algorithm_platform/evaluation_processing/analysis.py

# # -*- coding: utf-8 -*-
# import pandas as pd
# import matplotlib.pyplot as plt
# from pymongo import MongoClient
# import pickle
# import numpy as np
# import plotly.express as px
# from plotly.subplots import make_subplots
# import plotly.graph_objects as go
# from flask import Flask,request,jsonify
# from waitress import serve
# import time
# import random
# import argparse
# import logging
# import traceback
# import os
# import lightgbm as lgb
#
# app = Flask('analysis_report——service')
# def get_data_from_mongo(args):
#     # 1.读数据
#     mongodb_connection,mongodb_database,all_table,accuracy_table,model_table,model_name = "mongodb://root:sdhjfREWFWEF23e@192.168.1.43:30000/",args['mongodb_database'],args['train_table'],args['accuracy_table'],args['model_table'],args['model_name']
#     client = MongoClient(mongodb_connection)
#     # 选择数据库（如果数据库不存在，MongoDB 会自动创建）
#     db = client[mongodb_database]
#     # 将游标转换为列表，并创建 pandas DataFrame
#     df_all = pd.DataFrame(db[all_table].find({}, {'_id': 0}))
#
#     df_accuracy = pd.DataFrame(db[accuracy_table].find({}, {'_id': 0}))
#
#     model_data = db[model_table].find_one({"model_name": model_name})
#     if model_data is not None:
#         model_binary = model_data['model']  # 确保这个字段是存储模型的二进制数据
#         # 反序列化模型
#         model = pickle.loads(model_binary)
#     client.close()
#     return df_all,df_accuracy,model
#
#
# def draw_info(df_all,df_accuracy,model,features,args):
#     #1.数据描述 数据描述：
#     col_time = args['col_time']
#     label = args['label']
#     df_accuracy_beginTime = df_accuracy[col_time].min()
#     df_accuracy_endTime = df_accuracy[col_time].max()
#     df_train = df_all[df_all[col_time]<df_accuracy_beginTime][features+[col_time,label]]
#     df_train_beginTime = df_train[col_time].min()
#     df_train_endTime = df_train[col_time].max()
#     text_content = f"训练数据时间范围：{df_train_beginTime} 至 {df_train_endTime}，共{df_train.shape[0]}条记录，测试集数据时间范围：{df_accuracy_beginTime} 至 {df_accuracy_endTime}。<br>lightgbm模型参数：{model.params}"
#     return text_content
#
#
#
# def draw_global_scatter(df,args):
#     # --- 1. 实际功率和辐照度的散点图 ---
#     col_x = args['scatter_col_x']
#     col_y = args['label']
#     scatter_fig = px.scatter(
#         df,
#         x=col_x,
#         y=col_y,
#         title=f"{col_x}和{col_y}的散点图",
#         labels={"辐照度": "辐照度 (W/m²)", "实际功率": "实际功率 (kW)"}
#     )
#     return scatter_fig
#
#
#
# def draw_corr(df,features,args):
#
#     # --- 2. 相关性热力图 ---
#     # 计算相关性矩阵
#     label = args['label']
#     features_coor = features+[label]
#     corr_matrix = df[features_coor].corr()
#     # 使用 Plotly Express 绘制热力图
#     heatmap_fig = px.imshow(corr_matrix,
#                     text_auto=True,  # 显示数值
#                     color_continuous_scale='RdBu',  # 配色方案
#                     title="Correlation Heatmap")
#     heatmap_fig.update_coloraxes(showscale=False)
#
#     return heatmap_fig
#
# def draw_feature_importance(model,features):
#     # --- 3. 特征重要性排名 ---
#     # 获取特征重要性
#     importance = model.feature_importance()  # 'split' 或 'gain'，根据需求选择
#
#     # 转换为 DataFrame 方便绘图
#     feature_importance_df = pd.DataFrame({
#         'Feature': features,
#         'Importance': importance
#     })
#     feature_importance_df = feature_importance_df.sort_values(by='Importance', ascending=False)
#
#     # 使用 Plotly Express 绘制条形图
#     importance_fig = px.bar(feature_importance_df, x='Feature', y='Importance',
#                  title="特征重要性排名",
#                  labels={'Feature': '特征', 'Importance': '重要性'},
#                  color='Importance',
#                  color_continuous_scale='Viridis')
#     # 更新每个 trace，确保没有图例
#
#     importance_fig.update_layout(title="模型特征重要性排名",
#                                  showlegend=False  # 移除图例
#                                 )
#     importance_fig.update_coloraxes(showscale=False)
#     return importance_fig
#
#
# def draw_data_info_table(content):
#     # --- 4. 创建数据说明的表格 ---
#     # 转换为表格格式：1行1列，且填充文字说明
#     # 转换为表格格式
#     # 创建一个空的图
#     table_fig = go.Figure()
#
#     # 第一部分: 显示文字说明
#     table_fig.add_trace(go.Table(
#         header=dict(
#             values=["说明"],  # 表格只有一列：说明
#             fill_color="paleturquoise",
#             align="center"
#         ),
#         cells=dict(
#             values=[[content]] ,  # 第一行填入文本说明
#             fill_color="lavender",
#             align="center"
#         )
#     ))
#
#
#     return table_fig
#
#
#
# def draw_accuracy_table(df,content):
#
#     # --- 4. 每日的准确率表格 ---
#     # 转换为表格格式
#     table_fig = go.Figure(
#         data=[
#             go.Table(
#                 header=dict(
#                     values=list(df.columns),
#                     fill_color="paleturquoise",
#                     align="center"
#                 ),
#                 cells=dict(
#                     values=[df[col] for col in df.columns],
#                     fill_color="lavender",
#                     align="center"
#                 )
#             )
#         ]
#     )
#     table_fig.update_layout(title="准确率表", showlegend=False)
#     return table_fig
#
#
# @app.route('/analysis_report', methods=['POST'])
# def analysis_report():
#     start_time = time.time()
#     result = {}
#     success = 0
#     path = ""
#     print("Program starts execution!")
#     try:
#         args = request.values.to_dict()
#         print('args',args)
#         logger.info(args)
#         #获取数据
#         df_all, df_accuracy, model = get_data_from_mongo(args)
#         features = model.feature_name()
#         text_content = draw_info(df_all,df_accuracy,model,features,args)
#         text_fig,scatter_fig,heatmap_fig,importance_fig,table_fig=draw_data_info_table(text_content),draw_global_scatter(df_all,args),draw_corr(df_all,features,args),draw_feature_importance(model,features),\
#         draw_accuracy_table(df_accuracy,text_content)
#         # --- 合并图表并保存到一个 HTML 文件 ---
#         # 创建子图布局
#         combined_fig = make_subplots(
#             rows=5, cols=1,
#             subplot_titles=["数据-模型概览","辐照度和实际功率的散点图", "相关性","特征重要性排名", "准确率表"],
#             row_heights=[0.3, 0.6, 0.6, 0.6, 0.4],
#             specs=[[{"type": "table"}], [{"type": "xy"}], [{"type": "heatmap"}], [{"type": "xy"}],[{"type": "table"}]]  # 指定每个子图类型
#         )
#         # 添加文本信息到子图（第一行）
#         # 添加文字说明
#         for trace in text_fig.data:
#             combined_fig.add_trace(trace, row=1, col=1)
#
#         # 添加散点图
#         for trace in scatter_fig.data:
#             combined_fig.add_trace(trace, row=2, col=1)
#
#         # 添加相关性热力图
#         for trace in heatmap_fig.data:
#             combined_fig.add_trace(trace, row=3, col=1)
#
#         # 添加特征重要性排名图
#         for trace in importance_fig.data:
#             combined_fig.add_trace(trace, row=4, col=1)
#
#         # 添加表格
#         for trace in table_fig.data:
#             combined_fig.add_trace(trace, row=5, col=1)
#
#         # 更新布局
#         combined_fig.update_layout(
#         height=1500,
#         title_text="分析结果汇总",  # 添加换行符以适应文本内容
#         title_x=0.5,  # 中心对齐标题
#         showlegend=False,
#         )
#         combined_fig.update_coloraxes(showscale=False)
#         filename = f"{int(time.time() * 1000)}_{random.randint(1000, 9999)}.html"
#         # 保存为 HTML
#         directory = '/usr/share/nginx/html'
#         if not os.path.exists(directory):
#             os.makedirs(directory)
#         file_path = os.path.join(directory, filename)
#         # combined_fig.write_html(f"D://usr//{filename}")
#         combined_fig.write_html(file_path)
#         path = f"http://ds2:10093/{filename}"
#         success = 1
#     except Exception as e:
#         my_exception = traceback.format_exc()
#         my_exception.replace("\n","\t")
#         result['msg'] = my_exception
#     end_time = time.time()
#     result['success'] = success
#     result['args'] = args
#     result['start_time'] = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(start_time))
#     result['end_time'] = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime(end_time))
#     result['file_path'] = path
#     print("Program execution ends!")
#     return result
#
#
# if __name__=="__main__":
#     # print("Program starts execution!")
#     # logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
#     # logger = logging.getLogger("analysis_report log")
#     # from waitress import serve
#     # serve(app, host="0.0.0.0", port=10092)
#     # print("server start!")