|
|
ef56e5dcdb
|
Revert "tensorRT engines尝试精度没过，暂时先提交代码，后续再继续调试"
This reverts commit e1f8a83648.
|
2026-02-19 20:22:19 +08:00 |
|
|
|
e1f8a83648
|
tensorRT engines尝试精度没过，暂时先提交代码，后续再继续调试
|
2026-02-18 18:22:12 +08:00 |
|
|
|
afa12ba031
|
每步迭代保存异步
|
2026-02-10 19:54:53 +08:00 |
|
|
|
223a50f9e0
|
添加CrossAttention kv缓存，减少重复计算，提升性能，psnr=25.1201dB
|
2026-02-10 17:35:03 +08:00 |
|
|
|
91a9b0febc
|
DDIM loop 内小张量分配优化，attention mask 缓存到 GPU
|
2026-02-10 16:53:00 +08:00 |
|
|
|
ed637c972b
|
tf32推理
|
2026-02-10 16:39:14 +08:00 |
|
|
|
fffc5a9956
|
init
|
2026-02-08 03:29:15 +00:00 |
|
yuchen-x
|
edf8df63ac
|
upload submodule
|
2025-09-15 18:06:42 +08:00 |
|
