import numpy as np

ENV_NAME = "MountainCarContinuous-v0"
NUM_ENVS = 15             # 15 asynchronous environments running in parallel
ROLLOUT_STEPS = 200        # Steps collected per environment per iteration
TOTAL_ITERATIONS = 1500     # Replacing episode loops with iteration cycles. which is 4.5million timesteps
GAMMA = 0.99
CLIP_RATIO = 0.2
POLICY_LR = 3e-4
VALUE_LR = 1e-3
TRAIN_EPOCHS = 5
BATCH_SIZE = 128          # Increased to handle the larger batch sizes (15 * 200 = 3000 steps)

INITIAL_LOG_STD = -0.5 * np.ones(1, dtype=np.float32)

# Directory Management
LOG_DIR = "./ContinuousMountainCar_logs/"
CHECKPOINT_DIR = "./ContinuousMountainCar/"