def sort_card(by, reverse=True):
ma = MA()
login = ma.login(config.loginId, config.password)
ma.masterdata_card()
for card in ma.cards.values():
card.hp_power = card.hp + card.power
card.cp = card.hp_power / card.cost
card.lvmax_hp_power = card.lvmax_hp + card.lvmax_power
card.lvmax_cp = card.lvmax_hp_power / card.cost
return sorted(ma.cards.values(), key=lambda x: getattr(x, by),
reverse=reverse)
def all_in_one(invitation_id, name_prefix=None):
_hash = hashlib.md5(invitation_id).hexdigest()
password = _hash[:10]
if not name_prefix:
name_prefix = _hash[-5:]
mobile = tasoo.get_tel()
submit_smscode = smsVerify(mobile)
sms = None
time.sleep(10)
for _ in range(10):
try:
sms = tasoo.get_sms(mobile)
except AssertionError:
time.sleep(2)
continue
if not sms:
raise Exception("can't get sms")
submit_smscode(sms)
ma = MA()
#ma.check_inspection()
#ma.notification_post_devicetoken(mobile, config.password)
ma.regist(mobile, password, invitation_id)
ma.save_character(name_prefix+sms[:4])
ma.tutorial_next(7030)
ma.tutorial_next(8000)
return mobile, password, name_prefix+sms[:4]
def _get_ma_60(self, stock_code, date_obj):
ma = MA().calculate(stock_code, date_obj, 60)
return ma
import sys from ma import MA login_id = sys.argv[1] password = sys.argv[2] #regist = sys.argv[3] ma = MA() ma.login(login_id, password) #ma.check_inspection() #ma.notification_post_devicetoken(login_id, password) #try: #ma.regist(login_id, password, regist) #except: #pass ma.save_character(login_id) ma.tutorial_next(7030) ma.tutorial_next(8000)
(571, 444, 0, lambda: time.sleep(2)), #go
(560, 514, 0, None), # go
]
for i, (x, y, color, text) in enumerate(steps):
print 'step %d' % i
if color:
while color_like(get_color(x, y), color) > 10:
time.sleep(1)
click(x, y)
if text:
if callable(text):
text = text()
if text:
time.sleep(1)
send(text)
time.sleep(1)
# over
while get_color(619, 601) == 0:
click(619, 601)
time.sleep(1)
from ma import MA
print tel, code
ma = MA()
ma.login(tel, PASSWORD)
ma.save_character(tel)
ma.tutorial_next(7030)
ma.tutorial_next(8000)
os.system(r'"C:\Program Files\BlueStacks\HD-Quit.exe"')
(571, 444, 0, lambda: time.sleep(2)), #go
(560, 514, 0, None), # go
]
for i, (x, y, color, text) in enumerate(steps):
print 'step %d' % i
if color:
while color_like(get_color(x, y), color) > 10:
time.sleep(1)
click(x, y)
if text:
if callable(text):
text = text()
if text:
time.sleep(1)
send(text)
time.sleep(1)
# over
while get_color(619, 601) == 0:
click(619, 601)
time.sleep(1)
from ma import MA
print tel, code
ma = MA()
ma.login(tel, PASSWORD)
ma.save_character(tel)
ma.tutorial_next(7030)
ma.tutorial_next(8000)
os.system(r'"C:\Program Files\BlueStacks\HD-Quit.exe"')
# starting training loop
def ppo(env, brain_name, agents):
scores = []
for i in range(4000):
agents.learn(env, brain_name)
current_max_reward = agents.act(env, brain_name)
mean_100 = np.mean(np.array(scores[-100:]))
scores.append(current_max_reward)
if mean_100 > 0.5:
agents.save()
print(
'Environment solved in {} episodes. Last mean 100: {}'.format(
i + 1, mean_100))
break
print(
'Episode: {} Score of the current episode: {} Last 100 average: {}'
.format(i + 1, current_max_reward, mean_100))
return scores
# instantiating multi-agent class
agents = MA(config)
# starting training
all_scores = ppo(env, brain_name, agents)
# visualizing rewards
plt.plot(np.arange(1, len(all_scores) + 1), all_scores)
plt.ylabel('Rewards')
plt.xlabel('Episode #')
plt.show()
import sys from ma import MA login_id = sys.argv[1] password = sys.argv[2] #regist = sys.argv[3] ma = MA() #ma.login(login_id, password) ma.check_inspection() ma.notification_post_devicetoken(login_id, password) ma.regist(login_id, password, regist) ma.save_character(login_id) ma.tutorial_next(7030) ma.tutorial_next(8000)
else:
threshold_fitness = params["threshold_fitness"]
optimizer_options = {
"max_evaluations": params["max_generations"] * n_individuals + 1,
"seed": params["optimizer_seed"] + i, # undefined in the paper
"step_size": 1.0,
"initial_guess": initial_guess,
"threshold_fitness": threshold_fitness,
"len_fitness_data": 0,
"save_step_size_data": True
}
results_file = os.path.join(
data_dir, "{:s}-Dim-{:d}-Trial-{:d}.pickle".format(
params["benchmark_function"], params["ndim_problem"], i + 1))
if not (params["is_plot"]):
solver = MA(problem, optimizer_options)
results = solver.optimize(eval(params["benchmark_function"]))
with open(results_file, "wb") as results_handle:
pickle.dump(results,
results_handle,
protocol=pickle.HIGHEST_PROTOCOL)
else:
with open(results_file, 'rb') as results_handle:
results = pickle.load(results_handle)
if len(results["fitness_data"]) < min_evaluations:
min_evaluations = len(results["fitness_data"])
if len(results["step_size_data"]) < min_step_size_data:
min_step_size_data = len(results["step_size_data"])
if params["is_plot"]:
fitness_data = np.zeros((min_evaluations, 2))
import sys
from ma import MA
login_id = sys.argv[1]
password = sys.argv[2]
#regist = sys.argv[3]
ma = MA()
ma.login(login_id, password)
#ma.check_inspection()
#ma.notification_post_devicetoken(login_id, password)
#try:
#ma.regist(login_id, password, regist)
#except:
#pass
ma.save_character(login_id)
ma.tutorial_next(7030)
ma.tutorial_next(8000)
