state = env.reset()
done = False
score = 0
while not done:
env.render()
action = agent.get_action(state)
new_state, reward, done, _ = env.step(action)
agent.reward_history.append(reward)
score += reward
state = new_state
if not test_mode:
agent.learn()
scores.append(score)
if count % 100 == 0:
avg_score = np.mean(scores[-100:])
if score > best_score:
best_score = score
if not test_mode:
agent.save_model()
print(
f"Episode: {count+1}, score: {score}, current average score: {avg_score}"
)
env.close()
x = range(1, num_games + 1)
plt.plot(x, scores)
plt.show()
