40 lines
754 B
Python
40 lines
754 B
Python
# %%
|
|
from stable_baselines3 import PPO
|
|
from intersim.envs.intersimple import NRasterizedRandomAgent, speed_reward
|
|
import functools
|
|
|
|
model_name = "ppo_speed_image_random"
|
|
|
|
env = NRasterizedRandomAgent(
|
|
reward=functools.partial(
|
|
speed_reward,
|
|
collision_penalty=0
|
|
)
|
|
)
|
|
|
|
# %%
|
|
model = PPO(
|
|
"CnnPolicy", env,
|
|
verbose=1,
|
|
batch_size=2048,
|
|
)
|
|
model.learn(total_timesteps=2e5)
|
|
model.save(model_name)
|
|
|
|
print('Done training.')
|
|
|
|
del model # remove to demonstrate saving and loading
|
|
|
|
# %%
|
|
model = PPO.load(model_name)
|
|
|
|
obs = env.reset()
|
|
while True:
|
|
action, _states = model.predict(obs)
|
|
obs, rewards, done, info = env.step(action)
|
|
env.render(mode='post')
|
|
if done:
|
|
break
|
|
|
|
env.close(filestr='render/'+model_name)
|