def __log_result_csv(self, index, time):
from pgportfolio.trade import backtest
dataframe = None
csv_dir = './train_package/train_summary.csv'
json_dir = "./train_package/" + str(
int(index)) + "/weight_history" + str(int(index)) + ".json"
tflearn.is_training(False, self._agent.session)
#evaluate on the test set
v_pv, v_log_mean, benefit_array, v_log_mean_free =\
self._evaluate("test",
self._agent.portfolio_value,
self._agent.log_mean,
self._agent.pv_vector,
self._agent.log_mean_free)
#initialize backtest
backtest = backtest.BackTest(self.config.copy(),
net_dir=None,
agent=self._agent)
#method of trader, initialized by BackTest
backtest.start_trading()
#collects the result in a dic
result = Result(
test_pv=[v_pv],
test_log_mean=[v_log_mean],
test_log_mean_free=[v_log_mean_free],
test_history=[''.join(str(e) + ', ' for e in benefit_array)],
config=[json.dumps(self.config)],
net_dir=[index],
backtest_test_pv=[backtest.test_pv],
backtest_test_history=[
''.join(str(e) + ', ' for e in backtest.test_pc_vector)
],
backtest_test_log_mean=[np.mean(np.log(backtest.test_pc_vector))],
training_time=int(time))
new_data_frame = pd.DataFrame(result._asdict()).set_index("net_dir")
historyweights = []
for weight in backtest.test_updated_omega:
historyweights.append(weight.tolist())
backtesthistory = {"portfolioweights": historyweights}
if os.path.isfile(csv_dir):
dataframe = pd.read_csv(csv_dir).set_index("net_dir")
dataframe = dataframe.append(new_data_frame)
else:
dataframe = new_data_frame
if int(index) > 0:
#writes everything in train_package/train_summary.csv and creates it
dataframe.to_csv(csv_dir)
#writes everything in train_package/int(index)/weight_history int(index).json
with open(json_dir, 'w') as outfile:
json.dump(backtesthistory, outfile)
return result
def __log_result_csv(self, index, time, path):
from pgportfolio.trade import backtest
dataframe = None
csv_dir = './train_package/train_summary.csv'
tflearn.is_training(False, self._agent.session)
v_pv, v_log_mean, benefit_array, v_log_mean_free = self._evaluate("test", self._agent.portfolio_value,\
self._agent.log_mean, self._agent.pv_vector,\
self._agent.log_mean_free)
backtest = backtest.BackTest(self.config.copy(),
self.stockList,
self.featureList,
self.start_date,
self.end_date,
self.fake_data,
net_dir=None,
result_path=path,
agent=self._agent) # 回测
backtest.start_trading()
result = Result(
test_pv=[v_pv],
test_log_mean=[v_log_mean],
test_log_mean_free=[v_log_mean_free],
test_history=[''.join(str(e) + ', ' for e in benefit_array)],
config=[json.dumps(self.config)],
net_dir=[index],
backtest_test_pv=[backtest.test_pv],
backtest_test_history=[
''.join(str(e) + ', ' for e in backtest.test_pc_vector)
],
backtest_test_log_mean=[np.mean(np.log(backtest.test_pc_vector))],
training_time=int(time))
new_data_frame = pd.DataFrame(result._asdict()).set_index("net_dir")
if os.path.isfile(csv_dir):
dataframe = pd.read_csv(csv_dir).set_index("net_dir")
dataframe = dataframe.append(new_data_frame)
else:
dataframe = new_data_frame
if int(index) > 0:
dataframe.to_csv(csv_dir)
return result
def __log_result_csv(self, index, time):
from pgportfolio.trade import backtest
dataframe = None
csv_dir = './train_package/train_summary.csv'
tflearn.is_training(False, self._agent.session)
v_pv, v_log_mean, benefit_array, v_log_mean_free =\
self._evaluate("test",
self._agent.portfolio_value,
self._agent.log_mean,
self._agent.pv_vector,
self._agent.log_mean_free)
backtest = backtest.BackTest(self.config.copy(),
net_dir=None,
agent=self._agent)
backtest.start_trading()
result = Result(test_pv=[v_pv],
test_log_mean=[v_log_mean],
test_log_mean_free=[v_log_mean_free],
test_history=[''.join(str(e)+', ' for e in benefit_array)],
config=[json.dumps(self.config)],
net_dir=[index],
backtest_test_pv=[backtest.test_pv],
backtest_test_history=[''.join(str(e)+', ' for e in backtest.test_pc_vector)],
backtest_test_log_mean=[np.mean(np.log(backtest.test_pc_vector))],
training_time=int(time))
new_data_frame = pd.DataFrame(result._asdict()).set_index("net_dir")
if os.path.isfile(csv_dir):
dataframe = pd.read_csv(csv_dir).set_index("net_dir")
dataframe = dataframe.append(new_data_frame)
else:
dataframe = new_data_frame
if int(index) > 0:
dataframe.to_csv(csv_dir)
return result
def __log_result_csv(self, index, time): # here the index is folder name
from pgportfolio.trade import backtest
dataframe = None
csv_dir = './train_package/train_summary.csv'
csv_weights = './train_package/%s/portfolio_weights.csv' % index
tflearn.is_training(
False, self._agent.session
) # specifying layers defined by testing instead of training
weights = self._evaluate(
"test", self._agent.portfolio_weights) # here (66,339)
v_pv, v_log_mean, benefit_array, v_log_mean_free =\
self._evaluate("test",
self._agent.portfolio_value,
self._agent.log_mean,
self._agent.pv_vector,
self._agent.log_mean_free)
backtest = backtest.BackTest(self.config.copy(),
net_dir=None,
agent=self._agent)
backtest.start_trading() # start showing steps and total assets
# here weights is still (66,339)
result = Result(
test_pv=[v_pv],
test_log_mean=[v_log_mean],
test_log_mean_free=[v_log_mean_free],
test_history=[''.join(str(e) + ', ' for e in benefit_array)],
config=[json.dumps(self.config)],
net_dir=[index],
backtest_test_pv=[backtest.test_pv],
backtest_test_history=[
''.join(str(e) + ', ' for e in backtest.test_pc_vector)
],
backtest_test_log_mean=[np.mean(np.log(backtest.test_pc_vector))],
training_time=int(time))
new_data_frame = pd.DataFrame(result._asdict()).set_index("net_dir")
if os.path.isfile(csv_dir):
dataframe = pd.read_csv(csv_dir).set_index("net_dir")
dataframe = dataframe.append(new_data_frame)
else:
dataframe = new_data_frame
# portfolio weight: weights is a list with one element of np.array(66,339) in coin data [time, assets]
self._asset_list.insert(0, 'RMB')
self._time_index_test = self._time_index[-int(
len(self._time_index) * self._test_portion - self.__window_size):]
weight_df = pd.DataFrame(weights[0],
index=self._time_index_test,
columns=self._asset_list)
if int(index) > 0:
dataframe.to_csv(csv_dir)
weight_df.to_csv(csv_weights)
return result