From b4b270954bb87b48ec8fbd7549e3dc39ca5d8629 Mon Sep 17 00:00:00 2001
From: Charles <larrylandor@gmail.com>
Date: Tue, 26 Aug 2025 17:02:24 +0800
Subject: [PATCH] =?UTF-8?q?fix:=20=E8=A7=A3=E5=86=B3Kronos=E6=A8=A1?=
 =?UTF-8?q?=E5=9E=8B=E9=A2=84=E6=B5=8B=E8=BF=9E=E7=BB=AD=E6=80=A7=E9=97=AE?=
 =?UTF-8?q?=E9=A2=98=EF=BC=8C=E9=80=9A=E8=BF=87=E4=BB=B7=E6=A0=BC=E8=BD=AC?=
 =?UTF-8?q?=E6=8D=A2=E4=BF=9D=E6=8C=81=E6=95=B0=E6=8D=AE=E8=BF=9E=E7=BB=AD?=
 =?UTF-8?q?=E6=80=A7?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 webui/app.py | 51 +++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 49 insertions(+), 2 deletions(-)

diff --git a/webui/app.py b/webui/app.py
index 38de059..b478cd8 100644
--- a/webui/app.py
+++ b/webui/app.py
@@ -122,6 +122,31 @@ def load_data_file(file_path):
     except Exception as e:
         return None, f"加载文件失败: {str(e)}"
 
+def convert_price_to_returns(df, price_cols=['open', 'high', 'low', 'close']):
+    """将绝对价格转换为相对变化率，解决Kronos模型的连续性问题"""
+    returns_df = df.copy()
+    
+    # 计算相对变化率 (pct_change)
+    for col in price_cols:
+        if col in returns_df.columns:
+            returns_df[col] = returns_df[col].pct_change()
+    
+    # 第一行会是NaN，用0填充
+    returns_df = returns_df.fillna(0)
+    
+    return returns_df
+
+def convert_returns_to_price(returns_df, initial_prices, price_cols=['open', 'high', 'low', 'close']):
+    """将相对变化率转换回绝对价格"""
+    price_df = returns_df.copy()
+    
+    for col in price_cols:
+        if col in price_df.columns:
+            # 从初始价格开始，逐步计算绝对价格
+            price_df[col] = (1 + returns_df[col]) * initial_prices[col]
+    
+    return price_df
+
 def save_prediction_results(file_path, prediction_type, prediction_results, actual_data, input_data, prediction_params):
     """保存预测结果到文件"""
     try:
@@ -476,8 +501,21 @@ def predict():
                 if isinstance(y_timestamp, pd.DatetimeIndex):
                     y_timestamp = pd.Series(y_timestamp, name='timestamps')
                 
-                pred_df = predictor.predict(
-                    df=x_df,
+                # 解决Kronos模型连续性问题：将绝对价格转换为相对变化率
+                original_x_df = x_df.copy()
+                initial_prices = {
+                    'open': x_df['open'].iloc[0],
+                    'high': x_df['high'].iloc[0],
+                    'low': x_df['low'].iloc[0],
+                    'close': x_df['close'].iloc[0]
+                }
+                
+                # 转换为相对变化率
+                x_df_returns = convert_price_to_returns(x_df, ['open', 'high', 'low', 'close'])
+                
+                # 使用转换后的数据进行预测
+                pred_df_returns = predictor.predict(
+                    df=x_df_returns,
                     x_timestamp=x_timestamp,
                     y_timestamp=y_timestamp,
                     pred_len=pred_len,
@@ -486,6 +524,15 @@ def predict():
                     sample_count=sample_count
                 )
                 
+                # 将预测结果转换回绝对价格
+                pred_df = convert_returns_to_price(pred_df_returns, initial_prices, ['open', 'high', 'low', 'close'])
+                
+                # 保持volume列不变（如果存在）
+                if 'volume' in pred_df.columns:
+                    pred_df['volume'] = pred_df_returns['volume']
+                if 'amount' in pred_df.columns:
+                    pred_df['amount'] = pred_df_returns['amount']
+                
             except Exception as e:
                 return jsonify({'error': f'Kronos模型预测失败: {str(e)}'}), 500
         else: