feie9454/kronos-trader
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: 添加详细的调试信息，显示价格转换的每一步操作

Browse Source
This commit is contained in:
Charles 2025-08-26 17:13:18 +08:00
parent b4b270954b
commit 35d35f740f
1 changed files with 38 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
webui/app.py
View File

@ -124,27 +124,44 @@ def load_data_file(file_path):
def convert_price_to_returns(df, price_cols=['open', 'high', 'low', 'close']): 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""" """将绝对价格转换为相对变化率，解决Kronos模型的连续性问题"""
print(f"[DEBUG] convert_price_to_returns: 开始转换，输入数据形状: {df.shape}")
print(f"[DEBUG] convert_price_to_returns: 输入数据前3行:")
print(df[price_cols].head(3))
returns_df = df.copy() returns_df = df.copy()
# 计算相对变化率 (pct_change) # 计算相对变化率 (pct_change)
for col in price_cols: for col in price_cols:
if col in returns_df.columns: if col in returns_df.columns:
print(f"[DEBUG] convert_price_to_returns: 转换列 {col}")
returns_df[col] = returns_df[col].pct_change() returns_df[col] = returns_df[col].pct_change()
# 第一行会是NaN，用0填充 # 第一行会是NaN，用0填充
returns_df = returns_df.fillna(0) returns_df = returns_df.fillna(0)
print(f"[DEBUG] convert_price_to_returns: 转换完成，输出数据前3行:")
print(returns_df[price_cols].head(3))
return returns_df return returns_df
def convert_returns_to_price(returns_df, initial_prices, price_cols=['open', 'high', 'low', 'close']): 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""" """将相对变化率转换回绝对价格"""
print(f"[DEBUG] convert_returns_to_price: 开始转换，输入数据形状: {returns_df.shape}")
print(f"[DEBUG] convert_returns_to_price: 初始价格: {initial_prices}")
print(f"[DEBUG] convert_returns_to_price: 输入相对变化率前3行:")
print(returns_df[price_cols].head(3))
price_df = returns_df.copy() price_df = returns_df.copy()
for col in price_cols: for col in price_cols:
if col in price_df.columns: if col in price_df.columns:
print(f"[DEBUG] convert_returns_to_price: 转换列 {col}")
# 从初始价格开始，逐步计算绝对价格 # 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] price_df[col] = (1 + returns_df[col]) * initial_prices[col]
print(f"[DEBUG] convert_returns_to_price: 转换完成，输出绝对价格前3行:")
print(price_df[price_cols].head(3))
return price_df return price_df
def save_prediction_results(file_path, prediction_type, prediction_results, actual_data, input_data, prediction_params): def save_prediction_results(file_path, prediction_type, prediction_results, actual_data, input_data, prediction_params):
@ -495,12 +512,16 @@ def predict():
y_timestamp = df.iloc[lookback:lookback+pred_len]['timestamps'] y_timestamp = df.iloc[lookback:lookback+pred_len]['timestamps']
prediction_type = "Kronos模型预测 (最新数据)" prediction_type = "Kronos模型预测 (最新数据)"
print(f"[DEBUG] 步骤1: 检查并转换时间戳格式...")
# 确保时间戳是Series格式，不是DatetimeIndex，避免Kronos模型的.dt属性错误 # 确保时间戳是Series格式，不是DatetimeIndex，避免Kronos模型的.dt属性错误
if isinstance(x_timestamp, pd.DatetimeIndex): if isinstance(x_timestamp, pd.DatetimeIndex):
print(f"[DEBUG] 将x_timestamp从DatetimeIndex转换为Series")
x_timestamp = pd.Series(x_timestamp, name='timestamps') x_timestamp = pd.Series(x_timestamp, name='timestamps')
if isinstance(y_timestamp, pd.DatetimeIndex): if isinstance(y_timestamp, pd.DatetimeIndex):
print(f"[DEBUG] 将y_timestamp从DatetimeIndex转换为Series")
y_timestamp = pd.Series(y_timestamp, name='timestamps') y_timestamp = pd.Series(y_timestamp, name='timestamps')
print(f"[DEBUG] 步骤2: 准备价格转换，保存初始价格...")
# 解决Kronos模型连续性问题：将绝对价格转换为相对变化率 # 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() original_x_df = x_df.copy()
initial_prices = { initial_prices = {
@ -509,10 +530,15 @@ def predict():
'low': x_df['low'].iloc[0], 'low': x_df['low'].iloc[0],
'close': x_df['close'].iloc[0] 'close': x_df['close'].iloc[0]
} }
print(f"[DEBUG] 初始价格: {initial_prices}")
print(f"[DEBUG] 步骤3: 将绝对价格转换为相对变化率...")
# 转换为相对变化率 # 转换为相对变化率
x_df_returns = convert_price_to_returns(x_df, ['open', 'high', 'low', 'close']) x_df_returns = convert_price_to_returns(x_df, ['open', 'high', 'low', 'close'])
print(f"[DEBUG] 转换完成，输入数据前5行相对变化率:")
print(x_df_returns[['open', 'high', 'low', 'close']].head())
print(f"[DEBUG] 步骤4: 使用转换后的数据进行Kronos模型预测...")
# 使用转换后的数据进行预测 # 使用转换后的数据进行预测
pred_df_returns = predictor.predict( pred_df_returns = predictor.predict(
df=x_df_returns, df=x_df_returns,
@ -523,15 +549,27 @@ def predict():
top_p=top_p, top_p=top_p,
sample_count=sample_count sample_count=sample_count
) )
print(f"[DEBUG] 预测完成，预测结果前5行相对变化率:")
print(pred_df_returns[['open', 'high', 'low', 'close']].head())
print(f"[DEBUG] 步骤5: 将预测结果转换回绝对价格...")
# 将预测结果转换回绝对价格 # 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']) pred_df = convert_returns_to_price(pred_df_returns, initial_prices, ['open', 'high', 'low', 'close'])
print(f"[DEBUG] 转换完成，预测结果前5行绝对价格:")
print(pred_df[['open', 'high', 'low', 'close']].head())
print(f"[DEBUG] 步骤6: 处理volume和amount列...")
# 保持volume列不变（如果存在） # 保持volume列不变（如果存在）
if 'volume' in pred_df.columns: if 'volume' in pred_df.columns:
pred_df['volume'] = pred_df_returns['volume'] pred_df['volume'] = pred_df_returns['volume']
print(f"[DEBUG] 已复制volume列")
if 'amount' in pred_df.columns: if 'amount' in pred_df.columns:
pred_df['amount'] = pred_df_returns['amount'] pred_df['amount'] = pred_df_returns['amount']
print(f"[DEBUG] 已复制amount列")
print(f"[DEBUG] 步骤7: 检查最终预测结果...")
print(f"[DEBUG] 最终预测结果形状: {pred_df.shape}")
print(f"[DEBUG] 最终预测结果列: {list(pred_df.columns)}")
except Exception as e: except Exception as e:
return jsonify({'error': f'Kronos模型预测失败: {str(e)}'}), 500 return jsonify({'error': f'Kronos模型预测失败: {str(e)}'}), 500