DeepSpeedExamples icon indicating copy to clipboard operation
DeepSpeedExamples copied to clipboard

autotuning, error

Open EthanChen1234 opened this issue 1 year ago • 1 comments

hi, thanks for your great job.

I'm trying to start autotuning task, but stuck with some errors.

entry script

cd DeepSpeedExamples/autotuning/hf/gpt2-medium bash test_tune.sh z0, ok bash test_tune.sh tune, errors

environment

ngc-23.03 apt-get install nfs-common pdsh

error infos

image

autotuning.tar.gz

some questions

  1. is there any recommended image or dockerfile, to avoid the environment problem?
  2. how to solve the above errors

EthanChen1234 avatar Nov 10 '22 13:11 EthanChen1234