brandinho/space-evaders

Link to Environment: SpaceEvaders-v0

Param	Value
Algorithm	PPO
Total timesteps	10,000,000
Framework	Stable-Baselines3
Environment	SpaceEvaders-v0

Reproduce Results

To train a similar model, you can run the follow code:

from stable_baselines3 import PPO
from sai_rl import SAIClient

## Initialize the SAI client
sai = SAIClient("SpaceEvaders-v0")

## Make the environment
env = sai.make_env()

# Initialize PPO agent
model = PPO("MlpPolicy", env, verbose=1)

# Train
model.learn(total_timesteps=10_000_000)

# Save trained model
sai.save_model("Space Captain", model)