Any way to disable using GPU?

[gai00]

my environment:
Ubuntu 18.04 without GPU.
Docker Image: tensorflow/tensorflow:latest-py3-jupyter
Tensorflow Version: 2.1.0
TensorBoard Version: 2.1.0

errors:

ERROR: Failed to launch TensorBoard (exited with 0).
Contents of stderr:
2020-01-17 04:12:50.147356: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer.so.6'; dlerror: libnvinfer.so.6: cannot open shared object file: No such file or directory
2020-01-17 04:12:50.147440: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'libnvinfer_plugin.so.6'; dlerror: libnvinfer_plugin.so.6: cannot open shared object file: No such file or directory
2020-01-17 04:12:50.147452: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:30] Cannot dlopen some TensorRT libraries. If you would like to use Nvidia GPU with TensorRT, please make sure the missing libraries mentioned above are installed properly.

Should do I install lib even I don't want to use GPU?

[rmothukuru]

In order to expedite the trouble-shooting process, please provide a code snippet to reproduce the issue reported here. Thanks!

[aretas77]

I get the exact same error.
Arch Linux 5.4.11-arch1-1 without GPU.
Python 3.7.6
Tensorflow Version: 2.1.0
TensorBoard Version: 2.1.0

[wchargin]

Hi @gai00! The GPU warnings that you’re seeing are just warnings, and
shouldn’t stop TensorBoard from launching. It looks like you’re trying
to launch TensorBoard from a notebook context (with %tensorboard) but
it’s failing to launch. Can you please post the code that you’re using
to launch TensorBoard? When I run that container with

docker run --rm -it -p 8888:8888 -p 6006:6006 \
    tensorflow/tensorflow:latest-py3-jupyter 

and then launch TensorBoard from a new Jupyter notebook via

%load_ext tensorboard
%tensorboard --logdir logs --bind_all

it works for me:

[gai00]

thank you @wchargin !
it work!

I open the chrome console to check why "connection deny", tensorboard open on port:6007.
maybe some shell code effected.

Re-new one docker container, all done.