import os import pickle import gym from gym.wrappers.monitoring.video_recorder import VideoRecorder import torch import torch.nn as nn import torch.optim as optim import gc import optuna gc.enable() # Configuration TRAIN = True FINETUNE = False SAVE_MODEL = True SAVE_VIDEO = False RENDER_MODE = 'rgb_array' # Hyperparameters TRAINING_EPISODES = 1000 FINETUNE_TRIALS = 100 # Default hyperparameters if not fine-tuning GAMMA = 0.99 LEARNING_RATE = 1e-3