with Profile(f'log/{args.log_name}'):
for step in range(5):
action = player.act(bef_o)
aft_o,r,d,i = env.step(action)
player.step(bef_o,action,r,d,i)
if d :
bef_o = env.reset()
else:
bef_o = aft_o
if args.render :
env.render()
remaining_steps = total_steps - hp.Learn_start - 25
for step in range(remaining_steps):
if ((hp.Learn_start + 25 + step) % hp.Model_save) == 0 :
player.save_model()
score = evaluate_f(player, eval_env, vid_type)
print('eval_score:{0}'.format(score))
action = player.act(bef_o)
aft_o,r,d,i = env.step(action)
player.step(bef_o,action,r,d,i)
if d :
bef_o = env.reset()
else:
bef_o = aft_o
if args.render :
env.render()
else :
for step in range(total_steps):
if (step>0) and ((step % hp.Model_save) == 0) :
if not args.vm :
env.render()
for step in range(total_steps):
action = player.act(bef_o)
aft_o,r,d,i = env.step(action)
player.step(bef_o,action,r,d,i)
if d :
bef_o = env.reset()
else:
bef_o = aft_o
if not args.vm :
env.render()
my_tqdm.close()
next_save = player.save_model()
if not args.load:
save_dir = player.save_dir
else:
save_dir, _ = os.path.split(args.load)
next_dir = os.path.join(save_dir,str(next_save))
score = player.evaluate(gym.make(ENVIRONMENT), vid_type)
print('eval_score:{0}'.format(score))
print('{0}steps took {1} sec'.format(total_steps,time.time()-st))
total_loop -= 1
if total_loop <= 0 :
sys.exit()
else :
next_args = []
next_args.append('python')