6维评分通用模块 + 港股通T+2延迟标注

1. stock_scorer.py — 共享的6维评分模块
   - score_future_outlook(code, data) → (score, reasons)
   - rank_by_outlook(holdings, data) → 排序列表
   - settlement_delay_note(sell_code, buy_code) → 结算延迟说明
   - is_hk_stock(code) → 判断港股通标的

2. stale_push_wlin.py 改用共享模块(去掉本地函数定义)

3. 换仓评估增加港股通结算延迟检测:
   - 卖港股→买A股时标注⚠️T+2到账限制
   - 本次推荐(招商银行+A股→海博思创)无需标注,全是A股
This commit is contained in:
知微
2026-06-24 11:59:55 +08:00
parent c7f15ebe0b
commit 9a984dd4dc
4 changed files with 193 additions and 99 deletions
+156
View File
@@ -0,0 +1,156 @@
#!/usr/bin/env python3
"""
stock_scorer.py — 6维股票前景评分系统
用于全面评估持仓或自选股的前景。
评分越低(负数越大)= 前景越差,越值得考虑卖出。
评分越高(正数越大)= 前景越好,越值得持有或买入。
使用场景:
- 换仓评估(决定卖什么)
- 持仓审查(定期排名)
- 组合优化(识别需清理的票)
调用方式:
from stock_scorer import score_future_outlook
score, reasons = score_future_outlook(code, decisions_dict)
# 批量评估
from stock_scorer import rank_by_outlook
rankings = rank_by_outlook(portfolio_holdings, decisions_dict)
"""
def score_future_outlook(code, decisions_data):
"""6维评分:基于决策系统分析数据评估股票前景。
评分维度(按重要度排序):
1. timing_signal — 决策系统主信号(买入/持有/深套持有)
2. 技术形态 — bearish/bullish/neutral
3. 量价关系 — 买卖盘主导
4. 行业背景 — 板块强弱
5. 盈亏比RR — 预期收益/风险
6. 股票类别 — 蓝筹/深套/题材
Args:
code: 股票代码
decisions_data: decisions.json 的 "decisions" 数组或 dict(code→数据)
Returns:
(score, reasons) — score浮点数,reasons字符串列表
"""
# 支持两种输入格式
if isinstance(decisions_data, dict):
d = decisions_data.get(code, {})
elif isinstance(decisions_data, list):
d = {}
for e in decisions_data:
if e.get("code") == code:
d = e
break
else:
return -999, ["无数据"]
if not d:
return -999, ["无数据"]
score = 0.0
reasons = []
# 1. timing_signal — 最直接的信号
signal = (d.get('timing_signal') or '').strip()
if '买入' in signal or '加仓' in signal:
score += 3
reasons.append('有买入信号')
elif '深套持有' in signal or '弱势持有' in signal:
score -= 2
reasons.append('深套/弱势持有')
elif signal in ('持有', '') or not signal:
score -= 0.5 # 中性偏弱(没有积极信号就是消极信号)
reasons.append('无积极信号')
# 2. 技术形态
tech = (d.get('tech_snapshot') or '') or ''
if '/bearish' in tech:
score -= 1.5
reasons.append('技术偏空')
elif '/bullish' in tech:
score += 1.5
reasons.append('技术偏多')
# 3. 量价关系
if '主动卖盘占优' in tech:
score -= 1
reasons.append('卖盘主导')
elif '主动买盘占优' in tech:
score += 1
reasons.append('买盘主导')
# 4. 行业背景
sector = (d.get('sector_context') or '') or ''
if '大跌' in sector or '偏弱' in sector:
score -= 0.5
if '大涨' in sector or '偏强' in sector:
score += 0.5
# 5. 盈亏比RR
rr = d.get('rr_ratio', 0) or 0
if rr >= 2:
score += 1
reasons.append(f'RR{rr:.1f}')
elif rr < 1:
score -= 0.5
reasons.append(f'RR{rr:.1f}<1')
else:
reasons.append(f'RR{rr:.1f}')
# 6. 股票类别
cat = (d.get('stock_category') or '') or ''
if '蓝筹' in cat or '白马' in cat:
score += 0.5
elif '深套' in cat or '弱势' in cat:
score -= 0.5
return round(score, 1), reasons
def rank_by_outlook(holdings_list, decisions_data):
"""批量评估持仓的前景,返回排序后的列表(最差排前)。
Args:
holdings_list: 持仓列表,每项有 code, name, shares, cost, price 等
decisions_data: decisions.json 数据
Returns:
排序后的列表,每项增加了 score, reasons 字段
"""
results = []
for h in holdings_list:
code = h.get("code", "")
if not code:
continue
score, reasons = score_future_outlook(code, decisions_data)
results.append({**h, "score": score, "reasons": reasons})
results.sort(key=lambda x: x["score"])
return results
def is_hk_stock(code):
"""判断是否为港股(港股通标的代码通常5位)"""
return len(str(code)) <= 5
def is_a_stock(code):
"""判断是否为A股(6位代码)"""
return len(str(code)) == 6
def settlement_delay_note(sell_code, buy_code):
"""返回资金结算延迟说明(如有)。"""
sell_is_hk = is_hk_stock(sell_code)
buy_is_hk = is_hk_stock(buy_code)
if sell_is_hk and not buy_is_hk:
return "(港股通卖出需T+2到账后才能买A股,注意时间差)"
return ""