Keras multi GPU in vast.ai

1

I am trying to run a keras model on vast.ai using multiple GPUs. For that I am using keras.utils.multi_gpu_model , however I keep having this error:

if multi_GPU and n_GPUs > 1:
    model = multi_gpu_model(model)

AttributeError: module 'tensorflow_core._api.v2.config' has no attribute 'experimental_list_devices')


I am using this default docker :

Official docker images for deep learning framework TensorFlow

Successfully loaded tensorflow/tensorflow:nightly-gpu-py3

I have also checked the available GPUs and all the GPUs are detected correctly:

Any ideas?

cheers

Joaquin Rives

Posted 2020-02-05T10:48:38.583

Reputation: 11

Can you include full code that produces this error, including imports? – Dr. Snoopy – 2020-02-05T10:56:32.957

Solution:

Finally I found the solution myself. I just used another docker image with an older version of tensorflow (2.0.0), and the error disappeared. – Joaquin Rives – 2020-02-05T19:04:40.637

Answers

0

Solution:

Finally I found the solution myself. I just used another docker image with an older version of tensorflow (2.0.0), and the error disappeared.

Joaquin Rives

Posted 2020-02-05T10:48:38.583

Reputation: 11