Udhav Sethi

Results 2 comments of Udhav Sethi

I see that in your code the output is printed twice, once per GPU. Why is that? How to run the inference only once per example? Also, is deepspeed inference...

@chenzuozhou I am trying to run the fine-tuning code using deepspeed using a similar setting as yours - I have access to eight 32GB V100 GPUs. I am running the...