main.py

from environment import Environment
from agent import Agent
from collections import deque
import torch
import numpy as np

def main() -> None:
    """
    Main function to train an RL agent using Deep Q-Learning.

    This function initializes the environment and the agent, performs training episodes,
    and saves the trained agent's model if the environment is solved.

    Returns:
        None
    """

    env   : Environment = Environment()
    agent : Agent       = Agent(action_size=env.number_actions)
    number_episodes               : int   = 2000
    maximum_timesteps_per_episode : int   = 10000
    epsilon_starting_value        : float = 1.0
    episilon_ending_value         : float = 0.01
    episilon_decay_value          : float = 0.995
    epsilon                       : float = epsilon_starting_value
    scores_on_100_episodes        : deque = deque(maxlen=100)
    for episodes in range(1, number_episodes + 1):
        state, _ = env.reset()  # type: ignore
        score : float = 0
        for timesteps in range(0, maximum_timesteps_per_episode):
            action : int = agent.action(state, epsilon)
            next_state, reward, done, _, _ = env.step(action)
            agent.step(state, action, reward, next_state, done)  # type: ignore
            state = next_state
            score += reward  # type: ignore
            if done:
                break
        scores_on_100_episodes.append(score)
        epsilon = max(episilon_ending_value, episilon_decay_value * epsilon)
        print(f'\rEpisode: {episodes}\tAverage Score: {np.mean(scores_on_100_episodes):.2f}', end='')
        if episodes % 100 == 0:
            print(f'\rEpisode: {episodes}\tAverage Score: {np.mean(scores_on_100_episodes):.2f}')
        if np.mean(scores_on_100_episodes) >= 500.0:  # type: ignore
            print(f'\nEnvironment Solved in {episodes:d} episodes!\tAverage Score: {np.mean(scores_on_100_episodes):.2f}')
            torch.save(agent.local_qnet.state_dict(), 'checkpoint.pth')
            break

if __name__ == "__main__":
    main()