eval.py

from agent import *
from env import *

def rollout(agents: list[Agent], board: StrandsBoard,display_s:float = -1) -> int:
    """
    Returns +1 for a win for white, -1 for a win for black, 0 for a draw
    """
    board.reset()
    board.make_first_random_action()
    board.draw(display_s)

    with torch.no_grad():
        while not board.check_for_termination():
            i = board.round_idx%2 # 0 for "WHITE to play", 1 for "BLACK to play"
            agents[i].act_greedily(board)
            board.draw(display_s)

    reward = board.compute_reward()

    return(reward)

def evaluate(agents: list[Agent], board: StrandsBoard, n_rollouts: int) -> float:

    rewards_agent_WHITE = []
    
    for i in range(n_rollouts):
        
        reward_agent_WHITE = rollout(agents, board)
        rewards_agent_WHITE.append(reward_agent_WHITE)

    return(np.mean(rewards_agent_WHITE), -np.mean(rewards_agent_WHITE))