REINFORCE-CartPole-v1-updated

REINFORCE Agent for CartPole-v1

This is a REINFORCE agent trained on the CartPole-v1 environment.

Training Hyperparameters

{'h_size': 16, 'lr': 0.01, 'gamma': 0.99, 'n_training_episodes': 1000, 'max_t': 1000, 'print_every': 100}

Evaluation Results

• Mean Reward: 362.10
• Standard Deviation: 179.98

How to use

import gymnasium as gym
import torch
from your_policy_file import Policy # Assuming Policy class is in 'your_policy_file.py'

# Load the environment
env = gym.make("CartPole-v1")

# Instantiate the policy network (adjust sizes as per your model)
s_size = env.observation_space.shape[0]
a_size = env.action_space.n
h_size = 16
policy = Policy(s_size, a_size, h_size)

# Load the trained weights
policy.load_state_dict(torch.load("model.pt"))
policy.eval()

# Test the agent
state, info = env.reset()
total_reward = 0
terminated = False
truncated = False
while not terminated and not truncated:
    action, _ = policy.act(state)
    state, reward, terminated, truncated, info = env.step(action)
    total_reward += reward
print(f"Test Episode Reward: {total_reward}")
env.close()