liurui/PriceForecast
Template
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

聚烯烃月度逻辑添加图片结果

Browse Source
This commit is contained in:
workpc 2025-07-29 11:25:53 +08:00
parent 9647067f65
commit e44f3f7ecf
5 changed files with 240 additions and 209 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
config_juxiting_yuedu.py
View File

@ -88,6 +88,13 @@ bdwdname = [
'次三月', '次三月',
'次四月', '次四月',
] ]
# 数据库预测结果表八大维度列名
price_columns = [
'day_price', 'week_price', 'second_week_price', 'next_week_price',
'next_month_price', 'next_february_price', 'next_march_price', 'next_april_price'
]
modelsindex = [ modelsindex = [
{ {
"NHITS": "SELF0000275", "NHITS": "SELF0000275",
@ -406,8 +413,8 @@ bdwd_items = {
} }
# 报告中八大维度数据项重命名 # 报告中八大维度数据项重命名
columnsrename={'jxtppbdwdbz': '本周', 'jxtppbdwdcey': '次二月', 'jxtppbdwdcr': '次日', 'jxtppbdwdcsiy': '次四月', columnsrename = {'jxtppbdwdbz': '本周', 'jxtppbdwdcey': '次二月', 'jxtppbdwdcr': '次日', 'jxtppbdwdcsiy': '次四月',
'jxtppbdwdcsany': '次三月', 'jxtppbdwdcy': '次月', 'jxtppbdwdcz': '次周', 'jxtppbdwdgz': '隔周', } 'jxtppbdwdcsany': '次三月', 'jxtppbdwdcy': '次月', 'jxtppbdwdcz': '次周', 'jxtppbdwdgz': '隔周', }
# 北京环境数据库 # 北京环境数据库
host = '192.168.101.27' host = '192.168.101.27'
@ -459,7 +466,7 @@ print("数据库连接成功", host, dbname, dbusername)
# 数据截取日期 # 数据截取日期
start_year = 2000 # 数据开始年份 start_year = 2000 # 数据开始年份
end_time = '' # 数据截取日期 end_time = '2025-07-22' # 数据截取日期
freq = 'M' # 时间频率，"D": 天 "W": 周"M": 月"Q": 季度"A": 年 "H": 小时 "T": 分钟 "S": 秒 "B": 工作日 freq = 'M' # 时间频率，"D": 天 "W": 周"M": 月"Q": 季度"A": 年 "H": 小时 "T": 分钟 "S": 秒 "B": 工作日
delweekenday = True if freq == 'B' else False # 是否删除周末数据 delweekenday = True if freq == 'B' else False # 是否删除周末数据
is_corr = False # 特征是否参与滞后领先提升相关系数 is_corr = False # 特征是否参与滞后领先提升相关系数

26
lib/tools.py
View File

@ -864,8 +864,7 @@ def find_best_models(date='', global_config=None):
# 获取真实价格数据 # 获取真实价格数据
try: try:
true_price = pd.read_csv(os.path.join( true_price = pd.read_csv('juxitingdataset/指标数据.csv')[['ds', 'y']]
global_config['dataset'], '指标数据.csv'))[['ds', 'y']]
except FileNotFoundError: except FileNotFoundError:
global_config['logger'].error( global_config['logger'].error(
f"未找到文件: {os.path.join(global_config['dataset'], '指标数据.csv')}") f"未找到文件: {os.path.join(global_config['dataset'], '指标数据.csv')}")
@ -1082,23 +1081,31 @@ def plot_pp_predict_result(y_hat, global_config):
import seaborn as sns import seaborn as sns
# 获取y的真实值 # 获取y的真实值
y = pd.read_csv(os.path.join( # y = pd.read_csv(os.path.join(
global_config['dataset'], '指标数据.csv'))[['ds', 'y']] # global_config['dataset'], '指标数据.csv'))[['ds', 'y']]
y = pd.read_csv('juxitingdataset/指标数据.csv')[['ds', 'y']]
y['ds'] = pd.to_datetime(y['ds']) y['ds'] = pd.to_datetime(y['ds'])
y = y[y['ds'] < y_hat['ds'].iloc[0]][-30:] y = y[y['ds'] < y_hat['ds'].iloc[0]][-30:]
# 取y的最后一行数据追加到y_hat（将真实值最后一行作为预测值起点）
if not y.empty:
# 获取y的最后一行并将'y'列重命名为'predictresult'以匹配y_hat结构
y_last_row = y.tail(1).rename(columns={'y': 'predictresult'})
# 追加到y_hat
y_y_hat = pd.concat([y_last_row, y_hat], ignore_index=True)
# 创建图表和子图布局，为表格预留空间 # 创建图表和子图布局，为表格预留空间
fig, ax = plt.subplots(figsize=(16, 9)) fig, ax = plt.subplots(figsize=(16, 9))
# 对日期列进行排序，确保日期大的在右边 # 对日期列进行排序，确保日期大的在右边
y_hat = y_hat.sort_values(by='ds') y_y_hat = y_y_hat.sort_values(by='ds')
y = y.sort_values(by='ds') y = y.sort_values(by='ds')
# 绘制 y_hat 的折线图，颜色为橙色 # 绘制 y_hat 的折线图，颜色为橙色
sns.lineplot(x=y_hat['ds'], y=y_hat['predictresult'], sns.lineplot(x=y_y_hat['ds'], y=y_y_hat['predictresult'],
color='orange', label='y_hat', ax=ax, linestyle='--') color='orange', label='预测值', ax=ax, linestyle='--')
# 绘制 y 的折线图，颜色为蓝色 # 绘制 y 的折线图，颜色为蓝色
sns.lineplot(x=y['ds'], y=y['y'], color='blue', label='y', ax=ax) sns.lineplot(x=y['ds'], y=y['y'], color='blue', label='真实值', ax=ax)
# date_str = pd.Timestamp(y_hat["ds"].iloc[0]).strftime('%Y-%m-%d') # date_str = pd.Timestamp(y_hat["ds"].iloc[0]).strftime('%Y-%m-%d')
ax.set_title(f'{global_config["end_time"]} PP期货八大维度 预测价格走势') ax.set_title(f'{global_config["end_time"]} PP期货八大维度 预测价格走势')
@ -1129,7 +1136,8 @@ def plot_pp_predict_result(y_hat, global_config):
table.set_fontsize(14) table.set_fontsize(14)
plt.tight_layout(rect=[0, 0.1, 1, 1]) # 调整布局，为表格留出空间 plt.tight_layout(rect=[0, 0.1, 1, 1]) # 调整布局，为表格留出空间
plt.savefig('pp_predict_result.png') plt.savefig(os.path.join(
global_config['dataset'], 'pp_predict_result.png'))
if __name__ == '__main__': if __name__ == '__main__':

9
main_juxiting.py
View File

@ -554,14 +554,7 @@ if __name__ == '__main__':
# logger.info(f'预测失败：{e}') # logger.info(f'预测失败：{e}')
# continue # continue
# predict_main() predict_main()
# push_market_value() # push_market_value()
# sql_inset_predict(global_config) # sql_inset_predict(global_config)
from lib.tools import find_best_models
best_bdwd_price = find_best_models(
date='2025-07-22', global_config=global_config)
y_hat = pd.DataFrame(best_bdwd_price).T[['date', 'predictresult']]
y_hat['ds'] = pd.to_datetime(y_hat['date'])
# 绘制PP期货预测结果的图表
plot_pp_predict_result(y_hat, global_config)

379
main_juxiting_yuedu.py
View File

@ -3,7 +3,7 @@
from lib.dataread import * from lib.dataread import *
from config_juxiting_yuedu import * from config_juxiting_yuedu import *
from lib.tools import SendMail, convert_df_to_pydantic_pp, exception_logger, get_modelsname from lib.tools import SendMail, convert_df_to_pydantic_pp, exception_logger, get_modelsname
from models.nerulforcastmodels import ex_Model, model_losss_juxiting, pp_export_pdf from models.nerulforcastmodels import ex_Model, model_losss_juxiting, pp_bdwd_png, pp_export_pdf
import datetime import datetime
import torch import torch
torch.set_float32_matmul_precision("high") torch.set_float32_matmul_precision("high")
@ -24,6 +24,7 @@ global_config.update({
'settings': settings, 'settings': settings,
'bdwdname': bdwdname, 'bdwdname': bdwdname,
'columnsrename': columnsrename, 'columnsrename': columnsrename,
'price_columns': price_columns,
# 模型参数 # 模型参数
@ -291,210 +292,218 @@ def predict_main():
返回: 返回:
None None
""" """
# end_time = global_config['end_time'] end_time = global_config['end_time']
# signature = BinanceAPI(APPID, SECRET) signature = BinanceAPI(APPID, SECRET)
# etadata = EtaReader(signature=signature, etadata = EtaReader(signature=signature,
# classifylisturl=global_config['classifylisturl'], classifylisturl=global_config['classifylisturl'],
# classifyidlisturl=global_config['classifyidlisturl'], classifyidlisturl=global_config['classifyidlisturl'],
# edbcodedataurl=global_config['edbcodedataurl'], edbcodedataurl=global_config['edbcodedataurl'],
# edbcodelist=global_config['edbcodelist'], edbcodelist=global_config['edbcodelist'],
# edbdatapushurl=global_config['edbdatapushurl'], edbdatapushurl=global_config['edbdatapushurl'],
# edbdeleteurl=global_config['edbdeleteurl'], edbdeleteurl=global_config['edbdeleteurl'],
# edbbusinessurl=global_config['edbbusinessurl'], edbbusinessurl=global_config['edbbusinessurl'],
# classifyId=global_config['ClassifyId'], classifyId=global_config['ClassifyId'],
# ) )
# # 获取数据 # 获取数据
# if is_eta: if is_eta:
# logger.info('从eta获取数据...') logger.info('从eta获取数据...')
# df_zhibiaoshuju, df_zhibiaoliebiao = etadata.get_eta_api_pp_data( df_zhibiaoshuju, df_zhibiaoliebiao = etadata.get_eta_api_pp_data(
# data_set=data_set, dataset=dataset) # 原始数据，未处理 data_set=data_set, dataset=dataset) # 原始数据，未处理
# if is_market: if is_market:
# logger.info('从市场信息平台获取数据...') logger.info('从市场信息平台获取数据...')
# try: try:
# # 如果是测试环境，最高价最低价取excel文档 # 如果是测试环境，最高价最低价取excel文档
# if server_host == '192.168.100.53': if server_host == '192.168.100.53':
# logger.info('从excel文档获取最高价最低价') logger.info('从excel文档获取最高价最低价')
# df_zhibiaoshuju = get_high_low_data(df_zhibiaoshuju) df_zhibiaoshuju = get_high_low_data(df_zhibiaoshuju)
# else: else:
# logger.info('从市场信息平台获取数据') logger.info('从市场信息平台获取数据')
# df_zhibiaoshuju = get_market_data( df_zhibiaoshuju = get_market_data(
# end_time, df_zhibiaoshuju) end_time, df_zhibiaoshuju)
# except: except:
# logger.info('最高最低价拼接失败') logger.info('最高最低价拼接失败')
# # 保存到xlsx文件的sheet表 # 保存到xlsx文件的sheet表
# with pd.ExcelWriter(os.path.join(dataset, data_set)) as file: 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_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) df_zhibiaoliebiao.to_excel(file, sheet_name='指标列表', index=False)
# # 数据处理 # 数据处理
# df = datachuli_juxiting(df_zhibiaoshuju, df_zhibiaoliebiao, y=global_config['y'], dataset=dataset, add_kdj=add_kdj, is_timefurture=is_timefurture, df = datachuli_juxiting(df_zhibiaoshuju, df_zhibiaoliebiao, y=global_config['y'], dataset=dataset, add_kdj=add_kdj, is_timefurture=is_timefurture,
# end_time=end_time) end_time=end_time)
# else: else:
# # 读取数据 # 读取数据
# logger.info('读取本地数据：' + os.path.join(dataset, data_set)) logger.info('读取本地数据：' + os.path.join(dataset, data_set))
# df, df_zhibiaoliebiao = getdata_zhoudu_juxiting(filename=os.path.join(dataset, data_set), y=y, dataset=dataset, add_kdj=add_kdj, df, df_zhibiaoliebiao = getdata_zhoudu_juxiting(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 is_timefurture=is_timefurture, end_time=end_time) # 原始数据，未处理
# # 更改预测列名称 # 更改预测列名称
# df.rename(columns={y: 'y'}, inplace=True) df.rename(columns={y: 'y'}, inplace=True)
# if is_edbnamelist: if is_edbnamelist:
# df = df[edbnamelist] df = df[edbnamelist]
# df.to_csv(os.path.join(dataset, '指标数据.csv'), index=False) 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 # 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 # 取第一行数据存储到数据库中
# first_row = df[['ds', 'y']].tail(1) first_row = df[['ds', 'y']].tail(1)
# # 判断y的类型是否为float # 判断y的类型是否为float
# if not isinstance(first_row['y'].values[0], 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') logger.info(f'{end_time}预测目标数据为空，跳过')
# return None return None
# # 将最新真实值保存到数据库 # 将最新真实值保存到数据库
# if not sqlitedb.check_table_exists('trueandpredict'): if not sqlitedb.check_table_exists('trueandpredict'):
# first_row.to_sql('trueandpredict', sqlitedb.connection, index=False) first_row.to_sql('trueandpredict', sqlitedb.connection, index=False)
# else: else:
# for row in first_row.itertuples(index=False): for row in first_row.itertuples(index=False):
# row_dict = row._asdict() row_dict = row._asdict()
# config.logger.info(f'要保存的真实值：{row_dict}') 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 # 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)): if isinstance(row_dict['ds'], (pd.Timestamp, datetime.datetime)):
# row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d') row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d')
# elif not isinstance(row_dict['ds'], str): elif not isinstance(row_dict['ds'], str):
# try: try:
# row_dict['ds'] = pd.to_datetime( row_dict['ds'] = pd.to_datetime(
# row_dict['ds']).strftime('%Y-%m-%d') row_dict['ds']).strftime('%Y-%m-%d')
# except: except:
# logger.warning(f"无法解析的时间格式: {row_dict['ds']}") 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')
# # row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d %H:%M:%S') # row_dict['ds'] = row_dict['ds'].strftime('%Y-%m-%d %H:%M:%S')
# check_query = sqlitedb.select_data( check_query = sqlitedb.select_data(
# 'trueandpredict', where_condition=f"ds = '{row.ds}'") 'trueandpredict', where_condition=f"ds = '{row.ds}'")
# if len(check_query) > 0: if len(check_query) > 0:
# set_clause = ", ".join( set_clause = ", ".join(
# [f"{key} = '{value}'" for key, value in row_dict.items()]) [f"{key} = '{value}'" for key, value in row_dict.items()])
# sqlitedb.update_data( sqlitedb.update_data(
# 'trueandpredict', set_clause, where_condition=f"ds = '{row.ds}'") 'trueandpredict', set_clause, where_condition=f"ds = '{row.ds}'")
# continue continue
# sqlitedb.insert_data('trueandpredict', tuple( sqlitedb.insert_data('trueandpredict', tuple(
# row_dict.values()), columns=row_dict.keys()) row_dict.values()), columns=row_dict.keys())
# # 更新accuracy表的y值 # 更新accuracy表的y值
# if not sqlitedb.check_table_exists('accuracy'): if not sqlitedb.check_table_exists('accuracy'):
# pass pass
# else: else:
# update_y = sqlitedb.select_data( update_y = sqlitedb.select_data(
# 'accuracy', where_condition="y is null") 'accuracy', where_condition="y is null")
# if len(update_y) > 0: if len(update_y) > 0:
# logger.info('更新accuracy表的y值') logger.info('更新accuracy表的y值')
# # 找到update_y 中ds且df中的y的行 # 找到update_y 中ds且df中的y的行
# update_y = update_y[update_y['ds'] <= end_time] update_y = update_y[update_y['ds'] <= end_time]
# logger.info(f'要更新y的信息：{update_y}') logger.info(f'要更新y的信息：{update_y}')
# # try: # try:
# for row in update_y.itertuples(index=False): for row in update_y.itertuples(index=False):
# try: try:
# row_dict = row._asdict() row_dict = row._asdict()
# yy = df[df['ds'] == row_dict['ds']]['y'].values[0] yy = df[df['ds'] == row_dict['ds']]['y'].values[0]
# LOW = df[df['ds'] == row_dict['ds']]['Brentzdj'].values[0] LOW = df[df['ds'] == row_dict['ds']]['Brentzdj'].values[0]
# HIGH = df[df['ds'] == row_dict['ds']]['Brentzgj'].values[0] HIGH = df[df['ds'] == row_dict['ds']]['Brentzgj'].values[0]
# sqlitedb.update_data( sqlitedb.update_data(
# 'accuracy', f"y = {yy},LOW_PRICE = {LOW},HIGH_PRICE = {HIGH}", where_condition=f"ds = '{row_dict['ds']}'") 'accuracy', f"y = {yy},LOW_PRICE = {LOW},HIGH_PRICE = {HIGH}", where_condition=f"ds = '{row_dict['ds']}'")
# except: except:
# logger.info(f'更新accuracy表的y值失败：{row_dict}') logger.info(f'更新accuracy表的y值失败：{row_dict}')
# # except Exception as e: # except Exception as e:
# # logger.info(f'更新accuracy表的y值失败：{e}') # logger.info(f'更新accuracy表的y值失败：{e}')
# # 判断当前日期是不是周一 # 判断当前日期是不是周一
# is_weekday = datetime.datetime.now().weekday() == 0 is_weekday = datetime.datetime.now().weekday() == 0
# if is_weekday: if is_weekday:
# logger.info('今天是周一，更新预测模型') 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 # 计算最近60天预测残差最低的模型名称
# model_results = sqlitedb.select_data( model_results = sqlitedb.select_data(
# 'trueandpredict', order_by="ds DESC", limit="60") 'trueandpredict', order_by="ds DESC", limit="60")
# # 删除空值率为90%以上的列 # 删除空值率为90%以上的列
# if len(model_results) > 10: if len(model_results) > 10:
# model_results = model_results.dropna( model_results = model_results.dropna(
# thresh=len(model_results)*0.1, axis=1) thresh=len(model_results)*0.1, axis=1)
# # 删除空行 # 删除空行
# model_results = model_results.dropna() model_results = model_results.dropna()
# modelnames = model_results.columns.to_list()[2:-1] modelnames = model_results.columns.to_list()[2:-1]
# for col in model_results[modelnames].select_dtypes(include=['object']).columns: for col in model_results[modelnames].select_dtypes(include=['object']).columns:
# model_results[col] = model_results[col].astype(np.float32) 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 # 计算每个预测值与真实值之间的偏差率
# for model in modelnames: for model in modelnames:
# model_results[f'{model}_abs_error_rate'] = abs( model_results[f'{model}_abs_error_rate'] = abs(
# model_results['y'] - model_results[model]) / model_results['y'] model_results['y'] - model_results[model]) / model_results['y']
# # 获取每行对应的最小偏差率值 # 获取每行对应的最小偏差率值
# min_abs_error_rate_values = model_results.apply( min_abs_error_rate_values = model_results.apply(
# lambda row: row[[f'{model}_abs_error_rate' for model in modelnames]].min(), axis=1) 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 # 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( min_abs_error_rate_column_name = model_results.apply(
# lambda row: row[[f'{model}_abs_error_rate' for model in modelnames]].idxmin(), axis=1) 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 # 将列名索引转换为列名
# min_abs_error_rate_column_name = min_abs_error_rate_column_name.map( min_abs_error_rate_column_name = min_abs_error_rate_column_name.map(
# lambda x: x.split('_')[0]) lambda x: x.split('_')[0])
# # 取出现次数最多的模型名称 # 取出现次数最多的模型名称
# most_common_model = min_abs_error_rate_column_name.value_counts().idxmax() 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}") 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 # 保存结果到数据库
# if not sqlitedb.check_table_exists('most_model'): if not sqlitedb.check_table_exists('most_model'):
# sqlitedb.create_table( sqlitedb.create_table(
# 'most_model', columns="ds datetime, most_common_model TEXT") 'most_model', columns="ds datetime, most_common_model TEXT")
# sqlitedb.insert_data('most_model', (datetime.datetime.now().strftime( sqlitedb.insert_data('most_model', (datetime.datetime.now().strftime(
# '%Y-%m-%d %H:%M:%S'), most_common_model,), columns=('ds', 'most_common_model',)) '%Y-%m-%d %H:%M:%S'), most_common_model,), columns=('ds', 'most_common_model',))
# if is_corr: if is_corr:
# df = corr_feature(df=df) 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 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...") logger.info(f"开始训练模型...")
# row, col = df.shape row, col = df.shape
# now = datetime.datetime.now().strftime('%Y%m%d%H%M%S') now = datetime.datetime.now().strftime('%Y%m%d%H%M%S')
# ex_Model(df, ex_Model(df,
# horizon=global_config['horizon'], horizon=global_config['horizon'],
# input_size=global_config['input_size'], input_size=global_config['input_size'],
# train_steps=global_config['train_steps'], train_steps=global_config['train_steps'],
# val_check_steps=global_config['val_check_steps'], val_check_steps=global_config['val_check_steps'],
# early_stop_patience_steps=global_config['early_stop_patience_steps'], early_stop_patience_steps=global_config['early_stop_patience_steps'],
# is_debug=global_config['is_debug'], is_debug=global_config['is_debug'],
# dataset=global_config['dataset'], dataset=global_config['dataset'],
# is_train=global_config['is_train'], is_train=global_config['is_train'],
# is_fivemodels=global_config['is_fivemodels'], is_fivemodels=global_config['is_fivemodels'],
# val_size=global_config['val_size'], val_size=global_config['val_size'],
# test_size=global_config['test_size'], test_size=global_config['test_size'],
# settings=global_config['settings'], settings=global_config['settings'],
# now=now, now=now,
# etadata=etadata, etadata=etadata,
# modelsindex=global_config['modelsindex'], modelsindex=global_config['modelsindex'],
# data=data, data=data,
# is_eta=global_config['is_eta'], is_eta=global_config['is_eta'],
# end_time=global_config['end_time'], 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') logger.info('训练数据绘图ing')
# model_results3 = model_losss_juxiting(sqlitedb, end_time=global_config['end_time'],is_fivemodels=global_config['is_fivemodels']) model_results3 = model_losss_juxiting(
# logger.info('训练数据绘图end') sqlitedb, end_time=global_config['end_time'], is_fivemodels=global_config['is_fivemodels'])
logger.info('训练数据绘图end')
# push_market_value() push_market_value()
# # 模型报告
# logger.info('制作报告ing')
# title = f'{settings}--{end_time}-预测报告' # 报告标题
# reportname = f'聚烯烃PP大模型月度预测--{end_time}.pdf' # 报告文件名
# reportname = reportname.replace(':', '-') # 替换冒号
# pp_export_pdf(dataset=dataset, num_models=5 if is_fivemodels else 22, time=end_time,
# reportname=reportname, sqlitedb=sqlitedb),
# logger.info('制作报告end')
# logger.info('模型训练完成')
sql_inset_predict(global_config) sql_inset_predict(global_config)
模型报告
logger.info('制作报告ing')
title = f'{settings}--{end_time}-预测报告' # 报告标题
reportname = f'聚烯烃PP大模型月度预测--{end_time}.pdf' # 报告文件名
reportname = reportname.replace(':', '-') # 替换冒号
pp_export_pdf(dataset=dataset, num_models=5 if is_fivemodels else 22, time=end_time,
reportname=reportname, sqlitedb=sqlitedb),
logger.info('制作报告end')
logger.info('模型训练完成')
# 图片报告
logger.info('图片报告ing')
pp_bdwd_png(global_config=global_config)
logger.info('图片报告end')
# # LSTM 单变量模型 # # LSTM 单变量模型
# ex_Lstm(df,input_seq_len=input_size,output_seq_len=horizon,is_debug=is_debug,dataset=dataset) # ex_Lstm(df,input_seq_len=input_size,output_seq_len=horizon,is_debug=is_debug,dataset=dataset)

22
models/nerulforcastmodels.py
View File

@ -6,7 +6,7 @@ import seaborn as sns
import matplotlib.pyplot as plt import matplotlib.pyplot as plt
import matplotlib.dates as mdates import matplotlib.dates as mdates
import datetime import datetime
from lib.tools import Graphs, mse, rmse, mae, exception_logger from lib.tools import Graphs, find_best_models, mse, plot_pp_predict_result, rmse, mae, exception_logger
from lib.tools import save_to_database, get_week_date from lib.tools import save_to_database, get_week_date
from lib.dataread import * from lib.dataread import *
from neuralforecast import NeuralForecast from neuralforecast import NeuralForecast
@ -165,7 +165,8 @@ def ex_Model(df, horizon, input_size, train_steps, val_check_steps, early_stop_p
# VanillaTransformer(h=horizon, input_size=input_size, max_steps=train_steps, val_check_steps=val_check_steps, scaler_type='standard', ), //报错了 # VanillaTransformer(h=horizon, input_size=input_size, max_steps=train_steps, val_check_steps=val_check_steps, scaler_type='standard', ), //报错了
# Autoformer(h=horizon, input_size=input_size, max_steps=train_steps, val_check_steps=val_check_steps, scaler_type='standard', ), //报错了 # Autoformer(h=horizon, input_size=input_size, max_steps=train_steps, val_check_steps=val_check_steps, scaler_type='standard', ), //报错了
NBEATS(h=horizon, input_size=input_size, max_steps=train_steps, val_check_steps=val_check_steps, scaler_type='standard', ), NBEATS(h=horizon, input_size=input_size, max_steps=train_steps,
val_check_steps=val_check_steps, scaler_type='standard', ),
# NBEATSx(h=horizon, input_size=input_size, max_steps=train_steps, val_check_steps=val_check_steps, scaler_type='standard',activation='ReLU', ), //报错 # NBEATSx(h=horizon, input_size=input_size, max_steps=train_steps, val_check_steps=val_check_steps, scaler_type='standard',activation='ReLU', ), //报错
# HINT(h=horizon), # HINT(h=horizon),
@ -2359,7 +2360,8 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in
stime = df3['ds'].iloc[0] stime = df3['ds'].iloc[0]
etime = df3['ds'].iloc[-1] etime = df3['ds'].iloc[-1]
# 添加偏差率表格 # 添加偏差率表格
fivemodels = '、'.join(eval_df['模型(Model)'].values[:5]) # 字符串形式，后面写入字符串使用 fivemodels = '、'.join(
eval_df['模型(Model)'].values[:5]) # 字符串形式，后面写入字符串使用
content.append(Graphs.draw_text( content.append(Graphs.draw_text(
f'预测使用了{num_models}个模型进行训练，使用评估结果MAE前五的模型分别是 {fivemodels} ，模型上一预测区间 {stime} -- {etime}的偏差率(%)分别是：')) f'预测使用了{num_models}个模型进行训练，使用评估结果MAE前五的模型分别是 {fivemodels} ，模型上一预测区间 {stime} -- {etime}的偏差率(%)分别是：'))
# # 添加偏差率表格 # # 添加偏差率表格
@ -2370,7 +2372,8 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in
content.append(Graphs.draw_table(col_width, *data)) content.append(Graphs.draw_table(col_width, *data))
content.append(Graphs.draw_little_title('上一周预测准确率：')) content.append(Graphs.draw_little_title('上一周预测准确率：'))
df4 = sqlitedb.select_data('accuracy_rote', order_by='结束日期 desc', limit=1) df4 = sqlitedb.select_data(
'accuracy_rote', order_by='结束日期 desc', limit=1)
df4 = df4.T df4 = df4.T
df4 = df4.reset_index() df4 = df4.reset_index()
df4 = df4.T df4 = df4.T
@ -3524,6 +3527,17 @@ def pp_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, input
print(f"请求超时: {e}") print(f"请求超时: {e}")
@exception_logger
def pp_bdwd_png(global_config):
best_bdwd_price = find_best_models(
date=global_config['end_time'], global_config=global_config)
# y_hat = pd.DataFrame(best_bdwd_price).T[['date', 'predictresult']][-4:]
y_hat = pd.DataFrame(best_bdwd_price).T[['date', 'predictresult']]
y_hat['ds'] = pd.to_datetime(y_hat['date'])
# 绘制PP期货预测结果的图表
plot_pp_predict_result(y_hat, global_config)
def pp_export_pdf_v1(num_indicators=475, num_models=21, num_dayindicator=202, inputsize=5, dataset='dataset', time='2024-07-30', reportname='report.pdf'): def pp_export_pdf_v1(num_indicators=475, num_models=21, num_dayindicator=202, inputsize=5, dataset='dataset', time='2024-07-30', reportname='report.pdf'):
global y global y
# 创建内容对应的空列表 # 创建内容对应的空列表