deepmd-kit package is inconsitent with TensorFlow Runtime

[WU2JING]

I successfully install deepmd-kit 2.1.4, but the error raised as follow, when train a model.

[wanghan-iapcm]

@njzjz could you please take a look?

[njzjz]

How did you install deepmd-kit? Did you use devtoolset on RHEL6/7 as is shown in the error message?

[WU2JING]

I install deepmd-kit with conda command and set CONDA_OVERRIDE_CUDA="11.6"

[njzjz]

Failed to reproduce: https://nbviewer.org/gist/njzjz/e660ea2d5469d6b1a2fb44e306bcbc38

[njzjz]

You may check your packages (conda list, pip list) and environment variables (env) to see if anything is wrong.

[WU2JING]

Ok, thank you.

[WU2JING]

It works, thanks again.

[samuelzyzhang]

I successfully install deepmd-kit 2.1.4, but the error raised as follow, when train a model.

I encountered the same problem as yours, how did you solve it?

[WU2JING]

You can try to use the fllow command.
CONDA_OVERRIDE_CUDA=11.6 pwd/miniconda3/bin/conda create -n deepmd deepmd-kit=*=*gpu -c https://conda.deepmodeling.org -y