search

ufoym / deepo

Setup and customize deep learning environment in seconds.
http://ufoym.com/deepo
MIT License
6.32k stars 750 forks source link

Python dumped when trying to import tensorflow #123

Closed Frankr0 closed 4 years ago

Frankr0 commented 5 years ago

I am a beginner in docker, when I follow the quick start instruction: docker run --runtime=nvidia --rm ufoym/deepo nvidia-smi got:

docker: Error response from daemon: Unknown runtime specified nvidia.
See 'docker run --help'.

Is --runtime=nvidia deprecated?

As the instruction of the nvidia-docker I use --gpus instead of --runtime=nvidia: docker run --gpus all -it --ipc=host ufoym/deepo bash

But:

root@ea810a8cf7b7:/# python
Python 3.6.8 (default, Jan 14 2019, 11:02:34) 
[GCC 8.0.1 20180414 (experimental) [trunk revision 259383]] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow
Illegal instruction (core dumped)
root@ea810a8cf7b7:/#

Was there anything wrong with what I did? Thx

ufoym commented 4 years ago

Fixed. Feel free to reopen it if you still have issues with the latest images. BTW, --runtime=nvidia is deprecated and should be replaced by --gpus all. Thank you for pointing out our mistake.