AstrAI/astrai/parallel
ViperEkura 9b416c1bbb refactor : 并行启动 Strategy 模式重构,local_rank 解耦
- setup_parallel 接收 local_rank 参数,不再读环境变量推导
- TorchrunStrategy 从 env 读取 LOCAL_RANK,LocalStrategy 用 rank
- _detect_launcher() 分级检测替代内联 RANK 检查
- _run_single_rank 统一入口,消除 _run_single/_run_multi 重复
- 优雅退出:except BaseException 终止子进程并 re-join
- gradient_checkpointing_modules 判定提取到外部变量
2026-06-02 11:22:24 +08:00
..
__init__.py feat: 新增FSDP并行后端 2026-05-25 19:43:14 +08:00
executor.py fix : 修复存储层 rglob 死锁、DDP LOCAL_RANK 绑定 2026-06-02 01:01:00 +08:00
module.py refactor: 优化参数传递,清理导入样式 2026-04-03 22:06:32 +08:00
setup.py refactor : 并行启动 Strategy 模式重构,local_rank 解耦 2026-06-02 11:22:24 +08:00