# env.render()
# 4. Run agent on the state
action = mario.act(state)
# 5. Agent performs action
next_state, reward, done, info = env.step(action)
# 6. Remember
mario.cache(state, next_state, action, reward, done)
# 7. Learn
q, loss = mario.learn()
# 8. Logging
logger.log_step(reward, loss, q)
# 9. Update state
state = next_state
# 10. Check if end of game
if done or info['flag_get']:
break
logger.log_episode()
if e % print_e == 0:
logger.record(episode=e,
epsilon=mario.exploration_rate,
step=mario.curr_step)
for e in range(episodes):
state = env.reset()
while True:
env.render()
action = mario.act(state)
next_state, reward, done, info = env.step(action)
mario.cache(state, next_state, action, reward, done)
logger.log_step(reward, None, None)
state = next_state
if done or info['flag_get']:
break
logger.log_episode()
if e % 20 == 0:
logger.record(
episode=e,
epsilon=mario.exploration_rate,
step=mario.curr_step
)