arzhela/QuadOpt-RL-ppo-sb3

PPO Agent for QuadMesh-v0

This model is a PPO (Proximal Policy Optimization) agent trained with Stable-Baselines3 on the custom QuadMesh-v0 environment.

Description

The agent was trained to optimize quadrilateral meshes using reinforcement learning.

Usage

from huggingface_hub import hf_hub_download
from stable_baselines3 import PPO

# Download and load the model
model_path = hf_hub_download(
    repo_id="arzhela/QuadOpt-RL-ppo-sb3",
    filename="test_model.zip",
)
model = PPO.load(model_path)

# Use the model (requires the QuadMesh-v0 environment)
# obs, info = env.reset()
# action, _states = model.predict(obs, deterministic=True)

Training Details

• Algorithm: PPO
• Library: Stable-Baselines3
• Environment: QuadMesh-v0 (custom Gymnasium environment)

Hyperparameters