stock_x = pandas.get_stock_hist_data_cache(code, date)
if stock_x is None or stock_x.empty:
return pd.Series([code, date, 0.0, 0.0], index=stock_name_list)
stock_y = pd.Series(stock_x["close"].values)
stock_x_next = stock_x.iloc[len(stock_x) - 1]
stock_x = stock_x.drop(stock_x.index[len(stock_x) - 1])
stock_y = stock_y.drop(stock_y.index[0])
stock_x.drop('close', axis=1, inplace=True)
stock_x_next.drop(labels=['close'], inplace=True)
model = linear_model.LinearRegression()
model.fit(stock_x.values, stock_y)
next_close = model.predict([stock_x_next.values])
if len(next_close) == 1:
next_close = next_close[0]
sklearn_score = model.score(stock_x.values, stock_y)
return pd.Series([code, date, next_close, sklearn_score * 100],
index=stock_name_list)
calculate_stock_sklearn_index = CalculateStockSklearnIndex()
if __name__ == '__main__':
tmp_datetime = executor.run_with_args(calculate_stock_sklearn_index.run)
from model.ts_today_all import ts_today_all
class CalculatePeriodWaveIndex:
@staticmethod
def run(date):
data = ts_today_all.select(date)
statistics = data.apply(lambda row: stockstats.calculate_wave(
code=row['code'], date=row['date'], max_point=5),
axis=1)
statistics.drop('date', axis=1, inplace=True)
statistics = pd.merge(data, statistics, on=['code'], how='left')
statistics["up_rate"] = stockstats.calculate_up_rate(
trade=statistics["trade"],
wave_mean=statistics["wave_mean"],
wave_crest=statistics["wave_crest"])
period_wave_index.delete(date)
period_wave_index.insert(statistics)
calculate_period_wave_index = CalculatePeriodWaveIndex()
# main函数入口
if __name__ == '__main__':
# 使用方法传递。
tmp_datetime = executor.run_with_args(calculate_period_wave_index.run)
appendix={'date': date.strftime("%Y%m%d")})
class DownloadIndex:
@staticmethod
def run(date):
tushare.download_data(svc_name='get_index',
primary_keys=["date", "code"],
appendix={'date': date.strftime("%Y%m%d")})
class DownloadTodayAll:
@staticmethod
def run(date):
tushare.download_data(svc_name='get_today_all',
primary_keys=["date", "code"],
appendix={'date': date.strftime("%Y%m%d")})
del_his_cache_data = DelHisCacheData()
download_top_list = DownloadTopList()
download_index = DownloadIndex()
download_today_all = DownloadTodayAll()
if __name__ == '__main__':
executor.run_with_args(del_his_cache_data.run)
executor.run_with_args(download_top_list.run)
executor.run_with_args(download_index.run)
time.sleep(5) # 停止5秒
executor.run_with_args(download_today_all.run)
statistics = data.apply(lambda row: stockstats.calculate_wave(
code=row['code'],
date=row['date'],
max_point=3,
),
axis=1)
statistics.drop('date', axis=1, inplace=True)
statistics = pd.merge(data, statistics, on=['code'], how='left')
statistics = statistics[statistics["trade"] > statistics["wave_base"]]
statistics = statistics[statistics["trade"] < statistics["wave_crest"]]
statistics["up_rate"] = stockstats.calculate_up_rate(
wave_mean=statistics["wave_mean"],
trade=statistics["trade"],
wave_crest=statistics["wave_crest"])
statistics["buy"] = 1
statistics["sell"] = 0
statistics["today_trade"] = statistics["trade"]
statistics["income"] = 0
statistics = statistics.rename(columns={'date': 'buy_date'})
stock_statistics_buy.insert(statistics)
calculate_stock_statistics_buy = CalculateStockStatisticsBuy()
if __name__ == '__main__':
executor.run_with_args(calculate_stock_statistics_buy.run)
date=row['date'],
),
axis=1
)
statistics.drop('date', axis=1, inplace=True)
return pd.merge(data, statistics, on=['code'], how='left')
class StockStatisticsFilter:
@staticmethod
def filter(date):
stock_statistics.delete(date)
data = ss_stock_statistics.select(
date=date,
min_kdjk=80,
min_kdjd=70,
min_kdjj=90,
min_rsi_6=50,
min_cci=100
)
stock_statistics.insert(data)
stock_statistics_filter = StockStatisticsFilter()
stock_stats_index_calculator = StockStatsIndexCalculator()
if __name__ == '__main__':
executor.run_with_args(stock_stats_index_calculator.run)
executor.run_with_args(stock_statistics_filter.filter)
download_deposit_rate_job = DownloadDepositRateJob()
download_loan_rate_job = DownloadLoanRateJob()
download_rrr_job = DownloadRrrJob()
download_money_supply_job = DownloadMoneySupplyJob()
download_money_supply_bal_job = DownloadMoneySupplyBalJob()
download_gdp_year_job = DownloadGdpYearJob()
download_gdp_quarter_job = DownloadGdpQuarterJob()
download_gdp_for_job = DownloadGdpForJob()
download_gdp_pull_job = DownloadGdpPullJob()
download_gdp_contrib_job = DownloadGdpContribJob()
download_cpi_job = DownloadCpiJob()
download_ppi_job = DownloadPpiJob()
download_stock_basics_job = DownloadStockBasicsJob()
if __name__ == '__main__':
mysql.create_new_schema_if_necessary()
executor.run_with_args(download_deposit_rate_job.run)
executor.run_with_args(download_loan_rate_job.run)
executor.run_with_args(download_rrr_job.run)
executor.run_with_args(download_money_supply_job.run)
executor.run_with_args(download_money_supply_bal_job.run)
executor.run_with_args(download_gdp_year_job.run)
executor.run_with_args(download_gdp_quarter_job.run)
executor.run_with_args(download_gdp_for_job.run)
executor.run_with_args(download_gdp_pull_job.run)
executor.run_with_args(download_gdp_contrib_job.run)
executor.run_with_args(download_cpi_job.run)
executor.run_with_args(download_ppi_job.run)
executor.run_with_args(download_stock_basics_job.run)
stock_data_list = [code, date]
stock = pandas.get_stock_hist_data_cache(code, date)
stock_data_list.append(
numpy.get_valid_ele(stock.tail(1)['close'].to_list(), 0))
stock_stats = stockstats.StockDataFrame.retype(stock)
stock_data_list.append(
numpy.get_valid_ele(stock_stats["cci"].tail(1).to_list(), 0))
stock_data_list.append(
numpy.get_valid_ele(stock_stats["kdjj"].tail(1).to_list(), 0))
stock_data_list.append(
numpy.get_valid_ele(stock_stats["rsi_6"].tail(1).to_list(), 0))
sell_indicator = self.get_sell_indicator(stock_data_list[3],
stock_data_list[4],
stock_data_list[5])
stock_data_list.append(sell_indicator)
stock_data_list.append(1 - sell_indicator)
return pd.Series(stock_data_list, index=stock_name_list)
@staticmethod
def get_sell_indicator(kdjj, rsi_6, cci):
return 1 if kdjj > 80 and rsi_6 > 55 or cci > 100 else 0
calculate_stock_statistics_sell = CalculateStockStatisticsSell()
if __name__ == '__main__':
tmp_datetime = executor.run_with_args(calculate_stock_statistics_sell.run)