main.py

import numpy as np
import random as rnd
import time 
import matplotlib.pyplot as plt
import math
import gym
import pygame

import torch
import tensorflow as tf

# all of the libraries above can be installed with pip
# ex: pip install numpy or pip install torch


#from DQN import DQNAgent



# Hyperparams
input_dims = 4
output_dims = 2
# likely want to put in some other cool things here like batch size, learning rate, etc. 
episodes = 0

# Global Constants, change these
MAX_EPISODES = 1


if __name__ == "__main__":
    env = gym.make('CartPole-v1', render_mode='human')
    # agnet = DQNAgent(input_dims, output_dims)

    # Make the main game loop.  

    while episodes < MAX_EPISODES:
        time_step = 0
        rewards = []
        #agent.replay_memory.erase_memory()
        observation, info = env.reset()
        time_step = 0
        done = False

        while not done:
            
            
            # Get action, ideally through your agent
            action = env.action_space.sample()
            
            # Take the action and observe the result
            observation, reward, terminated, trunicated, info = env.step(action)
            
            # Accumulate the reward

            # Check if we lost
            if terminated or trunicated:
                done = True


            # Store our memory

            # learn?
            #agent.learn()
            time_step += 1

            env.render()
        
    # TODO: Check if reward normalization makes sense!
    # agent.save()
    env.close()