"...git@developer.sourcefind.cn:wuxk1/megatron-lm.git" did not exist on "a9738f63a738cd445b04a7251d939c1cae84c4a1"
- 08 Apr, 2025 5 commits
-
-
Dongz authored
* [minor]: optimize dockerfile for fewer layer * [feature]: add pre-commit lint, update readme for contribution guidance * [minor]: fix run shell privileges * [auto]: first lint without rule F, fix rule E * [minor]: fix docker file error
-
gushiqiao authored
Co-authored-by:gushiqiao <gushiqiao@sensetime.com>
-
TorynCurtis authored
* 修改了main.py, t5的model, wan的model、三个weights文件和三个infer文件, 并且在common的conv3d算子中注册新算子 * 修改了Conv3dWeightForceBF16算子，更新了wan的pre_weights中对此算子的使用 * 修复了import中的bug * 修复了WanPreWeights, WanTransformerWeights没有self.config的bug * 修复了WanPreWeights, WanTransformerWeights没有self.config的bug * 修复了config的bug，目前在使用cpu_offload的时候，vae阶段有tensor不在同一device的bug * 修复了vae阶段迁移的bug * 修复了scale在mean和inv_std迁移后仍需重新赋值的bug
-
helloyongyang authored
-
helloyongyang authored
-
