diff --git a/lib/dataread.py b/lib/dataread.py index 1393310..6d894b9 100644 --- a/lib/dataread.py +++ b/lib/dataread.py @@ -1,6 +1,5 @@ # 导入模块 -from config_jingbo_zhoudu import * from reportlab.lib.units import cm # 单位：cm from reportlab.graphics.shapes import Drawing # 绘图工具 from reportlab.graphics.charts.legends import Legend # 图例类 @@ -51,8 +50,97 @@ plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号 # from config_jingbo import logger +global_config = { + # 核心配置项 + 'logger': None, # 日志记录器 + 'dataset': None, # 数据集路径 + 'y': None, # 目标变量列名 + 'is_fivemodels': None, + # 模型参数 + 'data_set': None, # 数据集名称 + 'input_size': None, # 输入维度 + 'horizon': None, # 预测步长 + 'train_steps': None, # 训练步数 + 'val_check_steps': None, # 验证间隔 + + # 特征工程开关 + 'is_del_corr': None, # 是否删除相关性特征 + 'is_del_tow_month': None, # 是否删除近两月未更新特征 + 'is_eta': None, # ETA功能开关 + 'is_update_eta': None, # 更新ETA开关 + '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 + + # 时间参数 + 'start_year': None, # 起始年份 + 'end_time': None, # 新增结束时间参数 ← 增加缺失的配置项 + 'freq': [None], # 数据频率（保留列表结构） + + # 数据上传 + 'upload_url': None, # 主数据上传地址 + 'upload_headers': None, # 上传请求头 + 'upload_warning_url': None, # 预警数据上传地址 + 'upload_warning_data': None, # 预警数据结构 + + # 查询接口 + 'query_data_list_item_nos_url': None, # 数据项查询地址 + 'query_data_list_item_nos_data': None, # 数据项查询参数 + + # 字段映射 + 'offsite_col': None, # 站点字段 + 'avg_col': None, # 平均值字段 + 'offsite': None, # 站点名称 + 'edbcodenamedict': None, # EDB编码映射 + 'rote': None, # 绘图上下界阈值 + + # 接口配置（原有配置） + 'login_pushreport_url': None, + 'login_data': None, + 'upload_warning_headers': None, + + # ETA配置 + 'APPID': None, + 'SECRET': None, + + # 数据库配置 + 'sqlitedb': None, +} + +# logger = global_config['logger'] +# dataset = global_config['dataset'] +# y = global_config['y'] +# data_set = global_config['data_set'] +# input_size = global_config['input_size'] +# horizon = global_config['horizon'] +# train_steps = global_config['train_steps'] +# val_check_steps = global_config['val_check_steps'] +# is_del_corr = global_config['is_del_corr'] +# is_del_tow_month = global_config['is_del_tow_month'] +# is_eta = global_config['is_eta'] +# is_update_eta = global_config['is_update_eta'] +# is_update_eta_data = global_config['is_update_eta_data'] +# start_year = global_config['start_year'] +# end_time = global_config['end_time'] +# freq = global_config['freq'][0] +# offsite_col = global_config['offsite_col'] +# avg_cols = global_config['avg_col'] +# offsite = global_config['offsite'] +# edbcodenamedict = global_config['edbcodenamedict'] +# query_data_list_item_nos_url = global_config['query_data_list_item_nos_url'] +# query_data_list_item_nos_data = global_config['query_data_list_item_nos_data'] +# config.login_pushreport_url = global_config['config.login_pushreport_url'] +# login_data = global_config['login_data'] +# upload_url = global_config['upload_url'] +# upload_warning_url = global_config['upload_warning_url'] +# upload_warning_data = global_config['upload_warning_data'] +# warning_data = global_config['upload_warning_data'] +# APPID = global_config['APPID'] +# SECRET = global_config['SECRET'] # 定义函数 + + def loadcsv(filename): """ 读取指定文件名的 CSV 文件。 @@ -145,15 +233,16 @@ def get_head_auth_report(): 返回: str: 如果登录成功，返回认证令牌；否则返回 None。 """ - logger.info("获取token中...") - logger.info(f'url:{login_pushreport_url},login_data:{login_data}') + config.logger.info("获取token中...") + config.logger.info( + f'url:{config.login_pushreport_url},login_data:{config.login_data}') # 发送 POST 请求到登录 URL，携带登录数据 - login_res = requests.post(url=login_pushreport_url, - json=login_data, timeout=(3, 30)) + login_res = requests.post(url=config.login_pushreport_url, + json=config.login_data, timeout=(3, 30)) # 将响应内容转换为 JSON 格式 text = json.loads(login_res.text) - logger.info(f'token接口响应：{text}') + config.logger.info(f'token接口响应：{text}') # 如果响应状态为成功 if text["status"]: # 从响应数据中获取认证令牌 @@ -180,30 +269,30 @@ def upload_report_data(token, upload_data): headers = {"Authorization": token} # 打印日志，显示正在上传报告数据 - logger.info("报告上传中...") + config.logger.info("报告上传中...") # 打印日志，显示认证头部信息 - logger.info(f"token:{token}") + config.logger.info(f"token:{token}") # 打印日志，显示要上传的报告数据 - logger.info(f"upload_data:{upload_data}") + config.logger.info(f"upload_data:{upload_data}") # 发送POST请求，上传报告数据 upload_res = requests.post( - url=upload_url, headers=headers, json=upload_data, timeout=(3, 15)) + url=config.upload_url, headers=headers, json=upload_data, timeout=(3, 15)) # 将响应内容转换为 JSON 格式 upload_res = json.loads(upload_res.text) # 打印日志，显示响应内容 - logger.info(upload_res) + config.logger.info(upload_res) # 如果上传成功，返回响应对象 if upload_res: return upload_res # 如果上传失败，打印日志并返回None else: - logger.info("报告上传失败") + config.logger.info("报告上传失败") return None @@ -224,27 +313,27 @@ def upload_warning_data(warning_data): headers = {"Authorization": token} # 打印日志，显示正在上传预警数据 - logger.info("预警上传中...") + config.logger.info("预警上传中...") # 打印日志，显示上传的URL - logger.info(f"upload_warning_url:{upload_warning_url}") + config.logger.info(f"upload_warning_url:{config.upload_warning_url}") # 打印日志，显示认证头部信息 - logger.info(f"token:{token}") + config.logger.info(f"token:{token}") # 打印日志，显示要上传的预警数据 - logger.info(f"warning_data:{warning_data}") + config.logger.info(f"warning_data:{config.warning_data}") # 发送POST请求，上传预警数据 upload_res = requests.post( - url=upload_warning_url, headers=headers, json=warning_data, timeout=(3, 15)) + url=config.upload_warning_url, headers=headers, json=config.warning_data, timeout=(3, 15)) # 如果上传成功，返回响应对象 if upload_res: return upload_res # 如果上传失败，打印日志并返回None else: - logger.info("预警上传失败") + config.logger.info("预警上传失败") return None @@ -259,7 +348,7 @@ def upload_warning_info(df_count): None """ # 打印日志，显示正在上传预警信息 - logger.info(f'上传预警信息') + config.logger.info(f'上传预警信息') try: # 获取当前日期 @@ -270,17 +359,17 @@ def upload_warning_info(df_count): content = f'{warning_date}有{df_count}个停更' # 更新预警数据中的日期和内容 - warning_data['data']['WARNING_DATE'] = warning_date2 - warning_data['data']['WARNING_CONTENT'] = content + config.warning_data['data']['WARNING_DATE'] = warning_date2 + config.warning_data['data']['WARNING_CONTENT'] = content # 调用 upload_warning_data 函数上传预警数据 - upload_warning_data(warning_data) + upload_warning_data(config.warning_data) # 打印日志，显示上传预警信息成功 - logger.info(f'上传预警信息成功') + config.logger.info(f'上传预警信息成功') except Exception as e: # 打印日志，显示上传预警信息失败，并记录异常信息 - logger.error(f'上传预警信息失败：{e}') + config.logger.error(f'上传预警信息失败：{e}') def create_feature_last_update_time(df): @@ -317,7 +406,8 @@ def create_feature_last_update_time(df): 0]).total_seconds() / 3600 / 24 last_update_time_datetime = datetime.datetime.strptime( last_update_time, '%Y-%m-%d') - last_update_date = end_time if end_time != '' else datetime.datetime.now().strftime('%Y-%m-%d') + last_update_date = config.end_time if config.end_time != '' else datetime.datetime.now( + ).strftime('%Y-%m-%d') end_time_datetime = datetime.datetime.strptime( last_update_date, '%Y-%m-%d') early_warning_date = last_update_time_datetime + \ @@ -327,18 +417,19 @@ def create_feature_last_update_time(df): early_warning_date = early_warning_date.strftime('%Y-%m-%d') except KeyError: time_diff = 0 - early_warning_date = end_time + early_warning_date = config.end_time continue values = values + [time_diff, early_warning_date, stop_update_period] last_update_times_df.loc[len(last_update_times_df)] = values - logger.info(f"Column {column} was last updated at {last_update_time}") + config.logger.info( + f"Column {column} was last updated at {last_update_time}") y_last_update_time = last_update_times_df[last_update_times_df['feature'] == 'y']['warning_date'].values[0] last_update_times_df.to_csv(os.path.join( - dataset, 'last_update_times.csv'), index=False) - logger.info('特征停更信息保存到文件：last_update_times.csv') + config.dataset, 'last_update_times.csv'), index=False) + config.logger.info('特征停更信息保存到文件：last_update_times.csv') return last_update_times_df, y_last_update_time @@ -378,7 +469,7 @@ def featurePindu(dataset): try: count = max(set(count), key=count.count) except ValueError: - logger.info(f'{column}列数据为空') + config.logger.info(f'{column}列数据为空') continue # 存储到字典中 count_dict[column] = count @@ -402,7 +493,7 @@ def featurePindu(dataset): # nan替换为 ' ' pindu_dfs = pindu_dfs.fillna('') pindu_dfs.to_csv(os.path.join(dataset, '特征频度统计.csv'), index=False) - logger.info(pindu_dfs) + config.logger.info(pindu_dfs) featureInfo = f'特征信息：总共有{len(columns)-2}个' for i in pindu_dfs.columns: featureInfo += f',{i}' @@ -419,10 +510,10 @@ def featurePindu(dataset): -- 向前填充，举例：采集数据开始日期为2018年1月1日，那么周度数据可能是2018年1月3日，那么3日的数据向前填充，使1日2日都有数值 数据特征相关性分析： ''' - logger.info(featureInfo) + config.logger.info(featureInfo) with open(os.path.join(dataset, '特征频度统计.txt'), 'w', encoding='utf-8') as f: f.write(featureInfo) - logger.info('*'*200) + config.logger.info('*'*200) def featureAnalysis(df, dataset, y): @@ -477,10 +568,10 @@ def corr_feature(df): df_test_noscaler = df_test.copy() # 滞后处理备份 df_noscaler = df_test.copy() # 画出相关性热力图 - df_test.to_csv(os.path.join(dataset, '同步相关性.csv')) + df_test.to_csv(os.path.join(config.dataset, '同步相关性.csv')) corr = df_test.corr() # 保存相关系数 - corr.to_csv(os.path.join(dataset, '同步相关性系数.csv')) + corr.to_csv(os.path.join(config.dataset, '同步相关性系数.csv')) # plt.figure(figsize=(10, 10)) # sns.heatmap(corr, annot=True, cmap='coolwarm') # plt.savefig('dataset/同步相关性热力图.png') @@ -502,7 +593,7 @@ def corr_feature(df): == col]['滞后周期'].values[0] # 滞后处理 df[col] = df[col].shift(period) - df.to_csv(os.path.join(dataset, '滞后处理后的数据集.csv')) + df.to_csv(os.path.join(config.dataset, '滞后处理后的数据集.csv')) # corr_feture_noscaler = {} # 保存相关性最大的周期 # 遍历df_test的每一列，计算相关性 @@ -510,7 +601,7 @@ def corr_feature(df): # # 跳过y列 # if col in ['y']: # continue - # logger.info('特征：', col) + # config.logger.info('特征：', col) # # 特征滞后n个周期，计算与y的相关性 # corr_dict = {} # try: @@ -521,10 +612,10 @@ def corr_feature(df): # df_noscaler[col+'_'+str(i)] = df_noscaler[col].shift(i) # corr_dict[col+'_'+str(i)] = abs(df_noscaler[col+'_'+str(i)].corr(df_noscaler['y'])) # except : - # logger.info('特征：', col, '滑动错误，请查看') + # config.logger.info('特征：', col, '滑动错误，请查看') # continue # 输出相关性最大的特征 - # logger.info(max(corr_dict, key=corr_dict.get), corr_dict[max(corr_dict, key=corr_dict.get)]) + # config.logger.info(max(corr_dict, key=corr_dict.get), corr_dict[max(corr_dict, key=corr_dict.get)]) # corr_feture_noscaler[col] = max(corr_dict, key=corr_dict.get).split('_')[-1] # 画出最相关性最大的特征和y的折线图 # plt.figure(figsize=(10, 5)) @@ -541,7 +632,7 @@ def corr_feature(df): # plt.savefig('dataset/特征与y的折线图_'+max(corr_dict, key=corr_dict.get).replace(':','_').replace('/','_').replace('(','_').replace(')','_')+'.png') # plt.close() # 结果保存到txt文件 - # logger.info('不参与标准化的特征滞后相关性写入txt文件') + # config.logger.info('不参与标准化的特征滞后相关性写入txt文件') # with open('dataset/不参与标准化的特征滞后相关性.txt', 'w') as f: # for key, value in corr_feture_noscaler.items(): # f.write('%s:%s\n' % (key, value)) @@ -576,7 +667,7 @@ def corr_feature(df): # # 跳过y列 # if col == 'y': # continue - # logger.info('特征：', col) + # config.logger.info('特征：', col) # # 特征滞后n个周期，计算与y的相关性 # corr_dict = {} # try: @@ -587,10 +678,10 @@ def corr_feature(df): # df_test[col+'_'+str(i)] = df_test[col].shift(i) # corr_dict[col+'_'+str(i)] = abs(df_test[col+'_'+str(i)].corr(df_test['y'])) # except : - # logger.info('特征：', col, '滑动错误，请查看') + # config.logger.info('特征：', col, '滑动错误，请查看') # continue # # 输出相关性最大的特征 - # logger.info(max(corr_dict, key=corr_dict.get), corr_dict[max(corr_dict, key=corr_dict.get)]) + # config.logger.info(max(corr_dict, key=corr_dict.get), corr_dict[max(corr_dict, key=corr_dict.get)]) # corr_feture[col] = max(corr_dict, key=corr_dict.get).split('_')[-1] # # 结果保存到txt文件 @@ -684,8 +775,8 @@ def check_column(df, col_name, two_months_ago): return True # 判断相关系数大于0.6 - if is_del_corr > 0: - if abs(df_check_column[col_name].corr(df_check_column['y'])) < is_del_corr: + if config.is_del_corr > 0: + if abs(df_check_column[col_name].corr(df_check_column['y'])) < config.is_del_corr: print(f'相关系数小于0.6：{col_name}') return True @@ -710,50 +801,50 @@ def datachuli(df_zhibiaoshuju, df_zhibiaoliebiao, datecol='date', end_time='', y df.sort_values(by='ds', inplace=True) df['ds'] = pd.to_datetime(df['ds']) # 获取start_year年到end_time的数据 - df = df[df['ds'].dt.year >= start_year] + df = df[df['ds'].dt.year >= config.start_year] df = df[df['ds'] <= end_time] # last_update_times_df,y_last_update_time = create_feature_last_update_time(df) - # logger.info(f'删除预警的特征前数据量：{df.shape}') + # config.logger.info(f'删除预警的特征前数据量：{df.shape}') # columns_to_drop = last_update_times_df[last_update_times_df['warning_date'] < y_last_update_time ]['feature'].values.tolist() # df = df.drop(columns = columns_to_drop) - # logger.info(f'删除预警的特征后数据量：{df.shape}') + # config.logger.info(f'删除预警的特征后数据量：{df.shape}') # if is_update_warning_data: # upload_warning_info(last_update_times_df,y_last_update_time) # 去掉近最后数据对应的日期在六月以前的列，删除近2月的数据是常熟的列 - if is_del_tow_month: + if config.is_del_tow_month: current_date = datetime.datetime.now() two_months_ago = current_date - timedelta(days=180) - logger.info(f'删除两月不更新特征前数据量：{df.shape}') + config.logger.info(f'删除两月不更新特征前数据量：{df.shape}') columns_to_drop = [] for clo in df.columns: if check_column(df, clo, two_months_ago): columns_to_drop.append(clo) df = df.drop(columns=columns_to_drop) - logger.info(f'删除两月不更新特征后数据量：{df.shape}') + config.logger.info(f'删除两月不更新特征后数据量：{df.shape}') # 衍生时间特征 if is_timefurture: df = addtimecharacteristics(df=df, dataset=dataset) - if freq == 'WW': + if config.freq == 'WW': # 自定义周数据 # 按weekofmothe分组取均值得到新的数据 df = df.groupby(df['yearmonthweeks']).mean() # 时间列转换为日期格式字符串 df['ds'] = df['ds'].dt.strftime('%Y-%m-%d') - elif freq == 'W': + elif config.freq == 'W': # 按周取样 df = df.resample('W', on='ds').mean().reset_index() - elif freq == 'M': + elif config.freq == 'M': # 按月取样 df = df.resample('M', on='ds').mean().reset_index() # 删除预测列空值的行 ''' 工作日缺失，如果删除，会影响预测结果，导致统计准确率出错 ''' # df = df.dropna(subset=['y']) - logger.info(f'删除预测列为空值的行后数据量：{df.shape}') + config.logger.info(f'删除预测列为空值的行后数据量：{df.shape}') df = df.dropna(axis=1, how='all') - logger.info(f'删除全为空值的列后数据量：{df.shape}') + config.logger.info(f'删除全为空值的列后数据量：{df.shape}') df.to_csv(os.path.join(dataset, '未填充的特征数据.csv'), index=False) # 去掉指标列表中的columns_to_drop的行 df_zhibiaoliebiao = df_zhibiaoliebiao[df_zhibiaoliebiao['指标名称'].isin( @@ -797,40 +888,40 @@ def zhoududatachuli(df_zhibiaoshuju, df_zhibiaoliebiao, datecol='date', end_time df.sort_values(by='ds', inplace=True) df['ds'] = pd.to_datetime(df['ds']) # 获取start_year年到end_time的数据 - df = df[df['ds'].dt.year >= start_year] + df = df[df['ds'].dt.year >= config.start_year] df = df[df['ds'] <= end_time] # last_update_times_df,y_last_update_time = create_feature_last_update_time(df) - # logger.info(f'删除预警的特征前数据量：{df.shape}') + # config.logger.info(f'删除预警的特征前数据量：{df.shape}') # columns_to_drop = last_update_times_df[last_update_times_df['warning_date'] < y_last_update_time ]['feature'].values.tolist() # df = df.drop(columns = columns_to_drop) - # logger.info(f'删除预警的特征后数据量：{df.shape}') + # config.logger.info(f'删除预警的特征后数据量：{df.shape}') # if is_update_warning_data: # upload_warning_info(last_update_times_df,y_last_update_time) # 去掉近最后数据对应的日期在六月以前的列，删除近2月的数据是常熟的列 - if is_del_tow_month: + if config.is_del_tow_month: current_date = datetime.datetime.now() two_months_ago = current_date - timedelta(days=180) - logger.info(f'删除两月不更新特征前数据量：{df.shape}') + config.logger.info(f'删除两月不更新特征前数据量：{df.shape}') columns_to_drop = [] for clo in df.columns: if check_column(df, clo, two_months_ago): columns_to_drop.append(clo) df = df.drop(columns=columns_to_drop) - logger.info(f'删除两月不更新特征后数据量：{df.shape}') + config.logger.info(f'删除两月不更新特征后数据量：{df.shape}') - if freq == 'W': + if config.freq == 'W': # 按周取样 df = df.resample('W', on='ds').mean().reset_index() - elif freq == 'M': + elif config.freq == 'M': # 按月取样 df = df.resample('M', on='ds').mean().reset_index() # 删除预测列空值的行 ''' 工作日缺失，如果删除，会影响预测结果，导致统计准确率出错 ''' # df = df.dropna(subset=['y']) - logger.info(f'删除预测列为空值的行后数据量：{df.shape}') + config.logger.info(f'删除预测列为空值的行后数据量：{df.shape}') df = df.dropna(axis=1, how='all') - logger.info(f'删除全为空值的列后数据量：{df.shape}') + config.logger.info(f'删除全为空值的列后数据量：{df.shape}') df.to_csv(os.path.join(dataset, '未填充的特征数据.csv'), index=False) # 去掉指标列表中的columns_to_drop的行 df_zhibiaoliebiao = df_zhibiaoliebiao[df_zhibiaoliebiao['指标名称'].isin( @@ -872,11 +963,11 @@ def datachuli_juxiting(df_zhibiaoshuju, df_zhibiaoliebiao, datecol='date', end_t df.rename(columns={datecol: 'ds'}, inplace=True) # 指定列统一减少数值 - df[offsite_col] = df[offsite_col]-offsite + df[config.offsite_col] = df[config.offsite_col]-config.offsite # 预测列为avg_cols的均值 - df[y] = df[avg_cols].mean(axis=1) + df[y] = df[config.avg_cols].mean(axis=1) # 去掉多余的列avg_cols - df = df.drop(columns=avg_cols) + df = df.drop(columns=config.avg_cols) # 重命名预测列 df.rename(columns={y: 'y'}, inplace=True) @@ -887,7 +978,7 @@ def datachuli_juxiting(df_zhibiaoshuju, df_zhibiaoliebiao, datecol='date', end_t df = df[df['ds'].dt.year >= 2018] # 获取小于等于当前日期的数据 df = df[df['ds'] <= end_time] - logger.info(f'删除两月不更新特征前数据量：{df.shape}') + config.logger.info(f'删除两月不更新特征前数据量：{df.shape}') # 去掉近最后数据对应的日期在两月以前的列，删除近2月的数据是常数的列 current_date = datetime.datetime.now() two_months_ago = current_date - timedelta(days=40) @@ -907,13 +998,13 @@ def datachuli_juxiting(df_zhibiaoshuju, df_zhibiaoliebiao, datecol='date', end_t columns_to_drop = df.columns[df.columns.map(check_column)].tolist() df = df.drop(columns=columns_to_drop) - logger.info(f'删除两月不更新特征后数据量：{df.shape}') + config.logger.info(f'删除两月不更新特征后数据量：{df.shape}') # 删除预测列空值的行 df = df.dropna(subset=['y']) - logger.info(f'删除预测列为空值的行后数据量：{df.shape}') + config.logger.info(f'删除预测列为空值的行后数据量：{df.shape}') df = df.dropna(axis=1, how='all') - logger.info(f'删除全为空值的列后数据量：{df.shape}') + config.logger.info(f'删除全为空值的列后数据量：{df.shape}') df.to_csv(os.path.join(dataset, '未填充的特征数据.csv'), index=False) # 去掉指标列表中的columns_to_drop的行 df_zhibiaoliebiao = df_zhibiaoliebiao[df_zhibiaoliebiao['指标名称'].isin( @@ -942,7 +1033,7 @@ def datachuli_juxiting(df_zhibiaoshuju, df_zhibiaoliebiao, datecol='date', end_t def getdata(filename, datecol='date', y='y', dataset='', add_kdj=False, is_timefurture=False, end_time=''): - logger.info('getdata接收：'+filename+' '+datecol+' '+end_time) + config.logger.info('getdata接收：'+filename+' '+datecol+' '+end_time) # 判断后缀名 csv或excel if filename.endswith('.csv'): df = loadcsv(filename) @@ -959,7 +1050,7 @@ def getdata(filename, datecol='date', y='y', dataset='', add_kdj=False, is_timef def getzhoududata(filename, datecol='date', y='y', dataset='', add_kdj=False, is_timefurture=False, end_time=''): - logger.info('getdata接收：'+filename+' '+datecol+' '+end_time) + config.logger.info('getdata接收：'+filename+' '+datecol+' '+end_time) # 判断后缀名 csv或excel if filename.endswith('.csv'): df = loadcsv(filename) @@ -976,7 +1067,7 @@ def getzhoududata(filename, datecol='date', y='y', dataset='', add_kdj=False, is def getdata_juxiting(filename, datecol='date', y='y', dataset='', add_kdj=False, is_timefurture=False, end_time=''): - logger.info('getdata接收：'+filename+' '+datecol+' '+end_time) + config.logger.info('getdata接收：'+filename+' '+datecol+' '+end_time) # 判断后缀名 csv或excel if filename.endswith('.csv'): df = loadcsv(filename) @@ -1002,6 +1093,103 @@ def sanitize_filename(filename): return sanitized +class Config: + # 核心配置 + @property + def logger(self): return global_config['logger'] + @property + def dataset(self): return global_config['dataset'] + @property + def y(self): return global_config['y'] + @property + def is_fivemodels(self): return global_config['is_fivemodels'] + + # 模型参数 + @property + def data_set(self): return global_config['data_set'] + @property + def input_size(self): return global_config['input_size'] + @property + def horizon(self): return global_config['horizon'] + @property + def train_steps(self): return global_config['train_steps'] + @property + def val_check_steps(self): return global_config['val_check_steps'] + @property + def rote(self): return global_config['rote'] + + # 特征工程开关 + @property + def is_del_corr(self): return global_config['is_del_corr'] + @property + def is_del_tow_month(self): return global_config['is_del_tow_month'] + @property + def is_eta(self): return global_config['is_eta'] + @property + def is_update_eta(self): return global_config['is_update_eta'] + @property + def is_update_eta_data(self): return global_config['is_update_eta_data'] + @property + def is_update_report(self): return global_config['is_update_report'] + + # 时间参数 + @property + def start_year(self): return global_config['start_year'] + @property + def end_time(self): return global_config['end_time'] + @property + def freq(self): return global_config['freq'] + + # 接口配置 + @property + def upload_url(self): return global_config['upload_url'] + + @property + def login_pushreport_url( + self): return global_config['login_pushreport_url'] + + @property + def login_data(self): return global_config['login_data'] + @property + def upload_headers(self): return global_config['upload_headers'] + @property + def upload_warning_url(self): return global_config['upload_warning_url'] + @property + def upload_warning_data(self): return global_config['upload_warning_data'] + + # 查询接口 + @property + def query_data_list_item_nos_url( + self): return global_config['query_data_list_item_nos_url'] + + @property + def query_data_list_item_nos_data( + self): return global_config['query_data_list_item_nos_data'] + + # 字段映射 + @property + def offsite_col(self): return global_config['offsite_col'] + @property + def avg_col(self): return global_config['avg_col'] + @property + def offsite(self): return global_config['offsite'] + @property + def edbcodenamedict(self): return global_config['edbcodenamedict'] + + # ETA配置 + @property + def APPID(self): return global_config['APPID'] + @property + def SECRET(self): return global_config['SECRET'] + + # 数据库配置 + @property + def sqlitedb(self): return global_config['sqlitedb'] + + +config = Config() + + class BinanceAPI: ''' 获取 Binance API 请求头签名 @@ -1397,7 +1585,8 @@ class EtaReader(): else: # 请求失败，打印错误信息 - logger.info(f'Error: {response.status_code}, {response.text}') + config.logger.info( + f'Error: {response.status_code}, {response.text}') # 主动抛出异常 raise Exception(f'Error: {response.status_code}, {response.text}') @@ -1449,7 +1638,7 @@ class EtaReader(): data = response.json() # 假设接口返回的是JSON数据 # 请求成功，处理响应内容 - # logger.info(data.get('Data')) + # config.logger.info(data.get('Data')) # 定义你想要保留的固定值 fixed_value = 1214 @@ -1469,7 +1658,7 @@ class EtaReader(): url = self.classifyidlisturl+str(ClassifyId) response = requests.get(url, headers=self.headers) if response.status_code == 200: - # logger.info(response.text) + # config.logger.info(response.text) data2 = response.json() Data = data2.get('Data') for i in Data: @@ -1504,7 +1693,8 @@ class EtaReader(): edbname_df = edbname_df.dropna() if len(edbname_df) == 0: - logger.info(f'指标名称：{EdbName} 没有数据') + config.logger.info( + f'指标名称：{EdbName} 没有数据') continue try: time_sequence = edbname_df['DataTime'].values.tolist( @@ -1554,18 +1744,18 @@ class EtaReader(): # df = pd.merge(df, df2, how='outer') df = pd.concat([df, df2]) else: - logger.info(f'跳过指标 {EdbName}') + config.logger.info(f'跳过指标 {EdbName}') # 找到列表中不在指标列中的指标id，保存成新的list new_list = [ item for item in self.edbcodelist if item not in df['指标id'].tolist()] - logger.info(new_list) + config.logger.info(new_list) # 遍历new_list，获取指标数据，保存到df1 for item in new_list: - logger.info(item) + config.logger.info(item) # 将item 加入到 df['指标id']中 try: - itemname = edbcodenamedict[item] + itemname = config.edbcodenamedict[item] except: itemname = item @@ -1579,8 +1769,8 @@ class EtaReader(): # df1.dropna(inplace=True) # 去掉大于今天日期的行 df1 = df1[df1['date'] <= datetime.datetime.now().strftime('%Y-%m-%d')] - logger.info(df1.head()) - # logger.info(f'{df1.head()}') + config.logger.info(df1.head()) + # config.logger.info(f'{df1.head()}') df_zhibiaoshuju = df1.copy() df_zhibiaoliebiao = df.copy() @@ -1624,7 +1814,7 @@ class EtaReader(): data = response.json() # 假设接口返回的是JSON数据 # 请求成功，处理响应内容 - # logger.info(data.get('Data')) + # config.logger.info(data.get('Data')) # 定义你想要保留的固定值 fixed_value = ClassifyId @@ -1644,7 +1834,7 @@ class EtaReader(): url = self.classifyidlisturl+str(ClassifyId) response = requests.get(url, headers=self.headers) if response.status_code == 200: - # logger.info(response.text) + # config.logger.info(response.text) data2 = response.json() Data = data2.get('Data') for i in Data: @@ -1669,18 +1859,18 @@ class EtaReader(): df = pd.concat([df, df2]) df1 = self.edbcodegetdata(df1, EdbCode, EdbName) else: - logger.info(f'跳过指标 {EdbName}') + config.logger.info(f'跳过指标 {EdbName}') # 找到列表中不在指标列中的指标id，保存成新的list new_list = [ item for item in self.edbcodelist if item not in df['指标id'].tolist()] - logger.info(new_list) + config.logger.info(new_list) # 遍历new_list，获取指标数据，保存到df1 for item in new_list: - logger.info(item) + config.logger.info(item) # 将item 加入到 df['指标id']中 try: - itemname = edbcodenamedict[item] + itemname = config.edbcodenamedict[item] except: itemname = item @@ -1694,8 +1884,8 @@ class EtaReader(): # df1.dropna(inplace=True) # 去掉大于今天日期的行 df1 = df1[df1['date'] <= datetime.datetime.now().strftime('%Y-%m-%d')] - logger.info(df1.head()) - # logger.info(f'{df1.head()}') + config.logger.info(df1.head()) + # config.logger.info(f'{df1.head()}') # 保存到xlsx文件的sheet表 with pd.ExcelWriter(os.path.join(dataset, data_set)) as file: df1.to_excel(file, sheet_name='指标数据', index=False) @@ -1719,7 +1909,7 @@ class EtaReader(): } # 发送post请求 上传数据 - logger.info(f'请求参数：{data}') + config.logger.info(f'请求参数：{data}') response = requests.post( self.edbdatapushurl, headers=self.headers, data=json.dumps(data)) @@ -1727,11 +1917,12 @@ class EtaReader(): if response.status_code == 200: data = response.json() # 假设接口返回的是JSON数据 - logger.info(f'上传成功，响应为：{data}') + config.logger.info(f'上传成功，响应为：{data}') else: # 请求失败，打印错误信息 - logger.info(f'Error: {response.status_code}, {response.text}') + config.logger.info( + f'Error: {response.status_code}, {response.text}') # 主动抛出异常 raise Exception(f'Error: {response.status_code}, {response.text}') @@ -1758,11 +1949,12 @@ class EtaReader(): if response.status_code == 200: data = response.json() # 假设接口返回的是JSON数据 - logger.info('删除成功，响应为：', data) + config.logger.info('删除成功，响应为：', data) else: # 请求失败，打印错误信息 - logger.info(f'Error: {response.status_code}, {response.text}') + config.logger.info( + f'Error: {response.status_code}, {response.text}') # 主动抛出异常 raise Exception(f'Error: {response.status_code}, {response.text}') @@ -1797,11 +1989,12 @@ class EtaReader(): if response.status_code == 200: data = response.json() # 假设接口返回的是JSON数据 - logger.info('删除成功，响应为：', data) + config.logger.info('删除成功，响应为：', data) else: # 请求失败，打印错误信息 - logger.info(f'Error: {response.status_code}, {response.text}') + config.logger.info( + f'Error: {response.status_code}, {response.text}') # 主动抛出异常 raise Exception(f'Error: {response.status_code}, {response.text}') @@ -1813,15 +2006,15 @@ def get_market_data(end_time, df): # 获取token token = get_head_auth_report() # 定义请求参数 - query_data_list_item_nos_data['data']['dateEnd'] = end_time.replace( + config.query_data_list_item_nos_data['data']['dateEnd'] = end_time.replace( '-', '') # 发送请求 headers = {"Authorization": token} - logger.info('获取数据中...') - items_res = requests.post(url=query_data_list_item_nos_url, headers=headers, - json=query_data_list_item_nos_data, timeout=(3, 35)) + config.logger.info('获取数据中...') + items_res = requests.post(url=config.query_data_list_item_nos_url, headers=headers, + json=config.query_data_list_item_nos_data, timeout=(3, 35)) json_data = json.loads(items_res.text) - logger.info(f"获取到的数据:{json_data}") + config.logger.info(f"获取到的数据:{json_data}") df3 = pd.DataFrame(json_data['data']) # 按照dataItemNo 分组 得到多个dataframe ，最后根据dataDate merge 成一个dataframe df2 = pd.DataFrame() @@ -1843,7 +2036,7 @@ def get_market_data(end_time, df): def get_high_low_data(df): # 读取excel 从第五行开始 - df1 = pd.read_excel(os.path.join(dataset, '数据项下载.xls'), header=5, names=[ + df1 = pd.read_excel(os.path.join(config.dataset, '数据项下载.xls'), header=5, names=[ 'numid', 'date', 'Brentzdj', 'Brentzgj']) # 合并数据 df = pd.merge(df, df1, how='left', on='date') diff --git a/main_yuanyou_zhoudu.py b/main_yuanyou_zhoudu.py index 441b236..1e8e8b2 100644 --- a/main_yuanyou_zhoudu.py +++ b/main_yuanyou_zhoudu.py @@ -1,14 +1,67 @@ # 读取配置 from lib.dataread import * -# from config_jingbo_zhoudu import * +from config_jingbo_zhoudu import * from lib.tools import SendMail, exception_logger from models.nerulforcastmodels import ex_Model, model_losss, model_losss_juxiting, brent_export_pdf, tansuanli_export_pdf, pp_export_pdf, model_losss_juxiting - -import glob +import datetime import torch torch.set_float32_matmul_precision("high") +global_config.update({ + # 核心参数 + 'logger': logger, + 'dataset': dataset, + 'y': y, + 'is_debug': is_debug, + 'is_train': is_train, + 'is_fivemodels': is_fivemodels, + 'settings': settings, + + + # 模型参数 + 'data_set': data_set, + 'input_size': input_size, + 'horizon': horizon, + 'train_steps': train_steps, + 'val_check_steps': val_check_steps, + 'val_size': val_size, + 'test_size': test_size, + 'modelsindex': modelsindex, + 'rote': rote, + + # 特征工程开关 + 'is_del_corr': is_del_corr, + 'is_del_tow_month': is_del_tow_month, + 'is_eta': is_eta, + 'is_update_eta': is_update_eta, + 'early_stop_patience_steps': early_stop_patience_steps, + + # 时间参数 + 'start_year': start_year, + 'end_time': end_time, + 'freq': freq, # 保持列表结构 + + # 接口配置 + 'login_pushreport_url': login_pushreport_url, + 'login_data': login_data, + 'upload_url': upload_url, + 'upload_warning_url': upload_warning_url, + 'warning_data': warning_data, + + # 查询接口 + 'query_data_list_item_nos_url': query_data_list_item_nos_url, + 'query_data_list_item_nos_data': query_data_list_item_nos_data, + + # eta 配置 + 'APPID': APPID, + 'SECRET': SECRET, + 'etadata': data, + + # 数据库配置 + 'sqlitedb': sqlitedb, +}) + def predict_main(): """ @@ -49,228 +102,237 @@ def predict_main(): 返回: None """ - global end_time - signature = BinanceAPI(APPID, SECRET) - etadata = EtaReader(signature=signature, - classifylisturl=classifylisturl, - classifyidlisturl=classifyidlisturl, - edbcodedataurl=edbcodedataurl, - edbcodelist=edbcodelist, - edbdatapushurl=edbdatapushurl, - edbdeleteurl=edbdeleteurl, - edbbusinessurl=edbbusinessurl - ) - # 获取数据 - if is_eta: - logger.info('从eta获取数据...') - signature = BinanceAPI(APPID, SECRET) - etadata = EtaReader(signature=signature, - classifylisturl=classifylisturl, - classifyidlisturl=classifyidlisturl, - edbcodedataurl=edbcodedataurl, - edbcodelist=edbcodelist, - edbdatapushurl=edbdatapushurl, - edbdeleteurl=edbdeleteurl, - edbbusinessurl=edbbusinessurl, - ) - df_zhibiaoshuju, df_zhibiaoliebiao = etadata.get_eta_api_yuanyou_data( - data_set=data_set, dataset=dataset) # 原始数据，未处理 + # global end_time + # signature = BinanceAPI(APPID, SECRET) + # etadata = EtaReader(signature=signature, + # classifylisturl=classifylisturl, + # classifyidlisturl=classifyidlisturl, + # edbcodedataurl=edbcodedataurl, + # edbcodelist=edbcodelist, + # edbdatapushurl=edbdatapushurl, + # edbdeleteurl=edbdeleteurl, + # edbbusinessurl=edbbusinessurl + # ) + # # 获取数据 + # if is_eta: + # logger.info('从eta获取数据...') + # signature = BinanceAPI(APPID, SECRET) + # etadata = EtaReader(signature=signature, + # classifylisturl=classifylisturl, + # classifyidlisturl=classifyidlisturl, + # edbcodedataurl=edbcodedataurl, + # edbcodelist=edbcodelist, + # edbdatapushurl=edbdatapushurl, + # edbdeleteurl=edbdeleteurl, + # edbbusinessurl=edbbusinessurl, + # ) + # df_zhibiaoshuju, df_zhibiaoliebiao = etadata.get_eta_api_yuanyou_data( + # data_set=data_set, dataset=dataset) # 原始数据，未处理 - if is_market: - logger.info('从市场信息平台获取数据...') - try: - # 如果是测试环境，最高价最低价取excel文档 - if server_host == '192.168.100.53': - logger.info('从excel文档获取最高价最低价') - df_zhibiaoshuju = get_high_low_data(df_zhibiaoshuju) - else: - logger.info('从市场信息平台获取数据') - df_zhibiaoshuju = get_market_data( - end_time, df_zhibiaoshuju) + # if is_market: + # logger.info('从市场信息平台获取数据...') + # try: + # # 如果是测试环境，最高价最低价取excel文档 + # if server_host == '192.168.100.53': + # logger.info('从excel文档获取最高价最低价') + # df_zhibiaoshuju = get_high_low_data(df_zhibiaoshuju) + # else: + # logger.info('从市场信息平台获取数据') + # df_zhibiaoshuju = get_market_data( + # end_time, df_zhibiaoshuju) - except: - logger.info('最高最低价拼接失败') + # except: + # logger.info('最高最低价拼接失败') - # 保存到xlsx文件的sheet表 - with pd.ExcelWriter(os.path.join(dataset, data_set)) as file: - df_zhibiaoshuju.to_excel(file, sheet_name='指标数据', index=False) - df_zhibiaoliebiao.to_excel(file, sheet_name='指标列表', index=False) + # # 保存到xlsx文件的sheet表 + # with pd.ExcelWriter(os.path.join(dataset, data_set)) as file: + # df_zhibiaoshuju.to_excel(file, sheet_name='指标数据', index=False) + # df_zhibiaoliebiao.to_excel(file, sheet_name='指标列表', index=False) - # 数据处理 - df = datachuli(df_zhibiaoshuju, df_zhibiaoliebiao, y=y, dataset=dataset, add_kdj=add_kdj, is_timefurture=is_timefurture, - end_time=end_time) + # # 数据处理 + # df = datachuli(df_zhibiaoshuju, df_zhibiaoliebiao, y=y, dataset=dataset, add_kdj=add_kdj, is_timefurture=is_timefurture, + # end_time=end_time) - else: - # 读取数据 - logger.info('读取本地数据：' + os.path.join(dataset, data_set)) - df, df_zhibiaoliebiao = getdata(filename=os.path.join(dataset, data_set), y=y, dataset=dataset, add_kdj=add_kdj, - is_timefurture=is_timefurture, end_time=end_time) # 原始数据，未处理 + # else: + # # 读取数据 + # logger.info('读取本地数据：' + os.path.join(dataset, data_set)) + # df, df_zhibiaoliebiao = getdata(filename=os.path.join(dataset, data_set), y=y, dataset=dataset, add_kdj=add_kdj, + # is_timefurture=is_timefurture, end_time=end_time) # 原始数据，未处理 - # 更改预测列名称 - df.rename(columns={y: 'y'}, inplace=True) + # # 更改预测列名称 + # df.rename(columns={y: 'y'}, inplace=True) - if is_edbnamelist: - df = df[edbnamelist] - df.to_csv(os.path.join(dataset, '指标数据.csv'), index=False) - # 保存最新日期的y值到数据库 - # 取第一行数据存储到数据库中 - first_row = df[['ds', 'y']].tail(1) - # 判断y的类型是否为float - if not isinstance(first_row['y'].values[0], float): - logger.info(f'{end_time}预测目标数据为空，跳过') - return None + # if is_edbnamelist: + # df = df[edbnamelist] + # df.to_csv(os.path.join(dataset, '指标数据.csv'), index=False) + # # 保存最新日期的y值到数据库 + # # 取第一行数据存储到数据库中 + # first_row = df[['ds', 'y']].tail(1) + # # 判断y的类型是否为float + # if not isinstance(first_row['y'].values[0], float): + # logger.info(f'{end_time}预测目标数据为空，跳过') + # return None - # 将最新真实值保存到数据库 - if not sqlitedb.check_table_exists('trueandpredict'): - first_row.to_sql('trueandpredict', sqlitedb.connection, index=False) - else: - for row in first_row.itertuples(index=False): - row_dict = row._asdict() - # row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d') - # row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d %H:%M:%S') - check_query = sqlitedb.select_data( - 'trueandpredict', where_condition=f"ds = '{row.ds}'") - if len(check_query) > 0: - set_clause = ", ".join( - [f"{key} = '{value}'" for key, value in row_dict.items()]) - sqlitedb.update_data( - 'trueandpredict', set_clause, where_condition=f"ds = '{row.ds}'") - continue - sqlitedb.insert_data('trueandpredict', tuple( - row_dict.values()), columns=row_dict.keys()) + # # 将最新真实值保存到数据库 + # if not sqlitedb.check_table_exists('trueandpredict'): + # first_row.to_sql('trueandpredict', sqlitedb.connection, index=False) + # else: + # for row in first_row.itertuples(index=False): + # row_dict = row._asdict() + # config.logger.info(f'要保存的真实值：{row_dict}') + # # 判断ds是否为字符串类型,如果不是则转换为字符串类型 + # if isinstance(row_dict['ds'], (pd.Timestamp, datetime.datetime)): + # row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d') + # elif not isinstance(row_dict['ds'], str): + # try: + # row_dict['ds'] = pd.to_datetime( + # row_dict['ds']).strftime('%Y-%m-%d') + # except: + # logger.warning(f"无法解析的时间格式: {row_dict['ds']}") + # # row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d') + # # row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d %H:%M:%S') + # check_query = sqlitedb.select_data( + # 'trueandpredict', where_condition=f"ds = '{row.ds}'") + # if len(check_query) > 0: + # set_clause = ", ".join( + # [f"{key} = '{value}'" for key, value in row_dict.items()]) + # sqlitedb.update_data( + # 'trueandpredict', set_clause, where_condition=f"ds = '{row.ds}'") + # continue + # sqlitedb.insert_data('trueandpredict', tuple( + # row_dict.values()), columns=row_dict.keys()) - # 更新accuracy表的y值 - if not sqlitedb.check_table_exists('accuracy'): - pass - else: - update_y = sqlitedb.select_data( - 'accuracy', where_condition="y is null") - if len(update_y) > 0: - logger.info('更新accuracy表的y值') - # 找到update_y 中ds且df中的y的行 - update_y = update_y[update_y['ds'] <= end_time] - logger.info(f'要更新y的信息：{update_y}') - # try: - for row in update_y.itertuples(index=False): - try: - row_dict = row._asdict() - yy = df[df['ds'] == row_dict['ds']]['y'].values[0] - LOW = df[df['ds'] == row_dict['ds']]['Brentzdj'].values[0] - HIGH = df[df['ds'] == row_dict['ds']]['Brentzgj'].values[0] - sqlitedb.update_data( - 'accuracy', f"y = {yy},LOW_PRICE = {LOW},HIGH_PRICE = {HIGH}", where_condition=f"ds = '{row_dict['ds']}'") - except: - logger.info(f'更新accuracy表的y值失败：{row_dict}') - # except Exception as e: - # logger.info(f'更新accuracy表的y值失败：{e}') + # # 更新accuracy表的y值 + # if not sqlitedb.check_table_exists('accuracy'): + # pass + # else: + # update_y = sqlitedb.select_data( + # 'accuracy', where_condition="y is null") + # if len(update_y) > 0: + # logger.info('更新accuracy表的y值') + # # 找到update_y 中ds且df中的y的行 + # update_y = update_y[update_y['ds'] <= end_time] + # logger.info(f'要更新y的信息：{update_y}') + # # try: + # for row in update_y.itertuples(index=False): + # try: + # row_dict = row._asdict() + # yy = df[df['ds'] == row_dict['ds']]['y'].values[0] + # LOW = df[df['ds'] == row_dict['ds']]['Brentzdj'].values[0] + # HIGH = df[df['ds'] == row_dict['ds']]['Brentzgj'].values[0] + # sqlitedb.update_data( + # 'accuracy', f"y = {yy},LOW_PRICE = {LOW},HIGH_PRICE = {HIGH}", where_condition=f"ds = '{row_dict['ds']}'") + # except: + # logger.info(f'更新accuracy表的y值失败：{row_dict}') + # # except Exception as e: + # # logger.info(f'更新accuracy表的y值失败：{e}') - import datetime - # 判断当前日期是不是周一 - is_weekday = datetime.datetime.now().weekday() == 0 - if is_weekday: - logger.info('今天是周一，更新预测模型') - # 计算最近60天预测残差最低的模型名称 - model_results = sqlitedb.select_data( - 'trueandpredict', order_by="ds DESC", limit="60") - # 删除空值率为90%以上的列 - if len(model_results) > 10: - model_results = model_results.dropna( - thresh=len(model_results)*0.1, axis=1) - # 删除空行 - model_results = model_results.dropna() - modelnames = model_results.columns.to_list()[2:-1] - for col in model_results[modelnames].select_dtypes(include=['object']).columns: - model_results[col] = model_results[col].astype(np.float32) - # 计算每个预测值与真实值之间的偏差率 - for model in modelnames: - model_results[f'{model}_abs_error_rate'] = abs( - model_results['y'] - model_results[model]) / model_results['y'] - # 获取每行对应的最小偏差率值 - min_abs_error_rate_values = model_results.apply( - lambda row: row[[f'{model}_abs_error_rate' for model in modelnames]].min(), axis=1) - # 获取每行对应的最小偏差率值对应的列名 - min_abs_error_rate_column_name = model_results.apply( - lambda row: row[[f'{model}_abs_error_rate' for model in modelnames]].idxmin(), axis=1) - # 将列名索引转换为列名 - min_abs_error_rate_column_name = min_abs_error_rate_column_name.map( - lambda x: x.split('_')[0]) - # 取出现次数最多的模型名称 - most_common_model = min_abs_error_rate_column_name.value_counts().idxmax() - logger.info(f"最近60天预测残差最低的模型名称：{most_common_model}") - # 保存结果到数据库 - if not sqlitedb.check_table_exists('most_model'): - sqlitedb.create_table( - 'most_model', columns="ds datetime, most_common_model TEXT") - sqlitedb.insert_data('most_model', (datetime.datetime.now().strftime( - '%Y-%m-%d %H:%M:%S'), most_common_model,), columns=('ds', 'most_common_model',)) + # # 判断当前日期是不是周一 + # is_weekday = datetime.datetime.now().weekday() == 0 + # if is_weekday: + # logger.info('今天是周一，更新预测模型') + # # 计算最近60天预测残差最低的模型名称 + # model_results = sqlitedb.select_data( + # 'trueandpredict', order_by="ds DESC", limit="60") + # # 删除空值率为90%以上的列 + # if len(model_results) > 10: + # model_results = model_results.dropna( + # thresh=len(model_results)*0.1, axis=1) + # # 删除空行 + # model_results = model_results.dropna() + # modelnames = model_results.columns.to_list()[2:-1] + # for col in model_results[modelnames].select_dtypes(include=['object']).columns: + # model_results[col] = model_results[col].astype(np.float32) + # # 计算每个预测值与真实值之间的偏差率 + # for model in modelnames: + # model_results[f'{model}_abs_error_rate'] = abs( + # model_results['y'] - model_results[model]) / model_results['y'] + # # 获取每行对应的最小偏差率值 + # min_abs_error_rate_values = model_results.apply( + # lambda row: row[[f'{model}_abs_error_rate' for model in modelnames]].min(), axis=1) + # # 获取每行对应的最小偏差率值对应的列名 + # min_abs_error_rate_column_name = model_results.apply( + # lambda row: row[[f'{model}_abs_error_rate' for model in modelnames]].idxmin(), axis=1) + # # 将列名索引转换为列名 + # min_abs_error_rate_column_name = min_abs_error_rate_column_name.map( + # lambda x: x.split('_')[0]) + # # 取出现次数最多的模型名称 + # most_common_model = min_abs_error_rate_column_name.value_counts().idxmax() + # logger.info(f"最近60天预测残差最低的模型名称：{most_common_model}") + # # 保存结果到数据库 + # if not sqlitedb.check_table_exists('most_model'): + # sqlitedb.create_table( + # 'most_model', columns="ds datetime, most_common_model TEXT") + # sqlitedb.insert_data('most_model', (datetime.datetime.now().strftime( + # '%Y-%m-%d %H:%M:%S'), most_common_model,), columns=('ds', 'most_common_model',)) - try: - if is_weekday: - # if True: - logger.info('今天是周一，发送特征预警') - # 上传预警信息到数据库 - warning_data_df = df_zhibiaoliebiao.copy() - warning_data_df = warning_data_df[warning_data_df['停更周期'] > 3][[ - '指标名称', '指标id', '频度', '更新周期', '指标来源', '最后更新时间', '停更周期']] - # 重命名列名 - warning_data_df = warning_data_df.rename(columns={'指标名称': 'INDICATOR_NAME', '指标id': 'INDICATOR_ID', '频度': 'FREQUENCY', - '更新周期': 'UPDATE_FREQUENCY', '指标来源': 'DATA_SOURCE', '最后更新时间': 'LAST_UPDATE_DATE', '停更周期': 'UPDATE_SUSPENSION_CYCLE'}) - from sqlalchemy import create_engine - import urllib - global password - if '@' in password: - password = urllib.parse.quote_plus(password) + # try: + # if is_weekday: + # # if True: + # logger.info('今天是周一，发送特征预警') + # # 上传预警信息到数据库 + # warning_data_df = df_zhibiaoliebiao.copy() + # warning_data_df = warning_data_df[warning_data_df['停更周期'] > 3][[ + # '指标名称', '指标id', '频度', '更新周期', '指标来源', '最后更新时间', '停更周期']] + # # 重命名列名 + # warning_data_df = warning_data_df.rename(columns={'指标名称': 'INDICATOR_NAME', '指标id': 'INDICATOR_ID', '频度': 'FREQUENCY', + # '更新周期': 'UPDATE_FREQUENCY', '指标来源': 'DATA_SOURCE', '最后更新时间': 'LAST_UPDATE_DATE', '停更周期': 'UPDATE_SUSPENSION_CYCLE'}) + # from sqlalchemy import create_engine + # import urllib + # global password + # if '@' in password: + # password = urllib.parse.quote_plus(password) - engine = create_engine( - f'mysql+pymysql://{dbusername}:{password}@{host}:{port}/{dbname}') - warning_data_df['WARNING_DATE'] = datetime.date.today().strftime( - "%Y-%m-%d %H:%M:%S") - warning_data_df['TENANT_CODE'] = 'T0004' - # 插入数据之前查询表数据然后新增id列 - existing_data = pd.read_sql(f"SELECT * FROM {table_name}", engine) - if not existing_data.empty: - max_id = existing_data['ID'].astype(int).max() - warning_data_df['ID'] = range( - max_id + 1, max_id + 1 + len(warning_data_df)) - else: - warning_data_df['ID'] = range(1, 1 + len(warning_data_df)) - warning_data_df.to_sql( - table_name, con=engine, if_exists='append', index=False) - if is_update_warning_data: - upload_warning_info(len(warning_data_df)) - except: - logger.info('上传预警信息到数据库失败') + # engine = create_engine( + # f'mysql+pymysql://{dbusername}:{password}@{host}:{port}/{dbname}') + # warning_data_df['WARNING_DATE'] = datetime.date.today().strftime( + # "%Y-%m-%d %H:%M:%S") + # warning_data_df['TENANT_CODE'] = 'T0004' + # # 插入数据之前查询表数据然后新增id列 + # existing_data = pd.read_sql(f"SELECT * FROM {table_name}", engine) + # if not existing_data.empty: + # max_id = existing_data['ID'].astype(int).max() + # warning_data_df['ID'] = range( + # max_id + 1, max_id + 1 + len(warning_data_df)) + # else: + # warning_data_df['ID'] = range(1, 1 + len(warning_data_df)) + # warning_data_df.to_sql( + # table_name, con=engine, if_exists='append', index=False) + # if is_update_warning_data: + # upload_warning_info(len(warning_data_df)) + # except: + # logger.info('上传预警信息到数据库失败') - if is_corr: - df = corr_feature(df=df) + # if is_corr: + # df = corr_feature(df=df) - df1 = df.copy() # 备份一下，后面特征筛选完之后加入ds y 列用 - logger.info(f"开始训练模型...") - row, col = df.shape + # df1 = df.copy() # 备份一下，后面特征筛选完之后加入ds y 列用 + # logger.info(f"开始训练模型...") + # row, col = df.shape - now = datetime.datetime.now().strftime('%Y%m%d%H%M%S') - ex_Model(df, - horizon=horizon, - input_size=input_size, - train_steps=train_steps, - val_check_steps=val_check_steps, - early_stop_patience_steps=early_stop_patience_steps, - is_debug=is_debug, - dataset=dataset, - is_train=is_train, - is_fivemodels=is_fivemodels, - val_size=val_size, - test_size=test_size, - settings=settings, - now=now, - etadata=etadata, - modelsindex=modelsindex, - data=data, - is_eta=is_eta, - end_time=end_time, - ) + # now = datetime.datetime.now().strftime('%Y%m%d%H%M%S') + # ex_Model(df, + # horizon=global_config['horizon'], + # input_size=global_config['input_size'], + # train_steps=global_config['train_steps'], + # val_check_steps=global_config['val_check_steps'], + # early_stop_patience_steps=global_config['early_stop_patience_steps'], + # is_debug=global_config['is_debug'], + # dataset=global_config['dataset'], + # is_train=global_config['is_train'], + # is_fivemodels=global_config['is_fivemodels'], + # val_size=global_config['val_size'], + # test_size=global_config['test_size'], + # settings=global_config['settings'], + # now=now, + # etadata=global_config['etadata'], + # modelsindex=global_config['modelsindex'], + # data=data, + # is_eta=global_config['is_eta'], + # end_time=global_config['end_time'], + # ) - logger.info('模型训练完成') + # logger.info('模型训练完成') logger.info('训练数据绘图ing') model_results3 = model_losss(sqlitedb, end_time=end_time) diff --git a/models/nerulforcastmodels.py b/models/nerulforcastmodels.py index 621a0b3..7ce32d0 100644 --- a/models/nerulforcastmodels.py +++ b/models/nerulforcastmodels.py @@ -112,11 +112,11 @@ def ex_Model(df, horizon, input_size, train_steps, val_check_steps, early_stop_p df_test['unique_id'] = 1 # 显示划分后的数据集的前几行 - logger.info("Training set head:") - logger.info(df_train.head()) + config.logger.info("Training set head:") + config.logger.info(df_train.head()) - logger.info("\nTesting set head:") - logger.info(df_test.head()) + config.logger.info("\nTesting set head:") + config.logger.info(df_test.head()) models = [ NHITS(h=horizon, input_size=input_size, max_steps=train_steps, val_check_steps=val_check_steps, @@ -185,7 +185,8 @@ def ex_Model(df, horizon, input_size, train_steps, val_check_steps, early_stop_p models.append(model) # 创建NeuralForecast实例并训练模型 - nf = NeuralForecast(models=models, freq=freq[0]) + # freq = 'B' + nf = NeuralForecast(models=models, freq=config.freq[:1]) from joblib import dump, load if is_train: @@ -207,7 +208,7 @@ def ex_Model(df, horizon, input_size, train_steps, val_check_steps, early_stop_p import glob filename = max(glob.glob(os.path.join( dataset, '*.joblib')), key=os.path.getctime) - logger.info('读取模型：' + filename) + config.logger.info('读取模型：' + filename) nf = load(filename) # 测试集预测 nf_test_preds = nf.cross_validation( @@ -233,10 +234,10 @@ def ex_Model(df, horizon, input_size, train_steps, val_check_steps, early_stop_p df_predict.to_csv(os.path.join(dataset, "predict.csv"), index=False) # 将预测结果保存到数据库 - save_to_database(sqlitedb, df_predict, 'predict', end_time) + save_to_database(config.sqlitedb, df_predict, 'predict', end_time) # 把预测值上传到eta - if is_update_eta: + if config.is_update_eta: df_predict['ds'] = pd.to_datetime(df_predict['ds']) dates = df_predict['ds'].dt.strftime('%Y-%m-%d') @@ -942,12 +943,12 @@ def model_losss(sqlitedb, end_time): df_combined['CREAT_DATE'] = df_combined['cutoff'] df_combined4 = df_combined.copy() # 备份df_combined,后面画图需要 # 删除缺失值大于80%的列 - logger.info(df_combined.shape) + config.logger.info(df_combined.shape) df_combined = df_combined.loc[:, df_combined.isnull().mean() < 0.8] - logger.info(df_combined.shape) + config.logger.info(df_combined.shape) # 删除缺失值 df_combined.dropna(inplace=True) - logger.info(df_combined.shape) + config.logger.info(df_combined.shape) # 其他列转为数值类型 df_combined = df_combined.astype({col: 'float32' for col in df_combined.columns if col not in [ 'CREAT_DATE', 'ds', 'created_dt', 'cutoff']}) @@ -989,17 +990,17 @@ def model_losss(sqlitedb, end_time): model_results3 = model_results3.sort_values( by='平均平方误差(MSE)', ascending=True) model_results3.to_csv(os.path.join( - dataset, "model_evaluation.csv"), index=False) + config.dataset, "model_evaluation.csv"), index=False) modelnames = model_results3['模型(Model)'].tolist() most_model_name = modelnames[0] allmodelnames = modelnames.copy() # 保存5个最佳模型的名称 if len(modelnames) > 5: modelnames = modelnames[0:5] - if is_fivemodels: + if config.is_fivemodels: pass else: - with open(os.path.join(dataset, "best_modelnames.txt"), 'w') as f: + with open(os.path.join(config.dataset, "best_modelnames.txt"), 'w') as f: f.write(','.join(modelnames) + '\n') # 预测值与真实值对比图 @@ -1014,12 +1015,13 @@ def model_losss(sqlitedb, end_time): plt.ylabel('价格') plt.title(model+'拟合') plt.subplots_adjust(hspace=0.5) - plt.savefig(os.path.join(dataset, '预测值与真实值对比图.png'), bbox_inches='tight') + plt.savefig(os.path.join(config.dataset, '预测值与真实值对比图.png'), + bbox_inches='tight') plt.close() # # 历史数据+预测数据 # # 拼接未来时间预测 - df_predict = pd.read_csv(os.path.join(dataset, 'predict.csv')) + df_predict = pd.read_csv(os.path.join(config.dataset, 'predict.csv')) df_predict.drop('unique_id', inplace=True, axis=1) df_predict.dropna(axis=1, inplace=True) @@ -1071,7 +1073,7 @@ def model_losss(sqlitedb, end_time): def add_rote_column(row): columns = [] for r in names_df.columns: - if row[r] <= rote: + if row[r] <= config.rote: columns.append(r.split('-')[0]) return pd.Series([columns], index=['columns']) names_df['columns'] = names_df.apply(add_rote_column, axis=1) @@ -1124,7 +1126,7 @@ def model_losss(sqlitedb, end_time): df_combined3['ds'] = pd.to_datetime(df_combined3['ds']) df_combined3['ds'] = df_combined3['ds'].dt.strftime('%Y-%m-%d') - df_predict2 = df_combined3.tail(horizon) + df_predict2 = df_combined3.tail(config.horizon) # 保存到数据库 if not sqlitedb.check_table_exists('accuracy'): @@ -1167,7 +1169,8 @@ def model_losss(sqlitedb, end_time): sqlitedb.update_data( 'accuracy', f"min_price = {row['min_price'].values[0]},max_price = {row['max_price'].values[0]},mean={row['mean'].values[0]}", f"id = {id}") except: - logger.error(f'更新accuracy表中的min_price,max_price,mean值失败，row={row}') + config.loggererror( + f'更新accuracy表中的min_price,max_price,mean值失败，row={row}') df = accuracy_df.copy() df['ds'] = pd.to_datetime(df['ds']) @@ -1256,7 +1259,7 @@ def model_losss(sqlitedb, end_time): except ValueError: pass df_combined3.to_csv(os.path.join( - dataset, "testandpredict_groupby.csv"), index=False) + config.dataset, "testandpredict_groupby.csv"), index=False) # 历史价格+预测价格 sqlitedb.drop_table('testandpredict_groupby') @@ -1301,7 +1304,8 @@ def model_losss(sqlitedb, end_time): plt.xticks(rotation=45) # 日期标签旋转45度，防止重叠 plt.ylabel('价格') - plt.savefig(os.path.join(dataset, '历史价格-预测值.png'), bbox_inches='tight') + plt.savefig(os.path.join(config.dataset, '历史价格-预测值.png'), + bbox_inches='tight') plt.close() def _plt_modeltopten_predict_ture(df): @@ -1340,7 +1344,7 @@ def model_losss(sqlitedb, end_time): plt.ylabel('价格') - plt.savefig(os.path.join(dataset, '历史价格-预测值1.png'), + plt.savefig(os.path.join(config.dataset, '历史价格-预测值1.png'), bbox_inches='tight') plt.close() @@ -1361,7 +1365,8 @@ def model_losss(sqlitedb, end_time): table.set_fontsize(10) # 设置表格样式，列数据最小的用绿色标识 - plt.savefig(os.path.join(dataset, '预测值表格.png'), bbox_inches='tight') + plt.savefig(os.path.join(config.dataset, '预测值表格.png'), + bbox_inches='tight') plt.close() def _plt_model_results3(): @@ -1376,7 +1381,8 @@ def model_losss(sqlitedb, end_time): table.set_fontsize(10) # 设置表格样式，列数据最小的用绿色标识 - plt.savefig(os.path.join(dataset, '模型评估.png'), bbox_inches='tight') + plt.savefig(os.path.join(config.dataset, '模型评估.png'), + bbox_inches='tight') plt.close() # _plt_predict_ture(df_combined3) @@ -2198,7 +2204,7 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in # print(f'绘制第{i+1}个特征{col}与价格散点图时出错：{e}') # 添加标题 - content.append(Graphs.draw_title(f'{y}{time}预测报告')) + content.append(Graphs.draw_title(f'{config.y}{time}预测报告')) # 预测结果 content.append(Graphs.draw_little_title('一、预测结果：')) @@ -2321,7 +2327,7 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in '''相关系数接近0：表示两个变量之间不存在线性关系，即它们的变化不会随着对方的变化而变化。''')) for name, group in grouped: cols = group['指标名称'].tolist() - logger.info(f'开始绘制{name}类指标的相关性直方图') + config.logger.info(f'开始绘制{name}类指标的相关性直方图') cols_subset = cols feature_names = ['y'] + cols_subset correlation_matrix = df_zhibiaoshuju[feature_names].corr()['y'] @@ -2376,7 +2382,7 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in draw_feature_trend(feature_data_df, negative_corr_features) # 计算correlation_sum 第一行的相关性的绝对值的总和 correlation_sum = correlation_matrix.abs().sum() - logger.info(f'{name}类指标的相关性总和为：{correlation_sum}') + config.logger.info(f'{name}类指标的相关性总和为：{correlation_sum}') # 分组的相关性总和拼接到grouped_corr goup_corr = pd.DataFrame( {'指标分类': [name], '指标数量': [len(cols_subset)], '相关性总和': [correlation_sum]}) @@ -2384,7 +2390,7 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in [grouped_corr, goup_corr], axis=0, ignore_index=True) # 绘制相关性总和的气泡图 - logger.info(f'开始绘制相关性总和的气泡图') + config.logger.info(f'开始绘制相关性总和的气泡图') plt.figure(figsize=(10, 10)) sns.scatterplot(data=grouped_corr, x='相关性总和', y='指标数量', size='相关性总和', sizes=( grouped_corr['相关性总和'].min()*5, grouped_corr['相关性总和'].max()*5), hue='指标分类', palette='viridis') @@ -2396,7 +2402,7 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in content.append(Graphs.draw_img(os.path.join(dataset, '指标分类相关性总和的气泡图.png'))) content.append(Graphs.draw_text( '气泡图中，横轴为指标分类，纵轴为指标分类下的特征数量，气泡的面积越大表示该分类中特征的相关系数和越大。')) - logger.info(f'绘制相关性总和的气泡图结束') + config.logger.info(f'绘制相关性总和的气泡图结束') content.append(Graphs.draw_little_title('模型选择：')) content.append(Graphs.draw_text( f'预测使用了{num_models}个模型进行训练拟合，通过评估指标MAE从小到大排列，前5个模型的简介如下：')) @@ -2441,7 +2447,7 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in doc.build(content) # pdf 上传到数字化信息平台 try: - if is_update_report: + if config.is_update_report: with open(os.path.join(dataset, reportname), 'rb') as f: base64_data = base64.b64encode(f.read()).decode('utf-8') upload_data["data"]["fileBase64"] = base64_data