crypto_predictor
1bd9d25f60
- 删除测试脚本: test_2026_classification.py, test_lagllama.py, test_timesfm.py, test_ttm.py - 删除2019相关文件: analyze_2019_*, apply_2019_*, dlinear_2019_*, three_models_2019_* - 添加集成模型相关代码和分析脚本 - 添加N-BEATS模型和CatBoost分类模型
303 lines
11 KiB
Python
303 lines
11 KiB
Python
"""N-BEATS + DLinear 模型集成预测 - 交集策略"""
|
|
|
|
import sys
|
|
import os
|
|
sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
|
|
|
|
import torch
|
|
import pickle
|
|
import numpy as np
|
|
import pandas as pd
|
|
import pandas_ta as ta
|
|
from sklearn.preprocessing import StandardScaler
|
|
from sklearn.metrics import confusion_matrix
|
|
from data.fetcher import BinanceDataFetcher
|
|
|
|
|
|
# ==================== 预处理器定义 ====================
|
|
|
|
class ClassificationPreprocessor:
|
|
"""DLinear分类预处理器"""
|
|
def __init__(self, lookback=168, forecast_horizon=1):
|
|
self.lookback = lookback
|
|
self.forecast_horizon = forecast_horizon
|
|
self.scaler = StandardScaler()
|
|
self.is_fitted = False
|
|
|
|
def add_features(self, df):
|
|
df = df.copy()
|
|
df['rsi'] = ta.rsi(df['close'], length=14)
|
|
macd = ta.macd(df['close'])
|
|
if macd is not None and not macd.empty:
|
|
df['macd'] = macd.iloc[:, 0]
|
|
df['macd_hist'] = macd.iloc[:, 1]
|
|
df['macd_signal'] = macd.iloc[:, 2]
|
|
return df
|
|
|
|
def create_labels(self, df):
|
|
df = df.copy()
|
|
df['future_price'] = df['close'].shift(-self.forecast_horizon)
|
|
df['future_return'] = (df['future_price'] - df['close']) / df['close']
|
|
df['label'] = (df['future_return'] > 0).astype(int)
|
|
return df[:-self.forecast_horizon]
|
|
|
|
def preprocess(self, df, fit=True):
|
|
df = self.add_features(df)
|
|
df = self.create_labels(df)
|
|
df = df.dropna().reset_index(drop=True)
|
|
|
|
feature_cols = ['open', 'high', 'low', 'close', 'volume',
|
|
'rsi', 'macd', 'macd_hist', 'macd_signal']
|
|
|
|
if fit:
|
|
scaled_features = self.scaler.fit_transform(df[feature_cols])
|
|
self.is_fitted = True
|
|
self.feature_columns = feature_cols
|
|
else:
|
|
scaled_features = self.scaler.transform(df[feature_cols])
|
|
|
|
X, y = [], []
|
|
for i in range(len(scaled_features) - self.lookback):
|
|
X.append(scaled_features[i:i + self.lookback])
|
|
y.append(df['label'].iloc[i + self.lookback])
|
|
|
|
return np.array(X), np.array(y), df
|
|
|
|
|
|
class NBeatsPreprocessor:
|
|
"""N-BEATS预处理器"""
|
|
def __init__(self, lookback=168, forecast_horizon=1):
|
|
self.lookback = lookback
|
|
self.forecast_horizon = forecast_horizon
|
|
self.price_scaler = StandardScaler()
|
|
self.is_fitted = False
|
|
|
|
def preprocess(self, df, fit=True):
|
|
df = df.copy()
|
|
df['future_price'] = df['close'].shift(-self.forecast_horizon)
|
|
df = df[:-self.forecast_horizon].reset_index(drop=True)
|
|
|
|
prices = df['close'].values.reshape(-1, 1)
|
|
future_prices = df['future_price'].values.reshape(-1, 1)
|
|
|
|
if fit:
|
|
scaled_prices = self.price_scaler.fit_transform(prices)
|
|
scaled_future = self.price_scaler.transform(future_prices)
|
|
self.is_fitted = True
|
|
else:
|
|
scaled_prices = self.price_scaler.transform(prices)
|
|
scaled_future = self.price_scaler.transform(future_prices)
|
|
|
|
X, y = [], []
|
|
for i in range(len(scaled_prices) - self.lookback):
|
|
X.append(scaled_prices[i:i + self.lookback, 0])
|
|
y.append(scaled_future[i + self.lookback, 0])
|
|
|
|
return np.array(X), np.array(y), df
|
|
|
|
def inverse_transform_price(self, scaled_price):
|
|
return self.price_scaler.inverse_transform(scaled_price.reshape(-1, 1)).flatten()
|
|
|
|
|
|
def main():
|
|
print("="*80)
|
|
print("模型集成预测：N-BEATS + DLinear 交集策略")
|
|
print("="*80)
|
|
|
|
# ========== 步骤1: 加载两个模型 ==========
|
|
print("\n[1/5] 加载两个模型...")
|
|
|
|
device = torch.device('mps' if torch.backends.mps.is_available() else 'cpu')
|
|
|
|
# 加载DLinear分类模型
|
|
from models.dlinear.model import DLinear
|
|
dlinear_checkpoint = torch.load(
|
|
'./models/checkpoints/dlinear_classification_2026/best_model.pt',
|
|
map_location=device
|
|
)
|
|
dlinear_model = DLinear(dlinear_checkpoint['config'])
|
|
dlinear_model.load_state_dict(dlinear_checkpoint['model_state_dict'])
|
|
dlinear_model = dlinear_model.to(device)
|
|
dlinear_model.eval()
|
|
|
|
with open('./models/checkpoints/dlinear_classification_2026/preprocessor.pkl', 'rb') as f:
|
|
dlinear_preprocessor = pickle.load(f)
|
|
|
|
print(f"✓ DLinear分类模型已加载 (准确率: 53.10%)")
|
|
|
|
# 加载N-BEATS回归模型
|
|
from models.nbeats.model import NBeats
|
|
nbeats_checkpoint = torch.load(
|
|
'./models/checkpoints/nbeats_btc_1h/best_model.pt',
|
|
map_location=device
|
|
)
|
|
nbeats_model = NBeats(nbeats_checkpoint['config'])
|
|
nbeats_model.load_state_dict(nbeats_checkpoint['model_state_dict'])
|
|
nbeats_model = nbeats_model.to(device)
|
|
nbeats_model.eval()
|
|
|
|
with open('./models/checkpoints/nbeats_btc_1h/preprocessor.pkl', 'rb') as f:
|
|
nbeats_preprocessor = pickle.load(f)
|
|
|
|
print(f"✓ N-BEATS回归模型已加载 (准确率: 50.83%)")
|
|
|
|
# ========== 步骤2: 获取2026年测试数据 ==========
|
|
print("\n[2/5] 获取2026年测试数据...")
|
|
fetcher = BinanceDataFetcher()
|
|
|
|
try:
|
|
df_2026 = fetcher.fetch_ohlcv('BTC', '1h', start_date='2026-01-01',
|
|
end_date='2026-12-31', use_cache=False)
|
|
except:
|
|
df_2026 = fetcher.fetch_ohlcv('BTC', '1h', start_date='2025-12-01', use_cache=False)
|
|
|
|
print(f"✓ 测试数据: {len(df_2026)} 行")
|
|
print(f"✓ 时间: {df_2026['timestamp'].min()} 至 {df_2026['timestamp'].max()}")
|
|
|
|
# ========== 步骤3: 预处理数据 ==========
|
|
print("\n[3/5] 预处理数据...")
|
|
|
|
# DLinear预处理
|
|
X_dlinear, y_dlinear, df_dlinear = dlinear_preprocessor.preprocess(df_2026, fit=False)
|
|
|
|
# N-BEATS预处理
|
|
X_nbeats, y_nbeats, df_nbeats = nbeats_preprocessor.preprocess(df_2026, fit=False)
|
|
|
|
# 确保两个模型的样本数量一致
|
|
min_samples = min(len(X_dlinear), len(X_nbeats))
|
|
X_dlinear = X_dlinear[:min_samples]
|
|
y_dlinear = y_dlinear[:min_samples]
|
|
X_nbeats = X_nbeats[:min_samples]
|
|
y_nbeats = y_nbeats[:min_samples]
|
|
|
|
print(f"✓ 对齐样本数: {min_samples}")
|
|
|
|
# ========== 步骤4: 两个模型分别预测 ==========
|
|
print("\n[4/5] 两个模型分别预测...")
|
|
|
|
# DLinear预测
|
|
X_dlinear_tensor = torch.FloatTensor(X_dlinear).to(device)
|
|
with torch.no_grad():
|
|
dlinear_logits = dlinear_model(X_dlinear_tensor).cpu().numpy()
|
|
dlinear_probs = 1 / (1 + np.exp(-dlinear_logits)) # sigmoid
|
|
dlinear_preds = (dlinear_probs > 0.5).astype(int).flatten() # 1=涨, 0=跌
|
|
|
|
print(f"✓ DLinear预测完成")
|
|
print(f" 预测上涨: {(dlinear_preds==1).sum()} ({(dlinear_preds==1).sum()/len(dlinear_preds)*100:.1f}%)")
|
|
print(f" 预测下跌: {(dlinear_preds==0).sum()} ({(dlinear_preds==0).sum()/len(dlinear_preds)*100:.1f}%)")
|
|
|
|
# N-BEATS预测
|
|
X_nbeats_tensor = torch.FloatTensor(X_nbeats).to(device)
|
|
with torch.no_grad():
|
|
nbeats_pred_scaled = nbeats_model(X_nbeats_tensor).cpu().numpy()
|
|
|
|
# 还原真实价格
|
|
nbeats_pred_prices = nbeats_preprocessor.inverse_transform_price(nbeats_pred_scaled)
|
|
|
|
# 获取当前价格（用于判断方向）
|
|
current_prices = df_nbeats['close'].iloc[168:168+len(nbeats_pred_prices)].values
|
|
nbeats_preds = (nbeats_pred_prices > current_prices).astype(int) # 1=涨, 0=跌
|
|
|
|
print(f"✓ N-BEATS预测完成")
|
|
print(f" 预测上涨: {(nbeats_preds==1).sum()} ({(nbeats_preds==1).sum()/len(nbeats_preds)*100:.1f}%)")
|
|
print(f" 预测下跌: {(nbeats_preds==0).sum()} ({(nbeats_preds==0).sum()/len(nbeats_preds)*100:.1f}%)")
|
|
|
|
# ========== 步骤5: 交集策略 ==========
|
|
print("\n[5/5] 应用交集策略...")
|
|
|
|
# 真实标签
|
|
true_labels = y_dlinear
|
|
|
|
# 策略1: 两个模型都预测上涨 → 上涨
|
|
# 策略2: 两个模型都预测下跌 → 下跌
|
|
# 策略3: 两个模型预测不一致 → 不做决策（弃权）
|
|
|
|
both_agree_up = (dlinear_preds == 1) & (nbeats_preds == 1)
|
|
both_agree_down = (dlinear_preds == 0) & (nbeats_preds == 0)
|
|
disagree = (dlinear_preds != nbeats_preds)
|
|
|
|
n_agree_up = both_agree_up.sum()
|
|
n_agree_down = both_agree_down.sum()
|
|
n_disagree = disagree.sum()
|
|
|
|
print(f"\n📊 预测一致性分析:")
|
|
print(f" 双方都预测上涨: {n_agree_up} ({n_agree_up/len(dlinear_preds)*100:.1f}%)")
|
|
print(f" 双方都预测下跌: {n_agree_down} ({n_agree_down/len(dlinear_preds)*100:.1f}%)")
|
|
print(f" 双方预测不一致: {n_disagree} ({n_disagree/len(dlinear_preds)*100:.1f}%)")
|
|
|
|
# 只在两个模型一致时做决策
|
|
agree_mask = both_agree_up | both_agree_down
|
|
ensemble_preds = np.full(len(dlinear_preds), -1) # -1表示弃权
|
|
ensemble_preds[both_agree_up] = 1
|
|
ensemble_preds[both_agree_down] = 0
|
|
|
|
# 有效预测（非弃权）
|
|
valid_mask = ensemble_preds != -1
|
|
valid_preds = ensemble_preds[valid_mask]
|
|
valid_labels = true_labels[valid_mask]
|
|
|
|
print(f"\n📊 集成策略结果:")
|
|
print(f" 有效预测数: {valid_mask.sum()} ({valid_mask.sum()/len(dlinear_preds)*100:.1f}%)")
|
|
print(f" 弃权数: {n_disagree} ({n_disagree/len(dlinear_preds)*100:.1f}%)")
|
|
|
|
if len(valid_preds) > 0:
|
|
ensemble_accuracy = (valid_preds == valid_labels).mean()
|
|
print(f"\n✨ 交集策略准确率: {ensemble_accuracy*100:.2f}%")
|
|
|
|
# 混淆矩阵
|
|
cm = confusion_matrix(valid_labels, valid_preds)
|
|
tn, fp, fn, tp = cm.ravel()
|
|
|
|
print(f"\n详细分析（仅有效预测）:")
|
|
print(f" ✓ 预测涨且真涨: {tp} ({tp/len(valid_preds)*100:.1f}%)")
|
|
print(f" ✓ 预测跌且真跌: {tn} ({tn/len(valid_preds)*100:.1f}%)")
|
|
print(f" ✗ 预测涨但真跌: {fp} ({fp/len(valid_preds)*100:.1f}%)")
|
|
print(f" ✗ 预测跌但真涨: {fn} ({fn/len(valid_preds)*100:.1f}%)")
|
|
|
|
# ========== 对比分析 ==========
|
|
print("\n" + "="*80)
|
|
print("📊 性能对比")
|
|
print("="*80)
|
|
|
|
dlinear_acc = (dlinear_preds == true_labels).mean()
|
|
nbeats_acc = (nbeats_preds == true_labels).mean()
|
|
|
|
print(f"\n单模型性能:")
|
|
print(f" DLinear分类: {dlinear_acc*100:.2f}%")
|
|
print(f" N-BEATS回归: {nbeats_acc*100:.2f}%")
|
|
|
|
if len(valid_preds) > 0:
|
|
print(f"\n集成策略性能:")
|
|
print(f" 交集准确率: {ensemble_accuracy*100:.2f}%")
|
|
print(f" 覆盖率: {valid_mask.sum()/len(dlinear_preds)*100:.1f}%")
|
|
print(f" 弃权率: {n_disagree/len(dlinear_preds)*100:.1f}%")
|
|
|
|
# 性能评估
|
|
print("\n" + "="*80)
|
|
print("🎯 策略评估")
|
|
print("="*80)
|
|
|
|
if ensemble_accuracy > max(dlinear_acc, nbeats_acc):
|
|
improvement = ensemble_accuracy - max(dlinear_acc, nbeats_acc)
|
|
print(f"\n ✅ 交集策略优于单模型!")
|
|
print(f" 准确率提升: {improvement*100:.2f}个百分点")
|
|
print(f" 代价: 弃权{n_disagree}次预测 ({n_disagree/len(dlinear_preds)*100:.1f}%)")
|
|
else:
|
|
print(f"\n ⚠️ 交集策略未能提升准确率")
|
|
print(f" 可能原因: 两个模型一致时也可能都错")
|
|
|
|
# 计算如果弃权时不交易的盈利
|
|
print(f"\n💰 实际收益分析:")
|
|
print(f" 如果只在两个模型都同意时交易:")
|
|
print(f" - 交易次数: {valid_mask.sum()} ({valid_mask.sum()/len(dlinear_preds)*100:.1f}%)")
|
|
print(f" - 准确率: {ensemble_accuracy*100:.2f}%")
|
|
print(f" - 避免交易: {n_disagree}次 (规避不确定性)")
|
|
|
|
print("\n" + "="*80)
|
|
print("✓ 集成预测分析完成!")
|
|
print("="*80)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|