v2.0.0 GPU 加速版本 (GPU Acceleration Version)

主要改进 / Major Improvements

• GPU 加速支持 / GPU acceleration support

  • 使用 PyTorch 实现 GPU 并行计算 / Use PyTorch for GPU parallel computing
  • 性能提升数十倍（相比 v1.0）/ Dozens of times faster (vs v1.0)

• 两步运行 / Two-step workflow

  • python cs2_gen_preresult.py - 生成模拟数据 / Generate simulation data
  • python cs2_gen_final.py - GPU 优化 Pick'Em / GPU-accelerated Pick'Em

• YAML 配置文件 / YAML configuration

  • batchsize.yaml - 可调整 GPU/CPU、批大小等 / Adjustable GPU/CPU, batch size, etc.

变更 / Changes

• 需要两步执行（不再是单文件）/ Requires two-step execution (no longer single file)
• 需要安装 PyTorch / Requires PyTorch installation
• 旧版保留为 cs2_swiss_predictor_old.py / Old version kept as cs2_swiss_predictor_old.py

感谢 @Tenzray 的贡献 / Thanks to @Tenzray for the contribution

New Contributors

• @Tenzray made their first contribution in #1

Full Changelog: v1.0.0...v2.0.0

v1.0.0 相对稳定的多进程CPU版本 (Stable Multi-Process CPU Version)

功能特性 / Features

• 完整的 ELO 评分系统 / Complete ELO rating system
• Buchholz 瑞士轮配对（Valve 官方规则）/ Buchholz Swiss pairing (Valve official rules)
• 10 万次蒙特卡洛模拟 / 100k Monte Carlo simulations
• 多进程优化（16 workers）/ Multi-process optimization (16 workers)
• 16 核 CPU 约 20 小时运行时间 / ~20 hours on 16-core CPU

性能 / Performance

• CPU 多进程并行计算 / CPU multi-process parallel computing
• Boolean 数组预计算优化 / Boolean array pre-computation optimization

Full Changelog: https://github.com/holygodly/CS2_Major_Swiss_System_Predictor/commits/v1.0.0