def draw_predictions(self,
corp_name,
scaler_close,
data_params,
test_predict,
invest_predicts=None):
"""예측 그래프를 그린다."""
dir = 'predicts'
dir_chart = os.path.join(self.main.DIR_CHARTS, dir)
DataUtils.create_dir(dir_chart)
if invest_predicts is not None and test_predict is not None:
predicts = np.append(test_predict, invest_predicts)
#print(len(test_predict), len(invest_predicts))
elif test_predict is not None:
predicts = test_predict
else:
predicts = invest_predicts
dataY = np.append(data_params.testY, data_params.investY)
preds = []
for pred in predicts:
#print(pred, scaler_close)
preds.append(DataUtils.inverse_scaled_data(scaler_close, pred))
close_values = []
for y in dataY:
close_values.append(DataUtils.inverse_scaled_data(scaler_close, y))
df_data = self.get_predicts_data(data_params, close_values, preds)
df_data['date'] = pd.to_datetime(df_data['date'], format='%Y.%m.%d')
df_data_index = df_data.set_index('date')
self.draw_predictions_seaborn(df_data_index, dir, corp_name)
self.save_csv(df_data, dir, corp_name)
def invest_scaled_money(self, invest_scaled_predict, now_scaled_close,
scaler_close, now_money, now_stock_cnt):
"""예측 값에 따라 매수 매도를 실행한다."""
now_close = DataUtils.inverse_scaled_data(scaler_close,
now_scaled_close)
if invest_scaled_predict == -1:
invest_predict = -1
else:
invest_predict = DataUtils.inverse_scaled_data(
scaler_close, invest_scaled_predict)
return self.invest_money(invest_predict, now_close, now_money,
now_stock_cnt)
def invest_scaled_money_before(self, before_scaled_close,
before_scaled_predict, x, now_money,
now_stock_cnt, scaler_close):
x_high, x_low = self.get_high_n_low(x, scaler_close)
before_invest_predict = DataUtils.inverse_scaled_data(
scaler_close, before_scaled_predict)
before_close = DataUtils.inverse_scaled_data(scaler_close,
before_scaled_close)
return self.invest_money_before2(before_close, before_invest_predict,
x_high, x_low, now_money,
now_stock_cnt)
def invest_n_all(self,
comp_code,
dataX_last,
data_params,
params_all,
scaler_close=None):
"""학습 후 모의 주식 거래를 한다."""
ip = InvestParams()
investX = data_params.investX
invest_count = len(investX)
ip.invest_money = self.params.invest_money
now_stock_cnt = 0
ip.index_money = ip.invest_money
all_stock_count = 0
before_scaled_close = None
before_invest_predict = None
now_scaled_close = None
for i in range(invest_count):
x = investX[i:i + 1]
invest_predict = self._get_predict(self.params, comp_code, x)
invest_predict_all = self._get_predict(params_all, comp_code, x)
invest_predict = np.mean([invest_predict, invest_predict_all])
ip.predict_list.append([invest_predict])
x_last = x[:, -1][0]
now_scaled_close = x_last[0]
if i != 0:
ip.invest_money, now_stock_cnt = self.invest_scaled_money_before(
before_scaled_close, before_invest_predict, x,
ip.invest_money, now_stock_cnt, scaler_close)
ip.invest_money, now_stock_cnt = self.invest_scaled_money(
invest_predict, now_scaled_close, scaler_close,
ip.invest_money, now_stock_cnt)
if i == 0:
ip.index_money, all_stock_count = self.invest_scaled_money(
-1, now_scaled_close, scaler_close, ip.index_money,
all_stock_count)
before_scaled_close = now_scaled_close
before_invest_predict = invest_predict
if now_scaled_close != None:
now_close = DataUtils.inverse_scaled_data(scaler_close,
now_scaled_close)
ip.invest_money += self.to_money(now_stock_cnt, now_close)
ip.index_money += self.to_money(all_stock_count, now_close)
# 마지막 예측 값을 구한다.
ip.last_predict = self._get_predict(self.params, comp_code, dataX_last)
last_predict_all = self._get_predict(params_all, comp_code, dataX_last)
ip.last_predict = np.mean([ip.last_predict, last_predict_all])
return ip
def draw_invests(self, corp_name, invest_data, all_invest_data, scaler_close, data_params)->list:
"""모의 투자 결과 그래프를 그린다."""
dir = 'invests'
kospi = self.get_kospi_kosdaq('KOSPI')
kosdaq = self.get_kospi_kosdaq('KOSDAQ')
start_money = self.params.invest_money
if self.params.index_money is not None:
start_index_money = self.params.index_money
else:
start_index_money = start_money
start_kospi = self.get_value(kospi, 0, 'close')
start_kosdaq = self.get_value(kosdaq, 0, 'close')
investY_date = data_params.investY_date
invest_index = all_invest_data[0]
money_index = invest_index[1]
stock_cnt_index = invest_index[2]
date = investY_date[0:1].values[0]
date = DateUtils.add_days(date, -1, '%Y.%m.%d')
invest_chart_data = [[date, 0, 0, 0, 0]]
invest_daily = [[date, start_money, start_index_money, start_kospi, start_kosdaq]]
for i in range(len(invest_data)):
try:
date = investY_date[i:i+1].values[0]
invest = invest_data[i]
close_scaled = invest[0]
money = invest[1]
stock_cnt = invest[2]
kospi_amt = self.get_date_search_value(kospi, date, 'close')
if kospi_amt == 0: continue
kospi_percent = self.get_ratio(kospi_amt,start_kospi)
kosdaq_amt = self.get_date_search_value(kosdaq, date, 'close')
if kosdaq_amt == 0: continue
kosdaq_percent = self.get_ratio(kosdaq_amt, start_kosdaq)
close = DataUtils.inverse_scaled_data(scaler_close, close_scaled)
eval_amt = money + close * stock_cnt
eval_percent = self.get_ratio (eval_amt, start_money)
eval_index_amt = money_index + close * stock_cnt_index
eval_index_percent = self.get_ratio (eval_index_amt, start_money)
invest_chart_data.append([date, eval_percent, eval_index_percent, kospi_percent, kosdaq_percent])
invest_daily.append([date, eval_amt, eval_index_amt, kospi_amt, kosdaq_amt])
except Exception:
pass
#exc_type, exc_value, exc_traceback = sys.exc_info()
#traceback.print_exception(exc_type, exc_value, exc_traceback, file=sys.stdout)
#print(invest_chart_data)
if self.params.debug == True:
df_data = pd.DataFrame(invest_chart_data, columns=['date', 'invest', 'index', 'KOSPI', 'KOSDAQ'])
self.draw_invest_seaborn(df_data, dir, corp_name)
self.save_csv(df_data, dir, corp_name)
return invest_daily
def get_real_money(self, data_params, scaler_close, last_predict):
"""실제 가격을 가져온다."""
invest_max_percent = self.params.invest_max_percent
trade_min_percent = self.params.trade_min_percent
close_scaled_money = data_params.testY[-1][0]
close_money = DataUtils.inverse_scaled_data(scaler_close,
close_scaled_money)
predict_money = DataUtils.inverse_scaled_data(scaler_close,
last_predict)
if predict_money > close_money:
trade_min_money = close_money * (1 + trade_min_percent / 100)
trade_money = close_money + (predict_money - close_money) * (
invest_max_percent / 100)
if trade_money < trade_min_money:
trade_money = trade_min_money
else:
trade_max_money = close_money * (1 - trade_min_percent / 100)
trade_money = close_money - (close_money - predict_money) * (
invest_max_percent / 100)
if trade_money > trade_max_money:
trade_money = trade_max_money
return close_money, predict_money, trade_money
def get_high_n_low(self, x, scaler_close):
x_last = x[:, -1][0]
x_high = DataUtils.inverse_scaled_data(scaler_close, x_last[2])
x_low = DataUtils.inverse_scaled_data(scaler_close, x_last[3])
return x_high, x_low
def invest(self,
comp_code: str,
corp_name: str,
tps: dict,
invest_only: bool = False) -> InvestParams:
"""학습 후 모의 주식 거래를 한다."""
ip = InvestParams()
ip.predict_list, ip.last_predict, ip.test_predict = self.ensemble_predicts(
comp_code, tps, invest_only)
#print(ip.predict_list, ip.last_predict, ip.test_predict)
tp = next(iter(tps.values()))
investX = tp.data_params.investX
#investY = data_params.investY
invest_count = len(investX)
ip.invest_money = self.params.invest_money
#print('mockInvestment', 'invest', ip.invest_money)
invest_line_trading = self.params.invest_line_trading
now_stock_cnt = 0
if self.params.index_money is not None:
ip.index_money = self.params.index_money
else:
ip.index_money = ip.invest_money
all_stock_count = now_stock_cnt
invest_list = []
index_invest_list = []
before_scaled_close = None
before_invest_predict = None
now_scaled_close = None
for i in range(invest_count):
x = investX[i:i + 1]
invest_predict = ip.predict_list[i][0]
x_last = x[:, -1][0]
now_scaled_close = x_last[0]
if i != 0 and invest_line_trading == True:
ip.invest_money, now_stock_cnt = self.invest_scaled_money_before(
before_scaled_close, before_invest_predict, x,
ip.invest_money, now_stock_cnt, tp.scaler_close)
ip.invest_money, now_stock_cnt = self.invest_scaled_money(
invest_predict, now_scaled_close, tp.scaler_close,
ip.invest_money, now_stock_cnt)
if i == 0:
ip.index_money, all_stock_count = self.invest_scaled_money(
-1, now_scaled_close, tp.scaler_close, ip.index_money,
all_stock_count)
index_invest_list.append(
[now_scaled_close, ip.index_money, all_stock_count])
#print(ip.index_money, all_stock_count, scaler_close.inverse_transform(now_scaled_close)[0][0])
before_scaled_close = now_scaled_close
before_invest_predict = invest_predict
invest_list.append(
[now_scaled_close, ip.invest_money, now_stock_cnt])
if now_scaled_close is not None:
now_close = DataUtils.inverse_scaled_data(tp.scaler_close,
now_scaled_close)
ip.invest_money += self.to_money(now_stock_cnt, now_close)
ip.last_money = ip.invest_money
ip.index_money += self.to_money(all_stock_count, now_close)
invest_list.append([now_scaled_close, ip.invest_money, 0])
index_invest_list.append([now_scaled_close, ip.index_money, 0])
try:
visualizer = InvestVisualizer(self.params)
ip.daily_data = visualizer.draw_invests(
corp_name, invest_list, index_invest_list, tp.scaler_close,
tp.data_params)
except Exception:
exc_type, exc_value, exc_traceback = sys.exc_info()
traceback.print_exception(exc_type,
exc_value,
exc_traceback,
file=sys.stdout)
else:
ip.last_money = ip.invest_money
return ip