Dinger
2e9c7cd69e
Core changes: - Refactor FastAnalysisService: single LLM multi-factor analysis replaces 7-agent pipeline; add multi-timeframe consensus, threshold calibration, confidence calibration, multi-model ensemble voting - Add RAG memory injection and reflection validation (analysis_memory + reflection worker) - Simplify billing config: remove unused strategy_run/backtest/portfolio_monitor, add ai_code_gen separate billing (different token consumption scale) - Settings hot-reload after save, no backend restart needed Frontend: - Global dark theme overhaul: pure black palette replacing blue-tinted colors across sidebar/header/dashboard/analysis/K-line/user-manage/profile/settings/billing - Fix USDT payment modal dark theme (portal rendering broke CSS selectors) - Refactor position modal: direction + quantity + entry price, remove add/reduce logic, show raw DB values on re-open, save exactly what user inputs - Fix Polymarket prediction market dark text - i18n for position modal title Backend: - Position management: one record per symbol (DELETE+INSERT replacing ON CONFLICT with side), fixes PnL showing 0 when switching long/short - MarketDataCollector data fetching optimization - portfolio_monitor scheduled monitoring improvements - env.example reorganized: common config first, advanced config last Documentation: - README architecture diagram updated to FastAnalysisService flow - Add virtual position, AI tuning config, billing items documentation - Add INDICATOR_DEFINITIONS_CN.md, FRONTEND_FAST_ANALYSIS.md Made-with: Cursor
101 lines
3.7 KiB
Python
101 lines
3.7 KiB
Python
"""
|
|
Reflection Service - Post-trade validation and learning.
|
|
|
|
Validates historical AI decisions against actual price outcomes,
|
|
updates qd_analysis_memory with was_correct/actual_return_pct,
|
|
and optionally triggers AI calibration.
|
|
"""
|
|
import os
|
|
import threading
|
|
import time
|
|
from typing import Dict, Any, Optional
|
|
|
|
from app.utils.logger import get_logger
|
|
from app.services.analysis_memory import get_analysis_memory
|
|
|
|
logger = get_logger(__name__)
|
|
|
|
_reflection_thread: Optional[threading.Thread] = None
|
|
_reflection_stop = threading.Event()
|
|
|
|
|
|
class ReflectionService:
|
|
"""
|
|
Runs verification cycle: validate unvalidated decisions, optionally run calibration.
|
|
"""
|
|
|
|
def run_verification_cycle(self) -> Dict[str, Any]:
|
|
"""
|
|
Run one verification cycle:
|
|
1. Validate unvalidated analysis records (older than min_age_days)
|
|
2. Optionally run AI calibration for configured markets
|
|
"""
|
|
memory = get_analysis_memory()
|
|
min_age_days = int(os.getenv("REFLECTION_MIN_AGE_DAYS", "7"))
|
|
limit = int(os.getenv("REFLECTION_VALIDATE_LIMIT", "200"))
|
|
|
|
stats = memory.validate_unvalidated_older_than(
|
|
min_age_days=min_age_days,
|
|
limit=limit,
|
|
)
|
|
logger.info(f"Reflection validation: {stats}")
|
|
|
|
if stats.get("validated", 0) > 0:
|
|
self._maybe_run_calibration()
|
|
else:
|
|
logger.debug("No new validations, skipping calibration")
|
|
|
|
return stats
|
|
|
|
def _maybe_run_calibration(self) -> None:
|
|
"""Run AI calibration if enabled."""
|
|
if os.getenv("ENABLE_OFFLINE_AI_CALIBRATION", "true").lower() != "true":
|
|
return
|
|
try:
|
|
from app.services.ai_calibration import AICalibrationService
|
|
svc = AICalibrationService()
|
|
markets = (os.getenv("AI_CALIBRATION_MARKETS", "Crypto") or "Crypto").strip().split(",")
|
|
for market in markets:
|
|
market = market.strip()
|
|
if not market:
|
|
continue
|
|
result = svc.calibrate_market(
|
|
market=market,
|
|
lookback_days=int(os.getenv("AI_CALIBRATION_LOOKBACK_DAYS", "30")),
|
|
min_samples=int(os.getenv("AI_CALIBRATION_MIN_SAMPLES", "80")),
|
|
validate_before=False,
|
|
)
|
|
if result:
|
|
logger.info(
|
|
f"[Reflection] Calibration updated for {market}: "
|
|
f"accuracy={result.best_accuracy:.1f}% thr=±{result.buy_threshold:.1f}"
|
|
)
|
|
except Exception as e:
|
|
logger.warning(f"Reflection calibration failed: {e}", exc_info=True)
|
|
|
|
|
|
def start_reflection_worker() -> None:
|
|
"""Start background reflection worker (validates + calibrates periodically)."""
|
|
global _reflection_thread
|
|
# Default to ON to reduce environment-specific configuration needs.
|
|
if os.getenv("ENABLE_REFLECTION_WORKER", "true").lower() != "true":
|
|
logger.info("Reflection worker disabled (ENABLE_REFLECTION_WORKER != true).")
|
|
return
|
|
interval_sec = int(os.getenv("REFLECTION_WORKER_INTERVAL_SEC", "86400"))
|
|
if _reflection_thread and _reflection_thread.is_alive():
|
|
return
|
|
|
|
def _run():
|
|
_reflection_stop.clear()
|
|
logger.info(f"Reflection worker started, interval={interval_sec}s")
|
|
while not _reflection_stop.is_set():
|
|
try:
|
|
ReflectionService().run_verification_cycle()
|
|
except Exception as e:
|
|
logger.error(f"Reflection cycle failed: {e}", exc_info=True)
|
|
_reflection_stop.wait(timeout=interval_sec)
|
|
logger.info("Reflection worker stopped.")
|
|
|
|
_reflection_thread = threading.Thread(target=_run, daemon=True)
|
|
_reflection_thread.start()
|