聚烯烃日度添加涨跌停价格

config_juxiting.py

@@ -439,6 +439,7 @@ is_update_predict_value = True  # 是否上传预测值到市场信息平台
 is_del_corr = 0.6  # 是否删除相关性高的特征,取值为 0-1 ，0 为不删除，0.6 表示删除相关性小于0.6的特征
 is_del_tow_month = True  # 是否删除两个月不更新的特征
 is_bdwd = False  # 是否使用八大维度
+is_add_zt_price = True  # 是否添加涨跌停价格
 
 
 # 连接到数据库

lib/dataread.py

@@ -67,6 +67,7 @@ global_config = {
     'is_update_eta_data': None,        # ETA数据更新开关
     'early_stop_patience_steps': None,  # 早停步数
     'is_update_report': None,          # 是否更新报告开关
+    'is_add_zt_price': None,           # 是否添加涨跌停价格
 
     # 时间参数
     'start_year': None,                # 起始年份
@@ -516,8 +517,7 @@ def featureAnalysis(df, dataset, y):
     import matplotlib.pyplot as plt
     # 选择特征和标签列
     X = df.drop(['ds', 'y'], axis=1)  # 特征集，排除时间戳和标签列
-    yy = df['y'] # 标签集
-    
+    yy = df['y']  # 标签集
 
     # 标签集自相关函数分析
     from statsmodels.graphics.tsaplots import plot_acf
@@ -1037,6 +1037,9 @@ def datachuli_juxiting(df_zhibiaoshuju, df_zhibiaoliebiao, datecol='date', end_t
     if is_timefurture:
         df = addtimecharacteristics(df=df, dataset=dataset)
 
+    if config.is_add_zt_price:
+        df = addztprice(df=df)
+
     if config.freq == 'WW':
         # 自定义周数据
         # 按weekofmothe分组取均值得到新的数据
@@ -1391,6 +1394,10 @@ class Config:
     @property
     def columnsrename(self): return global_config['columnsrename']
 
+    # 涨跌停价格
+    @property
+    def is_add_zt_price(self): return global_config['is_add_zt_price']
+
 
 config = Config()
 
@@ -2440,11 +2447,13 @@ def get_high_low_data(df):
     df = pd.merge(df, df1, how='left', on='date')
     return df
 
+
 def get_shujuxiang_data(df):
     # 读取excel 从第五行开始
     df1 = pd.read_excel(os.path.join(config.dataset, '数据项下载.xls'), header=5, names=[
                         'numid', 'date', 'MAIN_CONFT_SETTLE_PRICE'])
-    df1['MAIN_CONFT_SETTLE_PRICE'] = df1['MAIN_CONFT_SETTLE_PRICE'].str.replace(',', '').astype(float)
+    df1['MAIN_CONFT_SETTLE_PRICE'] = df1['MAIN_CONFT_SETTLE_PRICE'].str.replace(
+        ',', '').astype(float)
     # 合并数据
     df = pd.merge(df, df1, how='left', on='date')
     return df
@@ -2523,6 +2532,17 @@ def addtimecharacteristics(df, dataset):
     return df
 
 
+def addztprice(df):
+    try:
+        df['ztprice'] = df['y'] * 1.06
+        df['dtprice'] = df['y'] * 0.94
+        df['ztprice'] = df['ztprice'].shift(1)
+        df['dtprice'] = df['dtprice'].shift(1)
+    except:
+        config.logger.info('添加涨跌停价格失败')
+    return df
+
+
 # 从数据库获取百川数据，接收一个百川id列表，返回df格式的数据
 def get_baichuan_data(baichuanidnamedict):
     baichuanidlist = [str(k) for k in baichuanidnamedict.keys()]

main_juxiting.py

@@ -45,6 +45,7 @@ global_config.update({
     'is_fivemodels': is_fivemodels,
     'is_update_predict_value': is_update_predict_value,
     'early_stop_patience_steps': early_stop_patience_steps,
+    'is_add_zt_price': is_add_zt_price,
 
     # 时间参数
     'start_year': start_year,
@@ -95,6 +96,7 @@ global_config.update({
     'is_bdwd': is_bdwd,
     'db_mysql': db_mysql,
     'DEFAULT_CONFIG': DEFAULT_CONFIG,
+
 })
 
 
@@ -545,17 +547,16 @@ def predict_main():
 if __name__ == '__main__':
     # global end_time
     # 遍历2024-11-25 到 2024-12-3 之间的工作日日期
-    for i_time in pd.date_range('2025-6-2', '2025-7-23', freq='B'):
-        try:
-            global_config['end_time'] = i_time.strftime('%Y-%m-%d')
-            global_config['db_mysql'].connect()    
-            predict_main()
-        except Exception as e:
-            logger.info(f'预测失败：{e}')
-            continue
-
-    # predict_main()
+    # for i_time in pd.date_range('2025-6-2', '2025-7-23', freq='B'):
+    #     try:
+    #         global_config['end_time'] = i_time.strftime('%Y-%m-%d')
+    #         global_config['db_mysql'].connect()
+    #         predict_main()
+    #     except Exception as e:
+    #         logger.info(f'预测失败：{e}')
+    #         continue
 
+    predict_main()
 
     # push_market_value()
     # sql_inset_predict(global_config)