褚宏光
8dea3fbccb
- Updated all_results.csv with additional stock data and breakout strength metrics. - Revised report.md to improve clarity and detail on stock selection criteria and results. - Expanded strong_breakout_down.csv and strong_breakout_up.csv with new entries reflecting recent analysis. - Introduced new chart images for selected stocks to visualize breakout patterns. - Added plot_converging_triangles.py script for generating visualizations of stocks meeting convergence criteria. - Enhanced report_converging_triangles.py to allow for date-specific reporting and improved output formatting. - Optimized run_converging_triangle.py for performance and added execution time logging.
Technical Patterns Lab
技术形态识别研究与实验仓库，聚焦收敛三角形等形态的批量检测、突破强度计算与可视化验证。
核心功能
- 收敛三角形检测：基于枢轴点 + 边界线拟合 + 几何约束
- 批量滚动计算：支持多股票 × 多交易日的历史回测
- 突破强度评分：0~1 连续分数，综合价格、收敛度、成交量
目录结构
technical-patterns-lab/
├── src/ # 核心算法
│ ├── converging_triangle.py # 收敛三角形检测
│ └── archive/ # 归档代码
├── scripts/ # 运行脚本
│ ├── run_converging_triangle.py # 批量检测（主脚本）
│ └── run_sym_triangle_pkl.py # 对称三角形检测
├── data/ # 数据文件
│ ├── open.pkl, high.pkl, low.pkl, close.pkl, volume.pkl
├── outputs/ # 输出结果
│ ├── converging_triangles/ # 收敛三角形结果
│ └── sym_triangles/ # 对称三角形结果
└── docs/ # 文档
快速开始
# 1. 激活环境
.\.venv\Scripts\Activate.ps1
# 2. 安装依赖（首次）
pip install numpy pandas matplotlib
# 3. 运行批量检测
python scripts/run_converging_triangle.py
输出示例
检测结果: 1837 个有效三角形
- 向上突破: 258 次
- 向下突破: 251 次
- 高强度突破: 375 个 (strength > 0.3)
相关文档
docs/突破强度计算方法.md- 突破强度的计算逻辑docs/对称三角形识别-Python实现.md- 算法原理说明