sunjunlishi

do you have trained model.

loogarch64 ---__lasx_xvpickve_d

1

macihne's is /* Assembly instruction format: xd, xj, ui2 */ /* Data types in instruction templates: V4DI, V4DI, V4DI, UQI */ #define __lasx_xvpickve_d(/*__m256i*/ _1, /*__m256i*/ _2, /*ui2*/ _3) ((__m256i)__builtin_lasx_xvpickve_d((v4i64)(_1), (v4i64)(_2),...

💡 [REQUEST] - <title>怎么能够继续昨天的训练，继续训练

3

### 起始日期 | Start Date _No response_ ### 实现PR | Implementation PR _No response_ ### 相关Issues | Reference Issues _No response_ ### 摘要 | Summary 昨天训练到checkpoint-1200，loss值是0.9 今天想继续昨天的训练，能解决吗？ 不然loss值要从头2.7开始往下降，比较耗时。 ### 基本示例...

 One account is ok,but another account is not ok