Jinghan Yao

Results 6 comments of Jinghan Yao

Hi, @derekmauro , thanks for the reply. Sure, so my project does not directly use Absl, but instead, it has dependencies on GRPC, GoogleTest, and Google-cloud-cpp. After building absl, I...

Sure, thanks a lot! Plan to rewrite it in Tensorflow~

Checks are failing because new changes require calling `initialize` with defined `comm`, as `initialize(comm=MPI.COMM_WORLD)`

> It looks you use random number for some argument like request_output_len and the random number is invalid. Hi, @byshiue , I see what you mean, basically --shape only specifies...

Hi @byshiue , Thanks for the reply. So I am a little bit confused here. When enabling tensor parallelism, FasterTransformer expects it happens intra-node. So for example, each node has...

Oh, I see. I totally get it, it basically serves for handling different requests. Thanks for the explanation!