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

石油焦日度预测调试

Browse Source
This commit is contained in:
jingboyitiji 2025-03-20 16:29:25 +08:00
parent 313e9e229d
commit 7cf3cda87a
2 changed files with 533 additions and 220 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

420
main_shiyoujiao_lvyong.py
View File

@ -3,7 +3,7 @@
from lib.dataread import * from lib.dataread import *
from config_shiyoujiao_lvyong import * from config_shiyoujiao_lvyong import *
from lib.tools import SendMail, exception_logger 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 from models.nerulforcastmodels import model_losss, shiyoujiao_lvyong_export_pdf
import datetime import datetime
import torch import torch
torch.set_float32_matmul_precision("high") torch.set_float32_matmul_precision("high")
@ -173,228 +173,228 @@ def predict_main():
返回: 返回:
None None
""" """
end_time = global_config['end_time'] # end_time = global_config['end_time']
# 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...')
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'],
) # )
df_zhibiaoshuju, df_zhibiaoliebiao = etadata.get_eta_api_shiyoujiao_lvyong_data( # df_zhibiaoshuju, df_zhibiaoliebiao = etadata.get_eta_api_shiyoujiao_lvyong_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(df_zhibiaoshuju, df_zhibiaoliebiao, y=global_config['y'], dataset=dataset, add_kdj=add_kdj, is_timefurture=is_timefurture, # df = datachuli(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(filename=os.path.join(dataset, data_set), y=y, dataset=dataset, add_kdj=add_kdj, # 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 # 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',))
try: # try:
if is_weekday: # if is_weekday:
# if True: # # if True:
logger.info('今天是周一，发送特征预警') # logger.info('今天是周一，发送特征预警')
# 上传预警信息到数据库 # # 上传预警信息到数据库
warning_data_df = df_zhibiaoliebiao.copy() # warning_data_df = df_zhibiaoliebiao.copy()
warning_data_df = warning_data_df[warning_data_df['停更周期'] > 3][[ # 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']] # '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 # # 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', # 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'}) # '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 # from sqlalchemy import create_engine
import urllib # import urllib
global password # global password
if '@' in password: # if '@' in password:
password = urllib.parse.quote_plus(password) # password = urllib.parse.quote_plus(password)
engine = create_engine( # engine = create_engine(
f'mysql+pymysql://{dbusername}:{password}@{host}:{port}/{dbname}') # f'mysql+pymysql://{dbusername}:{password}@{host}:{port}/{dbname}')
warning_data_df['WARNING_DATE'] = datetime.date.today().strftime( # warning_data_df['WARNING_DATE'] = datetime.date.today().strftime(
"%Y-%m-%d %H:%M:%S") # "%Y-%m-%d %H:%M:%S")
warning_data_df['TENANT_CODE'] = 'T0004' # warning_data_df['TENANT_CODE'] = 'T0004'
# 插入数据之前查询表数据然后新增id列 # # 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) # existing_data = pd.read_sql(f"SELECT * FROM {table_name}", engine)
if not existing_data.empty: # if not existing_data.empty:
max_id = existing_data['ID'].astype(int).max() # max_id = existing_data['ID'].astype(int).max()
warning_data_df['ID'] = range( # warning_data_df['ID'] = range(
max_id + 1, max_id + 1 + len(warning_data_df)) # max_id + 1, max_id + 1 + len(warning_data_df))
else: # else:
warning_data_df['ID'] = range(1, 1 + len(warning_data_df)) # warning_data_df['ID'] = range(1, 1 + len(warning_data_df))
warning_data_df.to_sql( # warning_data_df.to_sql(
table_name, con=engine, if_exists='append', index=False) # table_name, con=engine, if_exists='append', index=False)
if is_update_warning_data: # if is_update_warning_data:
upload_warning_info(len(warning_data_df)) # upload_warning_info(len(warning_data_df))
except: # except:
logger.info('上传预警信息到数据库失败') # logger.info('上传预警信息到数据库失败')
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=global_config['etadata'], # etadata=global_config['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(sqlitedb, end_time=end_time) model_results3 = model_losss(sqlitedb, end_time=end_time)
@ -403,15 +403,15 @@ def predict_main():
# 模型报告 # 模型报告
logger.info('制作报告ing') logger.info('制作报告ing')
title = f'{settings}--{end_time}-预测报告' # 报告标题 title = f'{settings}--{end_time}-预测报告' # 报告标题
reportname = f'Brent原油大模型日度预测--{end_time}.pdf' # 报告文件名 reportname = f'石油焦铝用大模型日度预测--{end_time}.pdf' # 报告文件名
reportname = reportname.replace(':', '-') # 替换冒号 reportname = reportname.replace(':', '-') # 替换冒号
brent_export_pdf(dataset=dataset, num_models=5 if is_fivemodels else 22, time=end_time, shiyoujiao_lvyong_export_pdf(dataset=dataset, num_models=5 if is_fivemodels else 22, time=end_time,
reportname=reportname, sqlitedb=sqlitedb), reportname=reportname, sqlitedb=sqlitedb),
logger.info('制作报告end') logger.info('制作报告end')
logger.info('模型训练完成') logger.info('模型训练完成')
push_market_value() # push_market_value()
# # 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)

333
models/nerulforcastmodels.py
View File

@ -866,7 +866,7 @@ def model_losss_yongan(sqlitedb, end_time, table_name_prefix):
plt.text(i, j, str(j), ha='center', va='bottom') plt.text(i, j, str(j), ha='center', va='bottom')
# 当前日期画竖虚线 # 当前日期画竖虚线
plt.axvline(x=df['ds'].iloc[-horizon], color='r', linestyle='--') plt.axvline(x=df['ds'].iloc[-config.horizon], color='r', linestyle='--')
plt.legend() plt.legend()
plt.xlabel('日期') plt.xlabel('日期')
plt.ylabel('价格') plt.ylabel('价格')
@ -881,8 +881,8 @@ def model_losss_yongan(sqlitedb, end_time, table_name_prefix):
ax.axis('off') # 关闭坐标轴 ax.axis('off') # 关闭坐标轴
# 数值保留2位小数 # 数值保留2位小数
df = df.round(2) df = df.round(2)
df = df[-horizon:] df = df[-config.horizon:]
df['Day'] = [f'Day_{i}' for i in range(1, horizon+1)] df['Day'] = [f'Day_{i}' for i in range(1, config.horizon+1)]
# Day列放到最前面 # Day列放到最前面
df = df[['Day'] + list(df.columns[:-1])] df = df[['Day'] + list(df.columns[:-1])]
table = ax.table(cellText=df.values, table = ax.table(cellText=df.values,
@ -1297,7 +1297,7 @@ def model_losss(sqlitedb, end_time):
# plt.plot(df['ds'], df[model], label=model,marker='o') # plt.plot(df['ds'], df[model], label=model,marker='o')
plt.plot(df['ds'], df[most_model_name], label=model, marker='o') plt.plot(df['ds'], df[most_model_name], label=model, marker='o')
# 当前日期画竖虚线 # 当前日期画竖虚线
plt.axvline(x=df['ds'].iloc[-horizon], color='r', linestyle='--') plt.axvline(x=df['ds'].iloc[-config.horizon], color='r', linestyle='--')
plt.legend() plt.legend()
plt.xlabel('日期') plt.xlabel('日期')
# 设置横轴日期格式为年-月-日 # 设置横轴日期格式为年-月-日
@ -1338,7 +1338,7 @@ def model_losss(sqlitedb, end_time):
plt.text(i, j, str(j), ha='center', va='bottom') plt.text(i, j, str(j), ha='center', va='bottom')
# 当前日期画竖虚线 # 当前日期画竖虚线
plt.axvline(x=df['ds'].iloc[-horizon], color='r', linestyle='--') plt.axvline(x=df['ds'].iloc[-config.horizon], color='r', linestyle='--')
plt.legend() plt.legend()
plt.xlabel('日期') plt.xlabel('日期')
# 自动设置横轴日期显示 # 自动设置横轴日期显示
@ -1357,8 +1357,8 @@ def model_losss(sqlitedb, end_time):
ax.axis('off') # 关闭坐标轴 ax.axis('off') # 关闭坐标轴
# 数值保留2位小数 # 数值保留2位小数
df = df.round(2) df = df.round(2)
df = df[-horizon:] df = df[-config.horizon:]
df['Day'] = [f'Day_{i}' for i in range(1, horizon+1)] df['Day'] = [f'Day_{i}' for i in range(1, config.horizon+1)]
# Day列放到最前面 # Day列放到最前面
df = df[['Day'] + list(df.columns[:-1])] df = df[['Day'] + list(df.columns[:-1])]
table = ax.table(cellText=df.values, table = ax.table(cellText=df.values,
@ -1388,10 +1388,10 @@ def model_losss(sqlitedb, end_time):
bbox_inches='tight') bbox_inches='tight')
plt.close() plt.close()
# _plt_predict_ture(df_combined3) _plt_predict_ture(df_combined3)
# _plt_modeltopten_predict_ture(df_combined4) # _plt_modeltopten_predict_ture(df_combined4)
# _plt_predict_table(df_combined3) _plt_predict_table(df_combined3)
# _plt_model_results3() _plt_model_results3()
return model_results3 return model_results3
@ -2461,6 +2461,319 @@ def brent_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, in
print(f"请求超时: {e}") print(f"请求超时: {e}")
@exception_logger
def shiyoujiao_lvyong_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, inputsize=5, dataset='dataset', time='2024-07-30', reportname='report.pdf', sqlitedb='jbsh_yuanyou.db'):
global y
# 创建内容对应的空列表
content = list()
# 获取特征的近一月值
import pandas as pd
feature_data_df = pd.read_csv(os.path.join(
config.dataset,'指标数据添加时间特征.csv'), parse_dates=['ds']).tail(60)
def draw_feature_trend(feature_data_df, features):
# 画特征近60天的趋势图
feature_df = feature_data_df[['ds', 'y']+features]
# 遍历X每一列，和yy画散点图 ，
for i, col in enumerate(features):
# try:
print(f'正在绘制第{i+1}个特征{col}与价格散点图...')
if col not in ['ds', 'y']:
fig, ax1 = plt.subplots(figsize=(10, 6))
# 在第一个坐标轴上绘制数据
sns.lineplot(data=feature_df, x='ds', y='y', ax=ax1, color='b')
ax1.set_xlabel('日期')
ax1.set_ylabel('y', color='b')
ax1.tick_params('y', colors='b')
# 在 ax1 上添加文本显示值，添加一定的偏移避免值与曲线重叠
for j in range(1, len(feature_df), 2):
value = feature_df['y'].iloc[j]
date = feature_df['ds'].iloc[j]
offset = 1.001
ax1.text(date, value * offset, str(round(value, 2)),
ha='center', va='bottom', color='b', fontsize=10)
# 创建第二个坐标轴
ax2 = ax1.twinx()
# 在第二个坐标轴上绘制数据
sns.lineplot(data=feature_df, x='ds', y=col, ax=ax2, color='r')
ax2.set_ylabel(col, color='r')
ax2.tick_params('y', colors='r')
# 在 ax2 上添加文本显示值，添加一定的偏移避免值与曲线重叠
for j in range(0, len(feature_df), 2):
value = feature_df[col].iloc[j]
date = feature_df['ds'].iloc[j]
offset = 1.0003
ax2.text(date, value * offset, str(round(value, 2)),
ha='center', va='bottom', color='r', fontsize=10)
# 添加标题
plt.title(col)
# 设置横坐标为日期格式并自动调整
locator = mdates.AutoDateLocator()
formatter = mdates.AutoDateFormatter(locator)
ax1.xaxis.set_major_locator(locator)
ax1.xaxis.set_major_formatter(formatter)
# 文件名特殊字符处理
col = col.replace('*', '-')
col = col.replace(':', '-')
col = col.replace(r'/', '-')
plt.savefig(os.path.join(config.dataset, f'{col}与价格散点图.png'))
content.append(Graphs.draw_img(
os.path.join(config.dataset, f'{col}与价格散点图.png')))
plt.close()
# except Exception as e:
# 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'{config.y}{time}预测报告'))
# 预测结果
content.append(Graphs.draw_little_title('一、预测结果：'))
# 添加历史走势及预测价格的走势图片
content.append(Graphs.draw_img(os.path.join(config.dataset, '历史价格-预测值.png')))
# 波动率画图逻辑
content.append(Graphs.draw_text('图示说明：'))
content.append(Graphs.draw_text(
' 确定置信区间：设置残差置信阈值，以每周最佳模型为基准，选取在置信区间的预测值作为置信区间;'))
# 取df中y列为空的行
import pandas as pd
df = pd.read_csv(os.path.join(config.dataset, 'predict.csv'), encoding='gbk')
df_true = pd.read_csv(os.path.join(
config.dataset,'指标数据添加时间特征.csv'), encoding='utf-8') # 获取预测日期对应的真实值
df_true = df_true[['ds', 'y']]
eval_df = pd.read_csv(os.path.join(
config.dataset,'model_evaluation.csv'), encoding='utf-8')
# 按评估指标排序，取前五
fivemodels_list = eval_df['模型(Model)'].values # 列表形式，后面当作列名索引使用
# 取 fivemodels_list 和 ds 列
df = df[['ds'] + fivemodels_list.tolist()]
# 拼接预测日期对应的真实值
df = pd.merge(df, df_true, on='ds', how='left')
# 删除全部为nan的列
df = df.dropna(how='all', axis=1)
# 选择除 'ds' 列外的数值列，并进行类型转换和四舍五入
num_cols = [col for col in df.columns if col !=
'ds' and pd.api.types.is_numeric_dtype(df[col])]
for col in num_cols:
df[col] = df[col].astype(float).round(2)
# 添加最大值、最小值、平均值三列
df['平均值'] = df[num_cols].mean(axis=1).round(2)
df['最大值'] = df[num_cols].max(axis=1)
df['最小值'] = df[num_cols].min(axis=1)
# df转置
df = df.T
# df重置索引
df = df.reset_index()
# 添加预测值表格
data = df.values.tolist()
col_width = 500/len(df.columns)
content.append(Graphs.draw_table(col_width, *data))
content.append(Graphs.draw_little_title('二、上一预测周期偏差率分析：'))
df = pd.read_csv(os.path.join(
config.dataset,'testandpredict_groupby.csv'), encoding='utf-8')
df4 = df.copy() # 计算偏差率使用
# 去掉created_dt 列
df4 = df4.drop(columns=['created_dt'])
# 计算模型偏差率
# 计算各列对于y列的差值百分比
df3 = pd.DataFrame() # 存储偏差率
# 删除有null的行
df4 = df4.dropna()
df3['ds'] = df4['ds']
for col in fivemodels_list:
df3[col] = round(abs(df4[col] - df4['y']) / df4['y'] * 100, 2)
# 找出决定系数前五的偏差率
df3 = df3[['ds']+fivemodels_list.tolist()][-inputsize:]
# 找出上一预测区间的时间
stime = df3['ds'].iloc[0]
etime = df3['ds'].iloc[-1]
# 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
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：'))
# # 添加偏差率表格
df3 = df3.T
df3 = df3.reset_index()
data = df3.values.tolist()
col_width = 500/len(df3.columns)
content.append(Graphs.draw_table(col_width, *data))
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 = df4.T
df4 = df4.reset_index()
df4 = df4.T
data = df4.values.tolist()
col_width = 500/len(df4.columns)
content.append(Graphs.draw_table(col_width, *data))
content.append(Graphs.draw_little_title('三、预测过程解析：'))
# 特征、模型、参数配置
content.append(Graphs.draw_little_title('模型选择：'))
content.append(Graphs.draw_text(
f'本次预测使用了一个专门收集时间序列的NeuralForecast库中的{num_models}个模型：'))
content.append(Graphs.draw_text(f'使用40天的数据预测未来{inputsize}天的数据。'))
content.append(Graphs.draw_little_title('指标情况：'))
with open(os.path.join(config.dataset, '特征频度统计.txt'), encoding='utf-8') as f:
for line in f.readlines():
content.append(Graphs.draw_text(line))
data = pd.read_csv(os.path.join(config.dataset, '指标数据添加时间特征.csv'),
encoding='utf-8') # 计算相关系数用
df_zhibiaofenlei = loadcsv(os.path.join(
config.dataset,'特征处理后的指标名称及分类.csv')) # 气泡图用
df_zhibiaoshuju = data.copy() # 气泡图用
# 绘制特征相关气泡图
grouped = df_zhibiaofenlei.groupby('指标分类')
grouped_corr = pd.DataFrame(columns=['指标分类', '指标数量', '相关性总和'])
content.append(Graphs.draw_little_title('按指标分类分别与预测目标进行皮尔逊相关系数分析：'))
content.append(Graphs.draw_text('''皮尔逊相关系数说明：'''))
content.append(Graphs.draw_text('''衡量两个特征之间的线性相关性。'''))
content.append(Graphs.draw_text('''
相关系数为1：表示两个变量之间存在完全正向的线性关系，即当一个变量增加时，另一个变量也相应增加，且变化是完全一致的。'''))
content.append(Graphs.draw_text(
'''相关系数为-1：表示两个变量之间存在完全负向的线性关系，即当一个变量增加时，另一个变量会相应减少，且变化是完全相反的'''))
content.append(Graphs.draw_text(
'''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()
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']
# 绘制特征相关性直方分布图
plt.figure(figsize=(10, 8))
sns.histplot(correlation_matrix.values.flatten(),
bins=20, kde=True, color='skyblue')
plt.title(f'{name}类指标(共{len(cols_subset)}个)相关性直方分布图')
plt.xlabel('相关系数')
plt.ylabel('频数')
plt.savefig(os.path.join(
config.dataset,f'{name}类指标相关性直方分布图.png'), bbox_inches='tight')
plt.close()
content.append(Graphs.draw_img(
os.path.join(config.dataset, f'{name}类指标相关性直方分布图.png')))
content.append(Graphs.draw_text(
f'{name}类指标(共{len(cols_subset)}个)的相关性直方分布图如上所示。'))
# 相关性大于0的特征
positive_corr_features = correlation_matrix[correlation_matrix > 0].sort_values(
ascending=False).index.tolist()[1:]
print(f'{name}下正相关的特征值有：', positive_corr_features)
if len(positive_corr_features) > 5:
positive_corr_features = positive_corr_features[0:5]
content.append(Graphs.draw_text(
f'{name}类指标中，与预测目标y正相关前五的特征有：{positive_corr_features}'))
draw_feature_trend(feature_data_df, positive_corr_features)
elif len(positive_corr_features) == 0:
pass
else:
positive_corr_features = positive_corr_features
content.append(Graphs.draw_text(
f'其中，与预测目标y正相关的特征有：{positive_corr_features}'))
draw_feature_trend(feature_data_df, positive_corr_features)
# 相关性小于0的特征
negative_corr_features = correlation_matrix[correlation_matrix < 0].sort_values(
ascending=True).index.tolist()
print(f'{name}下负相关的特征值有：', negative_corr_features)
if len(negative_corr_features) > 5:
negative_corr_features = negative_corr_features[:5]
content.append(Graphs.draw_text(
f'与预测目标y负相关前五的特征有：{negative_corr_features}'))
draw_feature_trend(feature_data_df, negative_corr_features)
elif len(negative_corr_features) == 0:
pass
else:
content.append(Graphs.draw_text(
f'{name}类指标中，与预测目标y负相关的特征有：{negative_corr_features}'))
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()
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]})
grouped_corr = pd.concat(
[grouped_corr, goup_corr], axis=0, ignore_index=True)
# 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')
plt.title('指标分类相关性总和的气泡图')
plt.ylabel('数量')
plt.savefig(os.path.join(config.dataset, '指标分类相关性总和的气泡图.png'),
bbox_inches='tight')
plt.close()
content.append(Graphs.draw_img(os.path.join(config.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。'))
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：'))
# 读取模型简介
with open(os.path.join(config.dataset, 'model_introduction.txt'), 'r', encoding='utf-8') as f:
for line in f:
line_split = line.strip().split('--')
if line_split[0] in fivemodels_list:
for introduction in line_split:
content.append(Graphs.draw_text(introduction))
content.append(Graphs.draw_little_title('模型评估：'))
df = pd.read_csv(os.path.join(
config.dataset,'model_evaluation.csv'), encoding='utf-8')
# 判断 df 的数值列转为float
for col in eval_df.columns:
if col not in ['模型(Model)']:
eval_df[col] = eval_df[col].astype(float)
eval_df[col] = eval_df[col].round(3)
# 筛选 fivemodels_list.tolist() 的行
eval_df = eval_df[eval_df['模型(Model)'].isin(fivemodels_list)]
# df转置
eval_df = eval_df.T
# df重置索引
eval_df = eval_df.reset_index()
eval_df = eval_df.T
# # 添加表格
data = eval_df.values.tolist()
col_width = 500/len(eval_df.columns)
content.append(Graphs.draw_table(col_width, *data))
content.append(Graphs.draw_text('评估指标释义：'))
content.append(Graphs.draw_text(
'1. 均方根误差(RMSE)：均方根误差是衡量预测值与实际值之间误差的一种方法，取值越小，误差越小，预测效果越好。'))
content.append(Graphs.draw_text(
'2. 平均绝对误差(MAE)：平均绝对误差是衡量预测值与实际值之间误差的一种方法，取值越小，误差越小，预测效果越好。'))
content.append(Graphs.draw_text(
'3. 平均平方误差(MSE)：平均平方误差是衡量预测值与实际值之间误差的一种方法，取值越小，误差越小，预测效果越好。'))
content.append(Graphs.draw_text('模型拟合：'))
# 添加图片
content.append(Graphs.draw_img(os.path.join(config.dataset, '预测值与真实值对比图.png')))
# 生成pdf文件
doc = SimpleDocTemplate(os.path.join(config.dataset, reportname), pagesize=letter)
doc.build(content)
# pdf 上传到数字化信息平台
try:
if config.is_update_report:
with open(os.path.join(config.dataset, reportname), 'rb') as f:
base64_data = base64.b64encode(f.read()).decode('utf-8')
upload_data["data"]["fileBase64"] = base64_data
upload_data["data"]["fileName"] = reportname
token = get_head_auth_report()
upload_report_data(token, upload_data)
except TimeoutError as e:
print(f"请求超时: {e}")
@exception_logger @exception_logger
def pp_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, inputsize=5, dataset='dataset', time='2024-07-30', reportname='report.pdf', sqlitedb='jbsh_yuanyou.db'): def pp_export_pdf(num_indicators=475, num_models=21, num_dayindicator=202, inputsize=5, dataset='dataset', time='2024-07-30', reportname='report.pdf', sqlitedb='jbsh_yuanyou.db'):
# 创建内容对应的空列表 # 创建内容对应的空列表