Li Dong
Li Dong
I have the same problem. The cloudwatch log generated by sagemaker endpoint have too much redundant info. For example, the timestamp are repetitive and the com.amazonaws.ml.mms.wlm.WorkderLifeCycle doesn't mean anything to...
I wonder if you consider adding this feature at some point since dataparallel is quite useful when data is large.
I'd like to see such a PR too!
This is really nice work! Look forward to Falcon 7b!
I wanted to help but writing cuda/cpp code is not really my strength. I'm happy to do some testing once it reaches that stage. I would like to try this...
> I actually have a question from you guys, has anyone tested the inference of this model on [text_generation_inference](https://github.com/huggingface/text-generation-inference) system from HuggingFace? I tried FLAN-T5-XXL on TGI and compare the...
> @zphang is having different batch sizes per task actually helpful? Would be interesting to know as it's not something I've come across as a technique used by any MTL...