Commit Graph

10 Commits

Author SHA1 Message Date
qhy
f386a5810b 补充上次提交 2026-02-11 16:24:40 +08:00
qhy
352a79035f 主干部分fp16,最敏感psnr=25.21,可以考虑对主干部分太敏感的部分回退fp32 2026-02-11 16:23:21 +08:00
qhy
9a08e27a19 KV 融合实现完成。改动总结: 速度微弱提升psnr略微上升
attention.py — 3处改动:
  1. __init__ 添加 _kv_fused = False 标志
  2.新增 fuse_kv() 方法:将 to_k + to_v → to_kv,同时处理 _ip/_as/_aa 辅助 KV 对
  2. bmm_forward 两个分支加_kv_fused 判断,用to_kv().chunk(2, dim=-1) 替代分别调用
2026-02-11 12:36:38 +08:00
qhy
afa12ba031 每步迭代保存异步 2026-02-10 19:54:53 +08:00
qhy
bf4d66c874 跳过模型加载 2026-02-10 19:36:17 +08:00
qhy
2a6068f9e4 减少了一路视频vae解码 2026-02-10 17:13:45 +08:00
qhy
ed637c972b tf32推理 2026-02-10 16:39:14 +08:00
yuchen-x
f12b478265 upload real-robot deployment code 2025-09-23 15:13:22 +08:00
yuchen-x
9c9942f1d4 upload eval_utils.py file 2025-09-16 21:48:15 +08:00
yuchen-x
d7be60f9fe init commit 2025-09-12 21:53:41 +08:00