| Param | Value |
|---|---|
| Algorithm | PPO |
| Total timesteps | 10,000,000 |
| Framework | Stable-Baselines3 |
| Environment | SpaceEvaders-v0 |
1from stable_baselines3 import PPO
2from sai_rl import SAIClient
3
4## Initialize the SAI client
5sai = SAIClient("SpaceEvaders-v0")
6
7## Make the environment
8env = sai.make_env()
9
10# Initialize PPO agent
11model = PPO("MlpPolicy", env, verbose=1)
12
13# Train
14model.learn(total_timesteps=10_000_000)
15
16# Save trained model
17sai.save_model("Space Captain", model)