Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fixed runtime issues of RLLib #209

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fixed runtime issues of RLLib #209

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions experiments/ppo_4x4grid.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@


if __name__ == "__main__":
print(os.getcwd())
ray.init()

env_name = "4x4grid"
Expand Down
30 changes: 6 additions & 24 deletions experiments/sb3_grid4x4.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,6 @@
import numpy as np
import supersuit as ss
import traci
from pyvirtualdisplay.smartdisplay import SmartDisplay
from stable_baselines3 import PPO
from stable_baselines3.common.callbacks import EvalCallback
from stable_baselines3.common.evaluation import evaluate_policy
Expand All @@ -16,34 +15,26 @@


if __name__ == "__main__":
RESOLUTION = (3200, 1800)

env = sumo_rl.grid4x4(use_gui=True, out_csv_name="outputs/grid4x4/ppo_test", virtual_display=RESOLUTION)
env = sumo_rl.grid4x4(use_gui=False, out_csv_name="outputs/grid4x4/ppo_train")

max_time = env.unwrapped.env.sim_max_time
delta_time = env.unwrapped.env.delta_time

print("Environment created")

env = ss.pettingzoo_env_to_vec_env_v1(env)
env = ss.concat_vec_envs_v1(env, 2, num_cpus=1, base_class="stable_baselines3")
env = ss.concat_vec_envs_v1(env, 2, num_cpus=16, base_class="stable_baselines3")
env = VecMonitor(env)

model = PPO(
"MlpPolicy",
env,
verbose=3,
gamma=0.95,
n_steps=256,
ent_coef=0.0905168,
learning_rate=0.00062211,
vf_coef=0.042202,
max_grad_norm=0.9,
gae_lambda=0.99,
n_epochs=5,
clip_range=0.3,
batch_size=256,
tensorboard_log="./logs/grid4x4/ppo_test",
tensorboard_log="./logs/grid4x4/ppo_train",
)

print("Starting training")
Expand All @@ -55,28 +46,19 @@
print(mean_reward)
print(std_reward)

model.save('ppo_output')

# Maximum number of steps before reset, +1 because I'm scared of OBOE
print("Starting rendering")
num_steps = (max_time // delta_time) + 1

obs = env.reset()

if os.path.exists("temp"):
shutil.rmtree("temp")

os.mkdir("temp")
# img = disp.grab()
# img.save(f"temp/img0.jpg")

img = env.render()
for t in trange(num_steps):
actions, _ = model.predict(obs, state=None, deterministic=False)
obs, reward, done, info = env.step(actions)
img = env.render()
img.save(f"temp/img{t}.jpg")

subprocess.run(["ffmpeg", "-y", "-framerate", "5", "-i", "temp/img%d.jpg", "output.mp4"])
env.render()

print("All done, cleaning up")
shutil.rmtree("temp")
env.close()
Loading