02071032

evaluation_processing/analysis_cdq.py

@@ -17,42 +17,41 @@ import numbers
 
 app = Flask('analysis_report——service')
 
-def create_fig(df_predict, col_time, label, label_pre):
+def create_fig(df_predict, col_time, label, label_pre, point):
     # 创建一个图表对象
-    fig = make_subplots(rows=1, cols=18, subplot_titles=['超短期-第{}点'.format(p) for p in range(1, 17)]+['超短期-平均值'])
-    for point in range(1, 18):
-        point_data = df_predict[df_predict['howLongAgo']==point]
-        # 获取所有的模型
-        models = df_predict['model'].unique()
-        # 添加实际功率曲线
+    fig = go.Figure()
+    point_data = df_predict[df_predict['howLongAgo']==point]
+    # 获取所有的模型
+    models = df_predict['model'].unique()
+    # 添加实际功率曲线
+    fig.add_trace(go.Scatter(
+        x=df_predict[col_time],
+        y=df_predict[label],
+        mode='lines+markers',
+        name='实际功率',  # 实际功率
+        line=dict(width=1),  # 虚线
+        marker=dict(symbol='circle'),
+    ))
+    # 为每个模型添加预测值和实际功率的曲线
+    for model in models:
+        # 筛选该模型的数据
+        model_data = point_data[point_data['model'] == model]
+
+        # 添加预测值曲线
         fig.add_trace(go.Scatter(
-            x=df_predict[col_time],
-            y=df_predict[label],
+            x=model_data[col_time],
+            y=model_data[label_pre],
             mode='lines+markers',
-            name='实际功率',  # 实际功率
-            line=dict(width=1),  # 虚线
+            name=f'{model} 预测值',  # 预测值
             marker=dict(symbol='circle'),
-        ), row=1, col=point)
-        # 为每个模型添加预测值和实际功率的曲线
-        for model in models:
-            # 筛选该模型的数据
-            model_data = point_data[point_data['model'] == model]
-
-            # 添加预测值曲线
-            fig.add_trace(go.Scatter(
-                x=model_data[col_time],
-                y=model_data[label_pre],
-                mode='lines+markers',
-                name=f'{model} 预测值',  # 预测值
-                marker=dict(symbol='circle'),
-                line=dict(width=2)
-            ), row=1, col=point)
-
+            line=dict(width=2)
+        ))
+    fig_name = '超短期-第{}点'.format(point) if point < 17 else '超短期-平均值'
     # 设置图表的标题和标签
     fig.update_layout(
         template='seaborn',  # 使用 seaborn 模板
         title=dict(
-            # text=f"{label_pre} 与 {label} 对比",  # 标题
+            text=fig_name,  # 标题
             x=0.5, font=dict(size=20, color='darkblue')  # 标题居中并设置字体大小和颜色
         ),
         plot_bgcolor='rgba(255, 255, 255, 0.8)',  # 背景色
@@ -94,9 +93,6 @@ def put_analysis_report_to_html(args, df_predict, df_accuracy):
     label_pre = args['label_pre']
     farmId = args['farmId']
     acc_flag = df_accuracy.shape[0]
-    df_predict = df_predict.applymap(lambda x: float(x.to_decimal()) if isinstance(x, Decimal128) else float(x) if isinstance(x,  numbers.Number) else x).sort_values(by=col_time)
-    if acc_flag > 0:
-        df_accuracy = df_accuracy.applymap(lambda x: float(x.to_decimal()) if isinstance(x, Decimal128) else float(x) if isinstance(x, numbers.Number) else x).sort_values(by=col_time)
     # 获取所有的模型
     models = df_predict['model'].unique()
     aves = []
@@ -114,11 +110,15 @@ def put_analysis_report_to_html(args, df_predict, df_accuracy):
         ave['howLongAgo'] = 17
         ave = ave.reindex(columns=df_predict.columns.tolist())
         aves.append(ave)
-
     df_predict = pd.concat([df_predict]+aves)
-    fig = create_fig(df_predict, col_time, label, label_pre)
+    df_predict = df_predict.applymap(lambda x: float(x.to_decimal()) if isinstance(x, Decimal128) else float(x) if isinstance(x, numbers.Number) else x).sort_values(by=col_time)
+    if acc_flag > 0:
+        df_accuracy = df_accuracy.applymap(lambda x: float(x.to_decimal()) if isinstance(x, Decimal128) else float(x) if isinstance(x, numbers.Number) else x).sort_values(by=col_time)
+    figs = [create_fig(df_predict, col_time, label, label_pre, p) for p in [1, 8, 16, 17]]
     # 将折线图保存为 HTML 片段
-    power_htmls = pio.to_html(fig, full_html=False)
+    power_htmls = [pio.to_html(f, full_html=False) for f in figs]
+    power_htmls = ["<div class='plot-container'>{}</div>".format(html) for html in power_htmls]
+
     # -------------------- 准确率表展示--------------------
     acc_html = ''
     if acc_flag > 0:
@@ -232,10 +232,10 @@ def put_analysis_report_to_html(args, df_predict, df_accuracy):
         <div class="container">
             <h1>分析报告</h1>
             <!-- 曲线对比 -->
-            <div class="plot-container">
-                <h2>1. 预测功率与实际功率曲线对比</h2>
-                {power_htmls}
-            </div>
+            <h2>1. 预测功率与实际功率曲线对比</h2>
+   
+            {''.join(power_htmls)}
+            
             <!-- Pandas DataFrame 表格 -->
             <div style="display:flex; justify-content: space-between;">
                 <h2>2. 准确率对比</h2>
@@ -257,7 +257,7 @@ def put_analysis_report_to_html(args, df_predict, df_accuracy):
     """
     filename = f"{farmId}_{int(time.time() * 1000)}_{random.randint(1000, 9999)}.html"
     # 保存为 HTML
-    directory = '/usr/share/nginx/html'
+    directory = './test'
     if not os.path.exists(directory):
         os.makedirs(directory)
     file_path = os.path.join(directory, filename)
@@ -303,5 +303,9 @@ if __name__ == "__main__":
     logger = logging.getLogger("analysis_report log")
     from waitress import serve
 
-    serve(app, host="0.0.0.0", port=10108)
-    print("server start!")
+    # serve(app, host="0.0.0.0", port=10108)
+    # print("server start!")
+    args_dict = {"mongodb_database": 'db_cdq', 'mongodb_read_table': 'j00234_neu_overwrite,j00234_neu_res', 'col_time': 'dateTime',
+                 'label': 'C_REAL_VALUE', 'label_pre': 'power_forecast', 'farmId': 'j00234'}
+    df_predict, df_accuracy = get_df_list_from_mongo(args_dict)[0], get_df_list_from_mongo(args_dict)[1]
+    path = put_analysis_report_to_html(args_dict, df_predict, df_accuracy)