envs.common import makefrom openrl.modules.common import PPONet as Netfrom openrl.runners.common import PPOAgent as Agentdef train ():# 创建 MPE 环境，使用异步环境，即每个智能体独立运行env = make ("simple_spread",env_num=100,asynchrnotallow=True,)# 创建 神经网络，使用 GPU 进行训练net = Net (env, device="cuda")agent = Agent (net) # 初始化训练器# 开始训练agent.train (total_time_steps=5000000)# 保存训练完成的智能体agent.save ("./ppo_agent/")if __name__ == "__main__":train ()</code>

<code># mpe_ppo.yamlseed: 0 # 设置 seed，保证每次实验结果一致lr: 7e-4 # 设置学习率episode_length: 25 # 设置每个 episode 的长度use_recurrent_policy: true # 设置是否使用 RNNuse_joint_action_loss: true # 设置是否使用 JRPO 算法use_valuenorm: true # 设置是否使用 value normalization</code>

<code>python train_ppo.py --config mpe_ppo.yaml</code>

<code>env = make ("simple_spread", env_num=9, render_mode="group_human")</code>

<code>from openrl.envs.wrappers import GIFWrapperenv = GIFWrapper (env, "test_simple_spread.gif")</code>

<code># test_ppo.pyfrom openrl.envs.common import makefrom openrl.modules.common import PPONet as Netfrom openrl.runners.common import PPOAgent as Agentfrom openrl.envs.wrappers import GIFWrapper # 用于生成 gifdef test ():# 创建 MPE 环境env = make ( "simple_spread", env_num=4)# 使用 GIFWrapper，用于生成 gifenv = GIFWrapper (env, "test_simple_spread.gif")agent = Agent (Net (env)) # 创建 智能体# 保存智能体agent.save ("./ppo_agent/")# 加载智能体agent.load ('./ppo_agent/')# 开始测试obs, _ = env.reset ()while True:# 智能体根据 observation 预测下一个动作action, _ = agent.act (obs)obs, r, done, info = env.step (action)if done.any ():breakenv.close ()if __name__ == "__main__":test ()</code>

<code># train_ppo.pyfrom openrl.envs.common import makefrom openrl.modules.common import PPONet as Netfrom openrl.runners.common import PPOAgent as Agentfrom openrl.configs.config import create_config_parserdef train ():# 添加读取配置文件的代码cfg_parser = create_config_parser ()cfg = cfg_parser.parse_args ()# 创建 NLP 环境env = make ("daily_dialog",env_num=2,asynchrnotallow=True,cfg=cfg,)net = Net (env, cfg=cfg, device="cuda")agent = Agent (net)agent.train (total_time_steps=5000000)if __name__ == "__main__":train ()</code>

<code># nlp_ppo.yamldata_path: daily_dialog # 数据集路径env: # 环境所用到的参数args: {'tokenizer_path': 'gpt2'} # 读取 tokenizer 的路径seed: 0 # 设置 seed，保证每次实验结果一致lr: 1e-6 # 设置 policy 模型的学习率critic_lr: 1e-6 # 设置 critic 模型的学习率episode_length: 20 # 设置每个 episode 的长度use_recurrent_policy: true</code>

<code># nlp_ppo.yaml# 预训练模型路径model_path: rajkumarrrk/gpt2-fine-tuned-on-daily-dialog use_share_model: true # 策略网络和价值网络是否共享模型ppo_epoch: 5 # ppo 训练迭代次数data_path: daily_dialog # 数据集名称或者路径env: # 环境所用到的参数args: {'tokenizer_path': 'gpt2'} # 读取 tokenizer 的路径lr: 1e-6 # 设置 policy 模型的学习率critic_lr: 1e-6 # 设置 critic 模型的学习率episode_length: 128 # 设置每个 episode 的长度num_mini_batch: 20</code>

<code># train_ppo.pyfrom openrl.envs.common import makefrom openrl.modules.common import PPONet as Netfrom openrl.runners.common import PPOAgent as Agentfrom openrl.configs.config import create_config_parserfrom openrl.modules.networks.policy_value_network_gpt import (PolicyValueNetworkGPT as PolicyValueNetwork,)def train ():# 添加读取配置文件的代码cfg_parser = create_config_parser ()cfg = cfg_parser.parse_args ()# 创建 NLP 环境env = make ("daily_dialog",env_num=2,asynchrnotallow=True,cfg=cfg,)# 创建自定义神经网络model_dict = {"model": PolicyValueNetwork}net = Net (env, cfg=cfg, model_dict=model_dict)# 创建训练智能体agent = Agent (net)agent.train (total_time_steps=5000000)if __name__ == "__main__":train ()</code>

<code>model_dict = {"policy": CustomPolicyNetwork,"critic": CustomValueNetwork,}net = Net (env, model_dict=model_dict)</code>

<code># nlp_ppo.yamlreward_class:id: NLPReward # 奖励模型名称args: {# 用于意图判断的模型的名称或路径"intent_model": rajkumarrrk/roberta-daily-dialog-intent-classifier,# 用于计算 KL 散度的预训练模型的名称或路径"ref_model": roberta-base, # 用于意图判断的 tokenizer 的名称或路径}</code>

<code># train_ppo.pyfrom openrl.rewards.nlp_reward import CustomRewardfrom openrl.rewards import RewardFactoryRewardFactory.register ("CustomReward", CustomReward)</code>

<code>reward_class:id: "CustomReward" # 自定义奖励模型名称args: {} # 用户自定义奖励函数可能用到的参数</code>

<code># nlp_ppo.yamlvec_info_class:id: "NLPVecInfo" # 调用 NLPVecInfo 类以打印 NLP 任务中奖励函数的信息# 设置 wandb 信息wandb_entity: openrl # 这里用于指定 wandb 团队名称，请把 openrl 替换为你自己的团队名称experiment_name: train_nlp # 这里用于指定实验名称run_dir: ./run_results/ # 这里用于指定实验数据保存的路径log_interval: 1 # 这里用于指定每隔多少个 episode 上传一次 wandb 数据# 自行填写其他参数...</code>

<code># train_ppo.pyagent.train (total_time_steps=100000, use_wandb=True)</code>

<code># train_ppo.py # 注册自定义输出信息类 VecInfoFactory.register ("CustomVecInfo", CustomVecInfo)</code>

<code># nlp_ppo.yamlvec_info_class:id: "CustomVecInfo" # 调用自定义 CustomVecInfo 类以输出自定义信息</code>

<code># nlp_ppo.yamluse_amp: true # 开启混合精度训练</code>

<code># chat.pyfrom openrl.runners.common import ChatAgent as Agentdef chat ():agent = Agent.load ("./ppo_agent", tokenizer="gpt2",)history = []print ("Welcome to OpenRL!")while True:input_text = input ("> User:")if input_text == "quit":breakelif input_text == "reset":history = []print ("Welcome to OpenRL!")continueresponse = agent.chat (input_text, history)print (f"> OpenRL Agent: {response}")history.append (input_text)history.append (response)if __name__ == "__main__":chat ()</code>