evaluation_loop.py

from tqdm import tqdm

n_episodes = 100
reward_per_episode = []
success_per_episode = []

for i in tqdm(range(0, n_episodes)):

    state = env.reset()
    total_reward = 0
    done = False
    reward = None
    
    while not done:
        action = agent.act(state)
        next_state, reward, done, info = env.step(action)
        total_reward += reward
        state = next_state
        
    reward_per_episode.append(total_reward)
    success_per_episode.append(1 if reward > 0 else 0)