Feat: Integration-sb3

[Zeyi-Lin]

Description

需要安装的环境：

pip install gymnasium stable_baselines3

测试代码：

import swanlab
import gymnasium as gym
from stable_baselines3 import PPO
from stable_baselines3.common.monitor import Monitor
from stable_baselines3.common.vec_env import DummyVecEnv
from swanlab.integration.sb3 import SwanLabCallback

config = {
    "policy_type": "MlpPolicy",
    "total_timesteps": 25000,
    "env_name": "CartPole-v1",
}


def make_env():
    env = gym.make(config["env_name"], render_mode="rgb_array")
    env = Monitor(env)
    return env


env = DummyVecEnv([make_env])
model = PPO(
    config["policy_type"],
    env,
    verbose=1,
)

model.learn(
    total_timesteps=config["total_timesteps"],
    callback=SwanLabCallback(
        project="PPO",
        experiment_name="MlpPolicy",
        verbose=2,
    ),
)

swanlab.finish()

Closes: #563

[SAKURA-CAT]

应该把一些集成脚本写入test文件夹，并附上必要的安装脚本，方便复现。